Re: [DISCUSS] OOM killer and Routing/System VM's = :(
On Wed, Sep 04, 2013 at 12:14:27PM -0700, Darren Shepherd wrote: On 09/04/2013 11:37 AM, Roeland Kuipers wrote: Hi Darren, Thanks for your reply! Could you share a bit more on your plans/ideas? We also have been braining on other approaches of managing the systemvm's, especially small customizations for specific tenants. And maybe even leveraging a config mgmt tools like chef or puppet with the ability to integrate CS with that in some way. I'll have to send the full details later but here's a rough idea. The basic approach is this. Logical changes to the VRs (or system vms in general) get mapped to configuration items. So add a LB rule maps to iptables config and haproxy config. When you change a LB rule we then bump up the requested version of the configuration for iptables/haproxy. So the requested version will be 4 maybe. The applied version will be 3 as the VR still has the old configuration. Since 4 != 3, the VR will be signaled to pull the latest iptables/haproxy config. So it will pull the configuration. Say in the mean time somebody else adds four other LB rules. So the requested version is now at 8. So when the VR pulls the config it will get version 8, and then reply back saying it applied version 8. The applied version is now 8 which is greater than 4 (the version the first LB rule change was waiting for) so basically all async jobs waiting for the LB change will be done. To pull the configuration from the VR, the VR will be hitting a templating configuration system. So it pulls the full iptables and haproxy config. Not incremental changes. So if the VR ever reboots itself, it can easily just pull the latest config of everything and apply it. So it will be consistent. I'd be interested to hear what type of customizations you would like to add. It will definitely be an extensible system, but the problem is if your extensions wants to touch the same configuration files that ACS wants to manage. That gets a bit tricky as its really easy for each to break each other. But I can definitely add some hooks that users can use to mess up things and void the warranty. I've thought about chef and puppet for this, but basically it comes down to two things. I'm really interested in this being fast and light weight. Ruby is neither of those. So the core ACS stuff will probably remain as very simple shell scripts. Simple in that they really just need to download configuration and restart services. They know nothing about the nature of the changes. If, as an extension, you want to do something with puppet, chef, I'd be open to that. That's your deal. How about ansible? https://github.com/ansible/ansible No custom agents, plain ssh, doesn't need root acces. all configuration is yaml based. you can extend in any language. This approach has many other benefits. Like, for example, we can ensure that as we deploy a new ACS release existing system VMs can be updated (without a reboot, unless the kernel changes). Additionally, its fast and updates happen in near constant time. So most changes will be just a couple of seconds, even if you have 4000 LB rules. Darren -- Prasanna., Powered by BigRock.com
Re: documentation/wiki is a mess
On Sep 5, 2013, at 1:11 AM, Prasanna Santhanam t...@apache.org wrote: Radhika, what I meant was everyone should help out with docs. Esp. those working on a feature should care most that the docs for their feature is perfect for an end user to understand, implement and use. We shouldn't piggy-back on those helping fix doc bugs all the time which I see happening too often. Someone files a doc bug and someone else fixes it and someone else reviews it and finally the users are still having trouble understanding it. We're just creating work for ourselves that way. Feature specs have architectural and implementation details and may not often be fully baked by the time the feature starts development. Changes happen during implementation and reviewing our docs after everything is merged is a good forcing-function to fix our docs. I'm +1 for style guides and markdown based docs. They should make this whole process a lot easier for everyone. I am a big +1 for literally forcing devs to submit docs for user facing features, otherwise the feature does not get added. Just like unit tests. No unitests, no commit, no docs, no commit. The devs are the best to write the docs of their own feature. On Wed, Sep 04, 2013 at 09:52:03PM -0700, Ahmad Emneina wrote: I think prasanna hit the nail on the head. I'm sure there are features there no one knows about, or ever will... +1 for developers documenting what features/functions are abound. On Wed, Sep 4, 2013 at 9:36 PM, Radhika Puthiyetath radhika.puthiyet...@citrix.com wrote: Hi, We do not have a doc team as such :-) We have a set of doc contributors that so far have not worked as a team. Probably, we should think about aligning the doc efforts and having a process and style guide in place. If the FS is good enough, we need not trouble the code committers to write own docs is what I feel. Regards -Radhika -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: Thursday, September 05, 2013 9:34 AM To: dev@cloudstack.apache.org Subject: Re: documentation/wiki is a mess On Tue, Sep 03, 2013 at 11:10:28AM -0700, Darren Shepherd wrote: The state of documentation of an open source projects says a lot about the community that develops it... Esp. for XML documentation: I say we write our own docs if we want our feature to be used. Or it dies a natural death in quiet isolation with no one ever using it. Docs team can handle the editing and organizing bits. That should go for Wikis too. If you want the wiki to be useful then write it, organize it and maintain it. Don't just put it there to fill a template. It'll never recieve any love. -- Prasanna., Powered by BigRock.com -- Prasanna., Powered by BigRock.com
Re: Review Request 13964: CLOUDSTACK-4203: Adding a test for migrating volumes of stopped vms to test_stopped_vm.py
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13964/ --- (Updated Sept. 5, 2013, 6:32 a.m.) Review request for cloudstack, venkata swamy babu budumuru, SrikanteswaraRao Talluri, and Prasanna Santhanam. Bugs: CLOUDSTACK-4203 Repository: cloudstack-git Description --- Adding test for migrating volumes of stopped vm. Diffs - test/integration/component/test_stopped_vm.py 41eeb46 Diff: https://reviews.apache.org/r/13964/diff/ Testing --- Yes Thanks, sanjeev n
RE: SystemVM downloading
Hi Donal, I had tried the hyperv systemvm generated from the Jenkins job. I was getting image corruption error. I don't think the systemvm generated for hyperv is proper one. I had tried xenserver systemvm as it is VHD, it worked fine without issues. Thanks Rajesh Battala -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: Thursday, September 5, 2013 10:24 AM To: dev@cloudstack.apache.org Subject: Re: SystemVM downloading Hey Donal, Soheil, I had to remove those artifacts from jenkins as they occupied a lot of space on the instance which is a m1.small with a small disk. I've been looking for some mirroring space or s3 bucket where I can push the system VMs. You could meanwhile build your own as described here if you want to experiment with the latest. https://cwiki.apache.org/confluence/x/UlHVAQ I'll see if I can put up the hyperv vhd somewhere meanwhile. On Wed, Sep 04, 2013 at 09:16:05PM +, Rayees Namathponnan wrote: Hi Soheil, You can get the template from below location; it's not latest but tested extensively http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06-12-ma ster-kvm.qcow2.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-2013-07-12-ma ster-xen.vhd.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-4.2-vh7.ova Regards, Rayees -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Wednesday, September 04, 2013 1:14 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Not just Hyper-V but looks like none of the sytemvm images e.g. Xen/VMware are posted on the site. -Soheil From: Donal Lafferty [donal.laffe...@citrix.com] Sent: Wednesday, September 04, 2013 12:35 PM To: dev@cloudstack.apache.org Subject: SystemVM downloading I thought I'd be able to download systemvmtemplate-2013-09-03-master-hyperv.vhd from http://jenkins.cloudstack.org/view/master/job/build-systemvm-master/ However, I don't see a link to do so. Is systemvmtemplate-2013-09-03-master-hyperv.vhd or other recent copy online anywhere? DL -- Prasanna., Powered by BigRock.com
[ANNOUNCE] New PMC member: Ilya Musayev
The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
RE: [ANNOUNCE] New PMC member: Ilya Musayev
Hearty Congratulations Ilya. Thanks Rajesh Battala -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 5, 2013 12:22 PM To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: Passing vlan parameter to createPortableIpRange
Ok that means in any case public or portable ip ranges should not overlap and right now this is not handled for public ranges. That removes my confusion. Thanks swamy Regards, Gaurav On Thu, Sep 5, 2013 at 10:27 AM, Venkata SwamyBabu Budumuru venkataswamybabu.budum...@citrix.com wrote: Passing a vlan id or not doesn't matter as long as your infra is configured accordingly. The only thing that user should not do and CS is allowing is passing the same public ip range in different vlans. On 05/09/13 9:52 AM, Girish Shilamkar gir...@clogeny.com wrote: Hi Swamy, Thanks for setting up a call, the other day, we were able resolve this issue quickly. In what scenario would vlan id be passed? How does passing vlan id change/affect Portable IP range to be created ? I am guessing that Portable IP range created would be within IP range assigned for the vlan. (which as of now does not work) Regards, Girish On 05-Sep-2013, at 9:41 AM, Venkata SwamyBabu Budumuru venkataswamybabu.budum...@citrix.com wrote: Hi Gaurav, The VLAN parameter is not mandatory. Depending on your environment you can either pass a VLAN tag / untagged one. We have updated our documents about the issue with adding same ip range across portable and public VLANs. This is punted for 4.2.1. Thanks, SWAMY On 04/09/13 8:26 PM, Gaurav Aradhye gaurav.arad...@clogeny.com wrote: Hi Swamy, As observed, if portable ip range is created with IPs overlapping with any of the existing public ip ranges, then it gives Entity already exists error while associating any portable ip from the created range. (But it doesn't give any error while creating portable ip range in this case) The question is if we can't use IPs from existing VLANs, why to pass vlan parameter while creating portable ip range? Keeping the default value untagged serves the purpose and works well. I didn't get this well. Can you please explain me the how we can use vlan ids to while creating portable range and still get it work correctly? Regards, Gaurav
RE: Password prompt like nine times
Hi Mike, I too had faced this issue long back, after changing my sudoers file it never appeared. Here is my sudoers file content. Am running jetty from rajesh user. Defaultsenv_reset # Host alias specification # User alias specification # Cmnd alias specification # User privilege specification rootALL=(ALL:ALL) ALL rajesh ALL=(ALL:ALL) ALL # Members of the admin group may gain root privileges %admin ALL=(ALL) ALL # Allow members of group sudo to execute any command %sudo ALL=(ALL:ALL) ALL %rajesh ALL=(ALL:ALL) ALL #includedir /etc/sudoers.d rajesh ALL=(ALL) NOPASSWD:ALL Thanks Rajesh Battala -Original Message- From: Ian Duffy [mailto:i...@ianduffy.ie] Sent: Thursday, September 5, 2013 9:22 AM To: CloudStack Dev Subject: Re: Password prompt like nine times do you see any error in what I'm doing in that little clip I provided of my sudoers file? Its to do with the way sudo permissions are resolved. To quote the sudoers man pages: When multiple entries match for a user, they are applied in order. Where there are multiple matches, the last match is used (which is not necessarily the most specific match). Your user mtutkowski is part of the admin group, or atleast thats my guess. Validate this with: $ id mtutkowski This means the access that gets applied for mtutkowski is ALL=(ALL) ALL which requires a password. Switch around your mtutkowski and %admin lines and it should work. e.g. rootALL=(ALL) ALL %admin ALL=(ALL) ALL mtutkowski ALL=(ALL) NOPASSWD: ALL On 5 September 2013 03:49, Mike Tutkowski mike.tutkow...@solidfire.comwrote: Interesting, Ian. Although maybe not as ideal as your setup, do you see any error in what I'm doing in that little clip I provided of my sudoers file? It doesn't seem to stop the password prompts. Thanks! On Wed, Sep 4, 2013 at 8:26 PM, Ian Duffy i...@ianduffy.ie wrote: This is what I have been using. I didn't want to give my cloudstack full root access via sudo. cloudstack ALL=NOPASSWD: /usr/bin/keytool* cloudstack ALL=NOPASSWD: /bin/mkdir -p /home/cloudstack/* cloudstack ALL=NOPASSWD: /bin/mount -o loop /home/cloudstack/* On 5 September 2013 03:05, Mike Tutkowski mike.tutkow...@solidfire.com wrote: This is what the relevant part of my sudoers file (on Mac OS X and Ubuntu 12.04.1) looks like: # User privilege specification rootALL=(ALL) ALL mtutkowski ALL=(ALL) NOPASSWD: ALL %admin ALL=(ALL) ALL On Wed, Sep 4, 2013 at 6:39 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Yeah, CSMS = CloudStack Management Server Should this work? root ALL=(ALL) NOPASSWD: ALL On Wed, Sep 4, 2013 at 5:30 PM, Kelven Yang kelven.y...@citrix.com wrote: You need to make sudo-ers no-password required Kelven On 9/4/13 4:10 PM, Tracy Phillips tracp...@mantoso.com wrote: On Wed, Sep 4, 2013 at 6:47 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Hi, I think this has been discussed a few times on the list before. When I start up the CSMS, I get prompted for my password about nine times. I thought updating the sudoers file would fix this, but it still happens. Any thoughts on this? Thanks! Mike, Do you mean the Cloudstack Management Server when you say CSMS? What do you have in your sudoers file? Tracy -- *Mike Tutkowski* *Senior CloudStack Developer, SolidFire Inc.* e: mike.tutkow...@solidfire.com o: 303.746.7302 Advancing the way the world uses the cloud http://solidfire.com/solution/overview/?video=play *™* -- *Mike Tutkowski* *Senior CloudStack Developer, SolidFire Inc.* e: mike.tutkow...@solidfire.com o: 303.746.7302 Advancing the way the world uses the cloudhttp://solidfire.com/solution/overview/?video=play *™* -- *Mike Tutkowski* *Senior CloudStack Developer, SolidFire Inc.* e: mike.tutkow...@solidfire.com o: 303.746.7302 Advancing the way the world uses the cloudhttp://solidfire.com/solution/overview/?video=play *™*
Re: NetworkElements: configure the provider to enable the provider?
Had a chat with Murali about the semantics of adding external elements. I summarize here: 1. addProvider (virtualrouter and derivatives|physical external-service-provider|external-controller) 2. configureProvider 3. enableProvider The VR/VpcVR/InternalLBVm do these three steps but since they are template based systems one can enable the provider soon as it is configured. Because management server is responsible for start up of these appliances we consider them to be ready (isReady = true). OTOH, the external physical elements like the JuniperSRX, BigIP and NetScaler are not controlled by CloudStack and need to be externally configured. They need to signal to CS that they are ready for operation (isReady = false until a device is brought up) SDN Controllers like Midonet, Nicira take control of all aspects of routing and network traffic mgmt so they are always ready (isReady=true) since CS offloads all operations to these elements. Inflobox is similar to the external elements I think so it needs to signal to CS that it is indeed ready for operation. So I'm not sure if it should be in the default set of providers that are enabled. To handle the tests: deployDataCenter.py only optionally enables the provider and the device if they are available. We should keep that to be able to prepare a deployment with devices if and when they are provided in the marvin configuration. We can workaround: 1. enable the provider, add the device and remove it to keep the provider enabled for subsequent tests 2. attempt to enable the provider per-test (as you've done) I'm ok with 2. temporarily On Wed, Sep 04, 2013 at 05:25:24PM +, Sowmya Krishnan wrote: So to get around this limitation, should we be looking at making appropriate changes to deployDataCenter.py to not try to add NS device if it's not supplied? The test would take care anyway of adding the device and configuring the provider. Thoughts? -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Tuesday, September 03, 2013 11:48 AM To: dev@cloudstack.apache.org Subject: RE: NetworkElements: configure the provider to enable the provider? Hi Prasanna, The limitation you have highlighted enable network service providers without actually adding the device is not enforced by the NetworkManager but by the design of the NetworkElement. If you have not registered a device with the Netscaler NetworkElement it is rejecting the Enable. The NetworkManager calls the NetworkElement isReady() and that determines if it is ready to accept the call to Enable. I assume that NetScaler NetworkElement code can not handle subsequent calls to Prepare() etc. unless a device is configured. I assume those other NetworkElements you have highlighted, have chosen to return True on that call, and handle the subsequent calls e.g. Prepare(), when there is not a device, my NetworkElement behaves this way so that it can be enabled early when there is no device. There is a design pattern to enable some Network Elements that offer service to the Physical and Guest network in the call to createPhysicalNetwork() in NetworkServiceImpl: ... // add VirtualRouter as the default network service provider addDefaultVirtualRouterToPhysicalNetwork(pNetwork.getId()); // add security group provider to the physical network addDefaultSecurityGroupProviderToPhysicalNetwork(pNetwork.getId()); // add VPCVirtualRouter as the defualt network service provider addDefaultVpcVirtualRouterToPhysicalNetwork(pNetwork.getId()); // add baremetal as the defualt network service provider addDefaultBaremetalProvidersToPhysicalNetwork(pNetwork.getId()); //Add Internal Load Balancer element as a default network service provider addDefaultInternalLbProviderToPhysicalNetwork(pNetwork.getId()); // add Infoblox as the default network service provider addDefaultInfobloxProvidersToPhysicalNetwork(pNetwork.getId()); -Soheil From: Prasanna Santhanam [t...@apache.org] Sent: Saturday, August 31, 2013 1:36 AM To: CloudStack Dev Subject: NetworkElements: configure the provider to enable the provider? I'm looking to enable network service providers without actually adding the device but it appears as if the addition of the device is coupled with the enabling of the provider. While this is not the case with default providers like VR, VpcVR, security groups and the InternalLbVm? What is the reason for the restriction placed on External NetworkElements? In marvin integration tests [1] we want to be able to enable the provider whilst deployment and later add VPX Netscalers per test but this is not possible since the Netscaler provider cannot be enabled
Re: Password prompt like nine times
On Sep 5, 2013, at 1:30 AM, Kelven Yang kelven.y...@citrix.com wrote: You need to make sudo-ers no-password required bad idea, someone compromises your account and he is root. Kelven On 9/4/13 4:10 PM, Tracy Phillips tracp...@mantoso.com wrote: On Wed, Sep 4, 2013 at 6:47 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Hi, I think this has been discussed a few times on the list before. When I start up the CSMS, I get prompted for my password about nine times. I thought updating the sudoers file would fix this, but it still happens. Any thoughts on this? Thanks! Mike, Do you mean the Cloudstack Management Server when you say CSMS? What do you have in your sudoers file? Tracy
RE: [ANNOUNCE] New PMC member: Ilya Musayev
Congrats Ilya! Well deserved :-) Regards, Sateesh -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: 05 September 2013 12:23 To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Great stuff, congrats Ilya. :) On Thu, Sep 5, 2013 at 4:13 PM, Sateesh Chodapuneedi sateesh.chodapune...@citrix.com wrote: Congrats Ilya! Well deserved :-) Regards, Sateesh -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: 05 September 2013 12:23 To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Congrats :-) On Sep 5, 2013, at 8:51 AM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
RE: documentation/wiki is a mess
+1 on no docs, no submit Kind Regards Giles D: +44 20 3603 0541 | M: +44 796 111 2055 giles.sir...@shapeblue.com -Original Message- From: Sebastien Goasguen [mailto:run...@gmail.com] Sent: 05 September 2013 07:14 To: dev@cloudstack.apache.org Cc: aemne...@gmail.com Subject: Re: documentation/wiki is a mess On Sep 5, 2013, at 1:11 AM, Prasanna Santhanam t...@apache.org wrote: Radhika, what I meant was everyone should help out with docs. Esp. those working on a feature should care most that the docs for their feature is perfect for an end user to understand, implement and use. We shouldn't piggy-back on those helping fix doc bugs all the time which I see happening too often. Someone files a doc bug and someone else fixes it and someone else reviews it and finally the users are still having trouble understanding it. We're just creating work for ourselves that way. Feature specs have architectural and implementation details and may not often be fully baked by the time the feature starts development. Changes happen during implementation and reviewing our docs after everything is merged is a good forcing-function to fix our docs. I'm +1 for style guides and markdown based docs. They should make this whole process a lot easier for everyone. I am a big +1 for literally forcing devs to submit docs for user facing features, otherwise the feature does not get added. Just like unit tests. No unitests, no commit, no docs, no commit. The devs are the best to write the docs of their own feature. On Wed, Sep 04, 2013 at 09:52:03PM -0700, Ahmad Emneina wrote: I think prasanna hit the nail on the head. I'm sure there are features there no one knows about, or ever will... +1 for developers documenting what features/functions are abound. On Wed, Sep 4, 2013 at 9:36 PM, Radhika Puthiyetath radhika.puthiyet...@citrix.com wrote: Hi, We do not have a doc team as such :-) We have a set of doc contributors that so far have not worked as a team. Probably, we should think about aligning the doc efforts and having a process and style guide in place. If the FS is good enough, we need not trouble the code committers to write own docs is what I feel. Regards -Radhika -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: Thursday, September 05, 2013 9:34 AM To: dev@cloudstack.apache.org Subject: Re: documentation/wiki is a mess On Tue, Sep 03, 2013 at 11:10:28AM -0700, Darren Shepherd wrote: The state of documentation of an open source projects says a lot about the community that develops it... Esp. for XML documentation: I say we write our own docs if we want our feature to be used. Or it dies a natural death in quiet isolation with no one ever using it. Docs team can handle the editing and organizing bits. That should go for Wikis too. If you want the wiki to be useful then write it, organize it and maintain it. Don't just put it there to fill a template. It'll never recieve any love. -- Prasanna., Powered by BigRock.com -- Prasanna., Powered by BigRock.com This email and any attachments to it may be confidential and are intended solely for the use of the individual to whom it is addressed. Any views or opinions expressed are solely those of the author and do not necessarily represent those of Shape Blue Ltd or related companies. If you are not the intended recipient of this email, you must neither take any action based upon its contents, nor copy or show it to anyone. Please contact the sender if you believe you have received this email in error. Shape Blue Ltd is a company incorporated in England Wales. ShapeBlue Services India LLP is operated under license from Shape Blue Ltd. ShapeBlue is a registered trademark.
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Congratulations!! 2013/9/5 sebgoa run...@gmail.com The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: cpvm and ssvm but no host entries
On 09/04/2013 10:01 PM, Prasanna Santhanam wrote: On Wed, Sep 04, 2013 at 04:38:05PM -0700, Darren Shepherd wrote: So I'm trying to setup a cloudstack environment from scratch. Not a devcloud environment, but more like a production setup. Anyhow, so I managed to get the cpvm and ssvm running, but there is no host entry for them. Is that supposed to happen? If you mean nothing about the cpvm/ssvm in the `cloud`.`host` table. that is not expected. I'm on 4.2 mysql select id,name,uuid,status,type from host; ++---+--+--++ | id | name | uuid | status | type | ++---+--+--++ | 1 | ap... | ef34fcab-7b16-49c7-a249-4d091113def0 | Up | Routing| | 2 | ap... | 2dc20e3e-0fdf-4458-b879-b38595bfa8b7 | Up | Routing| | 3 | ap... | 02c30b44-0356-4070-9230-b20c320b3e8e | Up | Routing| | 4 | v-2-VM| 2d78c5f3-b481-4af4-9674-1e554853efd5 | Disconnected | ConsoleProxy | | 5 | s-1-VM| 0bee83d7-77be-4891-8fa1-4441de410366 | Disconnected | SecondaryStorageVM | | 6 | s-34-QA | 75f608db-9c4a-4915-b1b1-d1914904b930 | Up | SecondaryStorageVM | | 7 | v-38-QA | e77828b3-35b0-42b1-ae08-b8d0c22ba156 | Up | ConsoleProxy | ++---+--+--++ 7 rows in set (0.00 sec) Alright, I'm at a total loss here. Obviously doing something wrong. So in 4.1 the AddSecondaryStorageCmd would call the regular host discovery code which would put the entry in the hosts table. In 4.2 the code was changed to call the storage lifecycle code which seems to only put an entry in the image_store table. I can't find for the life of me anything that would actually insert the row in the host table in 4.2. So whats specifically happening is its calling TemplateService.handleSysTemplateDownload() which eventually calls DefaultEndpointSelector which can't find a host entry so it falls back to LocalHostEndpoint which obviously fails to run the DownloadCommand cause executing on the localhost and not on the SSVM. So in short, I must be doing something wrong Darren
Re: Review Request 13882: CLOUDSTACK-4434: EN: Ubuntu: Direct input - _ , ? /, keyboard / , keyboard - keys are not working well for the US keyboard.
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13882/#review25926 --- Commit 7bff499bd3625f8a75a1c93cb9e3e2c90955df5e in branch refs/heads/master from Sanjay Tripathi [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=7bff499 ] CLOUDSTACK-4434: EN: Ubuntu: Direct input - _ , ? /, keyboard / ,keyboard - keys are not working well for the US keyboard. - ASF Subversion and Git Services On Aug. 28, 2013, 9:21 a.m., Sanjay Tripathi wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13882/ --- (Updated Aug. 28, 2013, 9:21 a.m.) Review request for cloudstack and Devdeep Singh. Bugs: CLOUDSTACK-4434 Repository: cloudstack-git Description --- CLOUDSTACK-4434: EN: Ubuntu: Direct input - _ , ? /, keyboard / ,keyboard - keys are not working well for the US keyboard. https://issues.apache.org/jira/browse/CLOUDSTACK-4434 Diffs - services/console-proxy/server/js/ajaxviewer.js d3f3aa9 Diff: https://reviews.apache.org/r/13882/diff/ Testing --- Verified the fix on Ubuntu 12.04 VM. Thanks, Sanjay Tripathi
Re: Review Request 13804: CLOUDSTACK-4495: systemVM template URL is pointing to old template location in upgrade file.
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13804/#review25927 --- Commit 108aab76652a1d8294ec2c263c58059d1003ee39 in branch refs/heads/master from Sanjay Tripathi [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=108aab7 ] CLOUDSTACK-4495: systemVM template URL is pointing to old template location in upgrade file. - ASF Subversion and Git Services On Aug. 26, 2013, 8:02 a.m., Sanjay Tripathi wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13804/ --- (Updated Aug. 26, 2013, 8:02 a.m.) Review request for cloudstack, Abhinandan Prateek and Devdeep Singh. Bugs: CLOUDSTACK-4495 Repository: cloudstack-git Description --- SystemVM template URL for XenServer hypervisor is pointing to old template location in the Upgrade410to420.java file. https://issues.apache.org/jira/browse/CLOUDSTACK-4495 Diffs - engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java c3276ef Diff: https://reviews.apache.org/r/13804/diff/ Testing --- Tests: 1. Setup CloudStack environment, 2. Check the systemVM template URL in vm_template table (it should point to the new location). Thanks, Sanjay Tripathi
RE: [ANNOUNCE] New PMC member: Ilya Musayev
Congratz Ilya... -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 12:22 PM To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: cpvm and ssvm but no host entries
On Thu, Sep 05, 2013 at 12:47:25AM -0700, Darren Shepherd wrote: On 09/04/2013 10:01 PM, Prasanna Santhanam wrote: On Wed, Sep 04, 2013 at 04:38:05PM -0700, Darren Shepherd wrote: So I'm trying to setup a cloudstack environment from scratch. Not a devcloud environment, but more like a production setup. Anyhow, so I managed to get the cpvm and ssvm running, but there is no host entry for them. Is that supposed to happen? If you mean nothing about the cpvm/ssvm in the `cloud`.`host` table. that is not expected. I'm on 4.2 mysql select id,name,uuid,status,type from host; ++---+--+--++ | id | name | uuid | status | type | ++---+--+--++ | 1 | ap... | ef34fcab-7b16-49c7-a249-4d091113def0 | Up | Routing| | 2 | ap... | 2dc20e3e-0fdf-4458-b879-b38595bfa8b7 | Up | Routing| | 3 | ap... | 02c30b44-0356-4070-9230-b20c320b3e8e | Up | Routing| | 4 | v-2-VM| 2d78c5f3-b481-4af4-9674-1e554853efd5 | Disconnected | ConsoleProxy | | 5 | s-1-VM| 0bee83d7-77be-4891-8fa1-4441de410366 | Disconnected | SecondaryStorageVM | | 6 | s-34-QA | 75f608db-9c4a-4915-b1b1-d1914904b930 | Up | SecondaryStorageVM | | 7 | v-38-QA | e77828b3-35b0-42b1-ae08-b8d0c22ba156 | Up | ConsoleProxy | ++---+--+--++ 7 rows in set (0.00 sec) Alright, I'm at a total loss here. Obviously doing something wrong. So in 4.1 the AddSecondaryStorageCmd would call the regular host discovery code which would put the entry in the hosts table. In 4.2 the code was changed to call the storage lifecycle code which seems to only put an entry in the image_store table. I can't find for the life of me anything that would actually insert the row in the host table in 4.2. If you are looking for the SecondaryStorage nfs host entry (that usually would go into Alert state soon as it was added), it no more goes into the host table. There is no 'discovery' process for storage. Also the AddSecondaryStorageCmd is now deprecated and the AddImageStoreCmd and AddStagingStoreCmds are used for the backing store operations. So whats specifically happening is its calling TemplateService.handleSysTemplateDownload() which eventually calls DefaultEndpointSelector which can't find a host entry so it falls back to LocalHostEndpoint which obviously fails to run the DownloadCommand cause executing on the localhost and not on the SSVM. This is how I see it reported in the logs but havent dug deeper to be honest. Perhaps Edison/Min will have the right answer here So in short, I must be doing something wrong Darren -- Prasanna., Powered by BigRock.com
Questions about 4.2 RC
Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. -Sebastien
RE: SystemVM downloading
Great! I may need to update the template to allow it to accept host data from a Hyper-V 2012. Unlike XenServer, a Hyper-V host passes data through a user mode daemon that write key/value pairs to a file. I need to investigate a bit further, but I believe that the daemon starts and startup scripts need updating to deal with the key/value pair file format. DL -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: 05 September 2013 05:54 To: dev@cloudstack.apache.org Subject: Re: SystemVM downloading Hey Donal, Soheil, I had to remove those artifacts from jenkins as they occupied a lot of space on the instance which is a m1.small with a small disk. I've been looking for some mirroring space or s3 bucket where I can push the system VMs. You could meanwhile build your own as described here if you want to experiment with the latest. https://cwiki.apache.org/confluence/x/UlHVAQ I'll see if I can put up the hyperv vhd somewhere meanwhile. On Wed, Sep 04, 2013 at 09:16:05PM +, Rayees Namathponnan wrote: Hi Soheil, You can get the template from below location; it's not latest but tested extensively http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06- 12-ma ster-kvm.qcow2.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-2013-07- 12-ma ster-xen.vhd.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-4.2- vh7.ova Regards, Rayees -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Wednesday, September 04, 2013 1:14 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Not just Hyper-V but looks like none of the sytemvm images e.g. Xen/VMware are posted on the site. -Soheil From: Donal Lafferty [donal.laffe...@citrix.com] Sent: Wednesday, September 04, 2013 12:35 PM To: dev@cloudstack.apache.org Subject: SystemVM downloading I thought I'd be able to download systemvmtemplate-2013-09-03-master-hyperv.vhd from http://jenkins.cloudstack.org/view/master/job/build-systemvm-master/ However, I don't see a link to do so. Is systemvmtemplate-2013-09-03-master-hyperv.vhd or other recent copy online anywhere? DL -- Prasanna., Powered by BigRock.com
Re: Review Request 13922: Minimal Hyper-V Plugin
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13922/ --- (Updated Sept. 5, 2013, 8:28 a.m.) Review request for cloudstack, Alex Huang, Chiradeep Vittal, and Devdeep Singh. Changes --- Re add dependency that was accidentally deleted at first submission. Repository: cloudstack-git Description --- A plugin for Hyper-V control is available for CloudStack. The plugin implements basic VM control; however, its architecture allows additional functionality to be easily added. Incorporating the plugin in CloudStack will allow the community to participate in improving the features available with Hyper-V. The plugin uses a Director Connect Agent architecture described here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Progress This links provides details of third party libraries used by the plugin including their licenses. No source for these libraries is used, and the binaries are downloaded from their distributors at build time. No proprietary tools are required for the build. For instance, C# compiled with Mono has been tested. Therefore, the plugin has been added to the default build and the default deployment. The plugin includes unit and functional tests that can be triggered at build time. See the link above for details. The Apache header is applied to source and where feasible to build config files. The source is the https://github.com/lafferty/cloudstack/tree/hyperv_plugin Diffs - client/pom.xml 1afe5b898d328b06b5da7c132eaed14fefc78718 client/tomcatconf/applicationContext.xml.in 956c77c5fa6c3d909be1fbee8ba26e8d8ba1dce7 client/tomcatconf/componentContext.xml.in f36d0ee02a2eb91733f6f2a76be60a1477529d4d client/tomcatconf/simulatorComponentContext.xml.in c46d1de086a5633b64c90884dbc02a03c3b2e8ad plugins/hypervisors/hyperv/DotNet/ServerResource/.gitignore PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.Config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.targets PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.settings PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShell.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShellException.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Program.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/CloudStackTypes.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResource.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResourceController.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Utils.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/WmiCalls.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/HypervResourceControllerTest.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/ServerResource.Tests.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.sln PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_OperatingSystem.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_Processor.cs PRE-CREATION
Re: Review Request 13922: Minimal Hyper-V Plugin
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13922/ --- (Updated Sept. 5, 2013, 8:28 a.m.) Review request for cloudstack, Alex Huang, Chiradeep Vittal, and Devdeep Singh. Changes --- Re add dependency that was accidentally deleted at first submission. Repository: cloudstack-git Description --- A plugin for Hyper-V control is available for CloudStack. The plugin implements basic VM control; however, its architecture allows additional functionality to be easily added. Incorporating the plugin in CloudStack will allow the community to participate in improving the features available with Hyper-V. The plugin uses a Director Connect Agent architecture described here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Progress This links provides details of third party libraries used by the plugin including their licenses. No source for these libraries is used, and the binaries are downloaded from their distributors at build time. No proprietary tools are required for the build. For instance, C# compiled with Mono has been tested. Therefore, the plugin has been added to the default build and the default deployment. The plugin includes unit and functional tests that can be triggered at build time. See the link above for details. The Apache header is applied to source and where feasible to build config files. The source is the https://github.com/lafferty/cloudstack/tree/hyperv_plugin Diffs - client/pom.xml 1afe5b898d328b06b5da7c132eaed14fefc78718 client/tomcatconf/applicationContext.xml.in 956c77c5fa6c3d909be1fbee8ba26e8d8ba1dce7 client/tomcatconf/componentContext.xml.in f36d0ee02a2eb91733f6f2a76be60a1477529d4d client/tomcatconf/simulatorComponentContext.xml.in c46d1de086a5633b64c90884dbc02a03c3b2e8ad plugins/hypervisors/hyperv/DotNet/ServerResource/.gitignore PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.Config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.targets PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.settings PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShell.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShellException.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Program.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/CloudStackTypes.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResource.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResourceController.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Utils.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/WmiCalls.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/HypervResourceControllerTest.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/ServerResource.Tests.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.sln PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_OperatingSystem.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_Processor.cs PRE-CREATION
Re: Review Request 13922: Minimal Hyper-V Plugin
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13922/ --- (Updated Sept. 5, 2013, 8:29 a.m.) Review request for cloudstack, Alex Huang, Chiradeep Vittal, and Devdeep Singh. Repository: cloudstack-git Description --- A plugin for Hyper-V control is available for CloudStack. The plugin implements basic VM control; however, its architecture allows additional functionality to be easily added. Incorporating the plugin in CloudStack will allow the community to participate in improving the features available with Hyper-V. The plugin uses a Director Connect Agent architecture described here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Progress This links provides details of third party libraries used by the plugin including their licenses. No source for these libraries is used, and the binaries are downloaded from their distributors at build time. No proprietary tools are required for the build. For instance, C# compiled with Mono has been tested. Therefore, the plugin has been added to the default build and the default deployment. The plugin includes unit and functional tests that can be triggered at build time. See the link above for details. The Apache header is applied to source and where feasible to build config files. The source is the https://github.com/lafferty/cloudstack/tree/hyperv_plugin Diffs - client/pom.xml 1afe5b898d328b06b5da7c132eaed14fefc78718 client/tomcatconf/applicationContext.xml.in 956c77c5fa6c3d909be1fbee8ba26e8d8ba1dce7 client/tomcatconf/componentContext.xml.in f36d0ee02a2eb91733f6f2a76be60a1477529d4d client/tomcatconf/simulatorComponentContext.xml.in c46d1de086a5633b64c90884dbc02a03c3b2e8ad plugins/hypervisors/hyperv/DotNet/ServerResource/.gitignore PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.Config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.targets PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.settings PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShell.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShellException.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Program.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/CloudStackTypes.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResource.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResourceController.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Utils.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/WmiCalls.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/HypervResourceControllerTest.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/ServerResource.Tests.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.sln PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_OperatingSystem.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_Processor.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.virtualization.Msvm_ComputerSystem.cs PRE-CREATION
Re: Review Request 13922: Minimal Hyper-V Plugin
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13922/ --- (Updated Sept. 5, 2013, 8:29 a.m.) Review request for cloudstack, Alex Huang, Chiradeep Vittal, and Devdeep Singh. Repository: cloudstack-git Description --- A plugin for Hyper-V control is available for CloudStack. The plugin implements basic VM control; however, its architecture allows additional functionality to be easily added. Incorporating the plugin in CloudStack will allow the community to participate in improving the features available with Hyper-V. The plugin uses a Director Connect Agent architecture described here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Progress This links provides details of third party libraries used by the plugin including their licenses. No source for these libraries is used, and the binaries are downloaded from their distributors at build time. No proprietary tools are required for the build. For instance, C# compiled with Mono has been tested. Therefore, the plugin has been added to the default build and the default deployment. The plugin includes unit and functional tests that can be triggered at build time. See the link above for details. The Apache header is applied to source and where feasible to build config files. The source is the https://github.com/lafferty/cloudstack/tree/hyperv_plugin Diffs - client/pom.xml 1afe5b898d328b06b5da7c132eaed14fefc78718 client/tomcatconf/applicationContext.xml.in 956c77c5fa6c3d909be1fbee8ba26e8d8ba1dce7 client/tomcatconf/componentContext.xml.in f36d0ee02a2eb91733f6f2a76be60a1477529d4d client/tomcatconf/simulatorComponentContext.xml.in c46d1de086a5633b64c90884dbc02a03c3b2e8ad plugins/hypervisors/hyperv/DotNet/ServerResource/.gitignore PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.Config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/.nuget/NuGet.targets PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentService.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.Designer.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentSettings.settings PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShell.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/AgentShellException.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Program.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/AgentShell/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/CloudStackTypes.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResource.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/HypervResourceController.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/Utils.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/WmiCalls.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/HypervResource/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/App.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/HypervResourceControllerTest.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/ServerResource.Tests.csproj PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.Tests/packages.config PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/ServerResource.sln PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/Properties/AssemblyInfo.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_OperatingSystem.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.CIMV2.Win32_Processor.cs PRE-CREATION plugins/hypervisors/hyperv/DotNet/ServerResource/WmiWrappers/ROOT.virtualization.Msvm_ComputerSystem.cs PRE-CREATION
RE: SystemVM downloading
Donal, Great to know that you would be doing changes to systemvm template. I am curious to know how the host data will be passed to the systemvm's running in hyperv. In my initial investigation I figured out with Integration Services installed in the Linux, we can get the guest details like (version etc) using KVP component. I was not able to figure out how we can send the data/values from host to systemvm's. I had read for windows based guest vm's we can send data from host to guest and we can see those in the guest registry. Once you figured out how to handle for linux please share it, it's very interesting to know. Thanks Rajesh Battala -Original Message- From: Donal Lafferty [mailto:donal.laffe...@citrix.com] Sent: Thursday, September 5, 2013 1:52 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Great! I may need to update the template to allow it to accept host data from a Hyper-V 2012. Unlike XenServer, a Hyper-V host passes data through a user mode daemon that write key/value pairs to a file. I need to investigate a bit further, but I believe that the daemon starts and startup scripts need updating to deal with the key/value pair file format. DL -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: 05 September 2013 05:54 To: dev@cloudstack.apache.org Subject: Re: SystemVM downloading Hey Donal, Soheil, I had to remove those artifacts from jenkins as they occupied a lot of space on the instance which is a m1.small with a small disk. I've been looking for some mirroring space or s3 bucket where I can push the system VMs. You could meanwhile build your own as described here if you want to experiment with the latest. https://cwiki.apache.org/confluence/x/UlHVAQ I'll see if I can put up the hyperv vhd somewhere meanwhile. On Wed, Sep 04, 2013 at 09:16:05PM +, Rayees Namathponnan wrote: Hi Soheil, You can get the template from below location; it's not latest but tested extensively http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06- 12-ma ster-kvm.qcow2.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-2013-07- 12-ma ster-xen.vhd.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-4.2- vh7.ova Regards, Rayees -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Wednesday, September 04, 2013 1:14 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Not just Hyper-V but looks like none of the sytemvm images e.g. Xen/VMware are posted on the site. -Soheil From: Donal Lafferty [donal.laffe...@citrix.com] Sent: Wednesday, September 04, 2013 12:35 PM To: dev@cloudstack.apache.org Subject: SystemVM downloading I thought I'd be able to download systemvmtemplate-2013-09-03-master-hyperv.vhd from http://jenkins.cloudstack.org/view/master/job/build-systemvm-master/ However, I don't see a link to do so. Is systemvmtemplate-2013-09-03-master-hyperv.vhd or other recent copy online anywhere? DL -- Prasanna., Powered by BigRock.com
RE: SystemVM downloading
The daemon involved in KVP transfer is mentioned here http://dlafferty.blogspot.co.uk/2013/07/installing-hyper-v-pv-drivers-for-linux.html I'll write an update discussing how to use it to pass system VM settings. DL -Original Message- From: Rajesh Battala [mailto:rajesh.batt...@citrix.com] Sent: 05 September 2013 09:47 To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Donal, Great to know that you would be doing changes to systemvm template. I am curious to know how the host data will be passed to the systemvm's running in hyperv. In my initial investigation I figured out with Integration Services installed in the Linux, we can get the guest details like (version etc) using KVP component. I was not able to figure out how we can send the data/values from host to systemvm's. I had read for windows based guest vm's we can send data from host to guest and we can see those in the guest registry. Once you figured out how to handle for linux please share it, it's very interesting to know. Thanks Rajesh Battala -Original Message- From: Donal Lafferty [mailto:donal.laffe...@citrix.com] Sent: Thursday, September 5, 2013 1:52 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Great! I may need to update the template to allow it to accept host data from a Hyper-V 2012. Unlike XenServer, a Hyper-V host passes data through a user mode daemon that write key/value pairs to a file. I need to investigate a bit further, but I believe that the daemon starts and startup scripts need updating to deal with the key/value pair file format. DL -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: 05 September 2013 05:54 To: dev@cloudstack.apache.org Subject: Re: SystemVM downloading Hey Donal, Soheil, I had to remove those artifacts from jenkins as they occupied a lot of space on the instance which is a m1.small with a small disk. I've been looking for some mirroring space or s3 bucket where I can push the system VMs. You could meanwhile build your own as described here if you want to experiment with the latest. https://cwiki.apache.org/confluence/x/UlHVAQ I'll see if I can put up the hyperv vhd somewhere meanwhile. On Wed, Sep 04, 2013 at 09:16:05PM +, Rayees Namathponnan wrote: Hi Soheil, You can get the template from below location; it's not latest but tested extensively http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06- 12-ma ster-kvm.qcow2.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-2013-07- 12-ma ster-xen.vhd.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-4.2- vh7.ova Regards, Rayees -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Wednesday, September 04, 2013 1:14 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Not just Hyper-V but looks like none of the sytemvm images e.g. Xen/VMware are posted on the site. -Soheil From: Donal Lafferty [donal.laffe...@citrix.com] Sent: Wednesday, September 04, 2013 12:35 PM To: dev@cloudstack.apache.org Subject: SystemVM downloading I thought I'd be able to download systemvmtemplate-2013-09-03-master-hyperv.vhd from http://jenkins.cloudstack.org/view/master/job/build-systemvm-master/ However, I don't see a link to do so. Is systemvmtemplate-2013-09-03-master-hyperv.vhd or other recent copy online anywhere? DL -- Prasanna., Powered by BigRock.com
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Congratulations! On 05-Sep-2013, at 12:21 PM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
RE: [ANNOUNCE] New PMC member: Ilya Musayev
Congratulations Ilya. Regards, Saksham -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 12:22 PM To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
RE: SystemVM downloading
Thanks Donal, I had gone through your blog its very informative. Once the KVP daemon is running inside linux guest then we should be able to transfer/exchange data between hosts and linux guest right.! Thanks Rajesh Battala -Original Message- From: Donal Lafferty [mailto:donal.laffe...@citrix.com] Sent: Thursday, September 5, 2013 2:29 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading The daemon involved in KVP transfer is mentioned here http://dlafferty.blogspot.co.uk/2013/07/installing-hyper-v-pv-drivers-for-linux.html I'll write an update discussing how to use it to pass system VM settings. DL -Original Message- From: Rajesh Battala [mailto:rajesh.batt...@citrix.com] Sent: 05 September 2013 09:47 To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Donal, Great to know that you would be doing changes to systemvm template. I am curious to know how the host data will be passed to the systemvm's running in hyperv. In my initial investigation I figured out with Integration Services installed in the Linux, we can get the guest details like (version etc) using KVP component. I was not able to figure out how we can send the data/values from host to systemvm's. I had read for windows based guest vm's we can send data from host to guest and we can see those in the guest registry. Once you figured out how to handle for linux please share it, it's very interesting to know. Thanks Rajesh Battala -Original Message- From: Donal Lafferty [mailto:donal.laffe...@citrix.com] Sent: Thursday, September 5, 2013 1:52 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Great! I may need to update the template to allow it to accept host data from a Hyper-V 2012. Unlike XenServer, a Hyper-V host passes data through a user mode daemon that write key/value pairs to a file. I need to investigate a bit further, but I believe that the daemon starts and startup scripts need updating to deal with the key/value pair file format. DL -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: 05 September 2013 05:54 To: dev@cloudstack.apache.org Subject: Re: SystemVM downloading Hey Donal, Soheil, I had to remove those artifacts from jenkins as they occupied a lot of space on the instance which is a m1.small with a small disk. I've been looking for some mirroring space or s3 bucket where I can push the system VMs. You could meanwhile build your own as described here if you want to experiment with the latest. https://cwiki.apache.org/confluence/x/UlHVAQ I'll see if I can put up the hyperv vhd somewhere meanwhile. On Wed, Sep 04, 2013 at 09:16:05PM +, Rayees Namathponnan wrote: Hi Soheil, You can get the template from below location; it's not latest but tested extensively http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06- 12-ma ster-kvm.qcow2.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-2013-07- 12-ma ster-xen.vhd.bz2 http://download.cloud.com/templates/4.2/systemvmtemplate-4.2- vh7.ova Regards, Rayees -Original Message- From: Soheil Eizadi [mailto:seiz...@infoblox.com] Sent: Wednesday, September 04, 2013 1:14 PM To: dev@cloudstack.apache.org Subject: RE: SystemVM downloading Not just Hyper-V but looks like none of the sytemvm images e.g. Xen/VMware are posted on the site. -Soheil From: Donal Lafferty [donal.laffe...@citrix.com] Sent: Wednesday, September 04, 2013 12:35 PM To: dev@cloudstack.apache.org Subject: SystemVM downloading I thought I'd be able to download systemvmtemplate-2013-09-03-master-hyperv.vhd from http://jenkins.cloudstack.org/view/master/job/build-systemvm-maste r/ However, I don't see a link to do so. Is systemvmtemplate-2013-09-03-master-hyperv.vhd or other recent copy online anywhere? DL -- Prasanna., Powered by BigRock.com
Re: Review Request 13701: Automation Tests for HA Proxy Stickiness
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13701/#review25925 --- test/integration/component/test_haproxy.py https://reviews.apache.org/r/13701/#comment50605 can we move configs dictionary to top along with lb_methods? test/integration/component/test_haproxy.py https://reviews.apache.org/r/13701/#comment50606 Didn't quite understand what this test is qualifying - if we create a sticky policy when router is down, we can see the same immediately with listLBStickinessPolicies even before the router starts. - I think the test adds value if we can login to the router and check whether the policy is configure once the router is back online. test/integration/component/test_haproxy.py https://reviews.apache.org/r/13701/#comment50607 destroying a router shouldn't delete the stickiness policy but the description says so. Can you confirm whether you saw some behaviour like that? test/integration/component/test_haproxy.py https://reviews.apache.org/r/13701/#comment50608 there is no test for #adding same rule again though the doc string says there is. - venkata swamy babu budumuru On Aug. 27, 2013, 5:36 a.m., Girish Shilamkar wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13701/ --- (Updated Aug. 27, 2013, 5:36 a.m.) Review request for cloudstack, suresh sadhu and Prasanna Santhanam. Repository: cloudstack-git Description --- Automation Tests for HA Proxy Stickiness Diffs - test/integration/component/test_haproxy.py PRE-CREATION Diff: https://reviews.apache.org/r/13701/diff/ Testing --- Thanks, Girish Shilamkar
Re: Review Request 13889: CLOUDSTACK-4487: Add Netscaler service provider if not already present
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/#review25928 --- Ship it! Ship It! - venkata swamy babu budumuru On Aug. 28, 2013, 5:59 p.m., Sowmya Krishnan wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/ --- (Updated Aug. 28, 2013, 5:59 p.m.) Review request for cloudstack, venkata swamy babu budumuru and Prasanna Santhanam. Bugs: CLOUDSTACK-4487 Repository: cloudstack-git Description --- For a brand new set up, Netscaler service provider won't be present by defualt. Fixed the same in common.py. Diffs - tools/marvin/marvin/integration/lib/common.py f27e87d Diff: https://reviews.apache.org/r/13889/diff/ Testing --- Tested on both new MS and an existing one. Thanks, Sowmya Krishnan
Re: cpvm and ssvm but no host entries
On Thu, Sep 05, 2013 at 02:21:24AM -0700, Darren Shepherd wrote: On 09/05/2013 01:14 AM, Prasanna Santhanam wrote: If you are looking for the SecondaryStorage nfs host entry (that usually would go into Alert state soon as it was added), it no more goes into the host table. There is no 'discovery' process for storage. Also the AddSecondaryStorageCmd is now deprecated and the AddImageStoreCmd and AddStagingStoreCmds are used for the backing store operations. Ah good to know. I actually was using addImageStore and I didn't know it. Basically I tried to setup an advanced zone by hand using cloudmonkey and quickly found out that that task was never intended to be done by mere mortals. So I've been using the GUI to setup the zone through the wizard. You can use the marvin [1] test client for quick configuration. This is what is used within the Citrix QA team and in our automated test deployments to setup cloudstack. There are example configurations available in setup/dev/*.cfg in the repo. [1] https://cwiki.apache.org/confluence/x/QQzMAQ (See section on Marvin Deployment Configurations) Regardless, I got it working. The hosts table entry is created when the SSVM calls home through 8250. So basically what was happening is that the IP was wrong for the management server. Darren -- Prasanna., Powered by BigRock.com
Re: Review Request 13889: CLOUDSTACK-4487: Add Netscaler service provider if not already present
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/#review25930 --- Commit 8b5cb374b93a1821216f8fba9fd5e908905a3ce8 in branch refs/heads/master from Sowmya Krishnan [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=8b5cb37 ] CLOUDSTACK-4487 Fix adding Netscaler service provider if not already done Signed-off-by: venkataswamybabu budumuru venkataswamybabu.budum...@citrix.com (cherry picked from commit c1ea61a0e736ce626c0d1f21e90cf9aaf7f89964) - ASF Subversion and Git Services On Aug. 28, 2013, 5:59 p.m., Sowmya Krishnan wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/ --- (Updated Aug. 28, 2013, 5:59 p.m.) Review request for cloudstack, venkata swamy babu budumuru and Prasanna Santhanam. Bugs: CLOUDSTACK-4487 Repository: cloudstack-git Description --- For a brand new set up, Netscaler service provider won't be present by defualt. Fixed the same in common.py. Diffs - tools/marvin/marvin/integration/lib/common.py f27e87d Diff: https://reviews.apache.org/r/13889/diff/ Testing --- Tested on both new MS and an existing one. Thanks, Sowmya Krishnan
Re: Review Request 13889: CLOUDSTACK-4487: Add Netscaler service provider if not already present
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/#review25929 --- Commit c1ea61a0e736ce626c0d1f21e90cf9aaf7f89964 in branch refs/heads/4.2-forward from Sowmya Krishnan [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=c1ea61a ] CLOUDSTACK-4487 Fix adding Netscaler service provider if not already done Signed-off-by: venkataswamybabu budumuru venkataswamybabu.budum...@citrix.com - ASF Subversion and Git Services On Aug. 28, 2013, 5:59 p.m., Sowmya Krishnan wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13889/ --- (Updated Aug. 28, 2013, 5:59 p.m.) Review request for cloudstack, venkata swamy babu budumuru and Prasanna Santhanam. Bugs: CLOUDSTACK-4487 Repository: cloudstack-git Description --- For a brand new set up, Netscaler service provider won't be present by defualt. Fixed the same in common.py. Diffs - tools/marvin/marvin/integration/lib/common.py f27e87d Diff: https://reviews.apache.org/r/13889/diff/ Testing --- Tested on both new MS and an existing one. Thanks, Sowmya Krishnan
Review Request 13991: Made misc changes in test_multiple_ip_ranges.py
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13991/ --- Review request for cloudstack, venkata swamy babu budumuru, SrikanteswaraRao Talluri, and Prasanna Santhanam. Bugs: CLOUDSTACK-702 Repository: cloudstack-git Description --- This patch is a combination of two commits. In first commit 1. Removed advanced tag from all the tests. At the movement scripts support only basic zone 2. Added a test to deploy guest vm after adding ip range in new CIDR In second commit: Removed test for deploying vm after adding new cidr 1.This test would block parallel execution of remianing tests 2.Removed list_Routers proc and unused code 3.Added few more debug statements Diffs - test/integration/component/test_multiple_ip_ranges.py 18409c5 Diff: https://reviews.apache.org/r/13991/diff/ Testing --- Yes. Thanks, sanjeev n
Re: Change network functionality?
Thank you for your response Chiradeep. I have not tried to use API (yet). Even if i can use something like cloudmonkey to use API from CLI, my users can not. So my question still stands, when it will be or if it's already possible to change network(s) using CS web GUI? And additional question: is it currently possible using API with cloudmonkey on 4.1.1? Has anyone tried? Tnx. F. On 4/9/13 10:45 PM, Chiradeep Vittal wrote: Have you tried addNicToVirtualMachine updateDefaultNicForVirtualMachine removeNicFromVirtualMachine http://goo.gl/qUKIgT On 9/4/13 5:50 AM, France mailingli...@isg.si wrote: Hi all. We're use CS 4.1.1. I think there is no way to change network or networks to which VM belongs _after_ it was created. To change the network in which it is, it's required to create template from it and then a new VM. Is there another way? Is functionality to change network in which VM is, in plan for any future releases? If not, would anyone consider adding it? Regards, F.
FYI, issue in Portable IP feature
Hello, We found an issue while automating testcases for Portable IP. Suppose I have existing public IP range with a VLAN, and I use this range (or subset of this range) for creating a new portable IP range, portable IP range gets created successfully, but if you try to associate a portable IP to a network, it throws error saying Entity already exists. This happens because Portable IP range should not overlap with any of existing public ip ranges (irrespective of their vlan id). Ideally, CS should not allow two public or portable IP ranges to exist within same subnet (same netmask). Each range should have different netmask (hence different subnet). This also means that two public (or portable) ranges with same start and end IPs even in different VLAN ids cannot exist. This should be fixed in the 4.2.1. https://issues.apache.org/jira/browse/CLOUDSTACK-4596 If we have to create a new portable or public IP range through code, we can't create it by finding out free IPs range in existing subnets. Hence will need to define it somewhere i.e. in the config file used by automation testcases. Regards, Girish
RE: Change network functionality?
In 4.2, UI functionality for adding/removing and updating default NIC has been added. You may find it in the NIC tab of an instance. I haven't explicitly tried the APIs on 4.1.1, but the APIs should be present there also. Thanks, Saksham -Original Message- From: France [mailto:mailingli...@isg.si] Sent: Thursday, September 05, 2013 3:50 PM To: dev@cloudstack.apache.org Cc: Chiradeep Vittal Subject: Re: Change network functionality? Thank you for your response Chiradeep. I have not tried to use API (yet). Even if i can use something like cloudmonkey to use API from CLI, my users can not. So my question still stands, when it will be or if it's already possible to change network(s) using CS web GUI? And additional question: is it currently possible using API with cloudmonkey on 4.1.1? Has anyone tried? Tnx. F. On 4/9/13 10:45 PM, Chiradeep Vittal wrote: Have you tried addNicToVirtualMachine updateDefaultNicForVirtualMachine removeNicFromVirtualMachine http://goo.gl/qUKIgT On 9/4/13 5:50 AM, France mailingli...@isg.si wrote: Hi all. We're use CS 4.1.1. I think there is no way to change network or networks to which VM belongs _after_ it was created. To change the network in which it is, it's required to create template from it and then a new VM. Is there another way? Is functionality to change network in which VM is, in plan for any future releases? If not, would anyone consider adding it? Regards, F.
Re: [ANNOUNCE] New PMC member: Ilya Musayev
At it Ilya, how about a 'speech' on your take on the future of cloudstack? congrats, Daan On Thu, Sep 5, 2013 at 11:15 AM, Saksham Srivastava saksham.srivast...@citrix.com wrote: Congratulations Ilya. Regards, Saksham -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 12:22 PM To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: Questions about 4.2 RC
On Thu, Sep 5, 2013 at 10:21 AM, sebgoa run...@gmail.com wrote: Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? I suggested this and got no answers, meaning +1 on this one. 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. Is this due to nonoss remnants? it sounds like it (*Vmware*) 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. -Sebastien
Re: documentation/wiki is a mess
-1 on no docs no submits. Docs are important to people that need a contribution they didn't write themselves. The first ones are the ones to write docs where missing. You contribute because you need code, not because you need docs on it. On Thu, Sep 5, 2013 at 9:25 AM, Giles Sirett giles.sir...@shapeblue.com wrote: +1 on no docs, no submit Kind Regards Giles D: +44 20 3603 0541 | M: +44 796 111 2055 giles.sir...@shapeblue.com -Original Message- From: Sebastien Goasguen [mailto:run...@gmail.com] Sent: 05 September 2013 07:14 To: dev@cloudstack.apache.org Cc: aemne...@gmail.com Subject: Re: documentation/wiki is a mess On Sep 5, 2013, at 1:11 AM, Prasanna Santhanam t...@apache.org wrote: Radhika, what I meant was everyone should help out with docs. Esp. those working on a feature should care most that the docs for their feature is perfect for an end user to understand, implement and use. We shouldn't piggy-back on those helping fix doc bugs all the time which I see happening too often. Someone files a doc bug and someone else fixes it and someone else reviews it and finally the users are still having trouble understanding it. We're just creating work for ourselves that way. Feature specs have architectural and implementation details and may not often be fully baked by the time the feature starts development. Changes happen during implementation and reviewing our docs after everything is merged is a good forcing-function to fix our docs. I'm +1 for style guides and markdown based docs. They should make this whole process a lot easier for everyone. I am a big +1 for literally forcing devs to submit docs for user facing features, otherwise the feature does not get added. Just like unit tests. No unitests, no commit, no docs, no commit. The devs are the best to write the docs of their own feature. On Wed, Sep 04, 2013 at 09:52:03PM -0700, Ahmad Emneina wrote: I think prasanna hit the nail on the head. I'm sure there are features there no one knows about, or ever will... +1 for developers documenting what features/functions are abound. On Wed, Sep 4, 2013 at 9:36 PM, Radhika Puthiyetath radhika.puthiyet...@citrix.com wrote: Hi, We do not have a doc team as such :-) We have a set of doc contributors that so far have not worked as a team. Probably, we should think about aligning the doc efforts and having a process and style guide in place. If the FS is good enough, we need not trouble the code committers to write own docs is what I feel. Regards -Radhika -Original Message- From: Prasanna Santhanam [mailto:t...@apache.org] Sent: Thursday, September 05, 2013 9:34 AM To: dev@cloudstack.apache.org Subject: Re: documentation/wiki is a mess On Tue, Sep 03, 2013 at 11:10:28AM -0700, Darren Shepherd wrote: The state of documentation of an open source projects says a lot about the community that develops it... Esp. for XML documentation: I say we write our own docs if we want our feature to be used. Or it dies a natural death in quiet isolation with no one ever using it. Docs team can handle the editing and organizing bits. That should go for Wikis too. If you want the wiki to be useful then write it, organize it and maintain it. Don't just put it there to fill a template. It'll never recieve any love. -- Prasanna., Powered by BigRock.com -- Prasanna., Powered by BigRock.com This email and any attachments to it may be confidential and are intended solely for the use of the individual to whom it is addressed. Any views or opinions expressed are solely those of the author and do not necessarily represent those of Shape Blue Ltd or related companies. If you are not the intended recipient of this email, you must neither take any action based upon its contents, nor copy or show it to anyone. Please contact the sender if you believe you have received this email in error. Shape Blue Ltd is a company incorporated in England Wales. ShapeBlue Services India LLP is operated under license from Shape Blue Ltd. ShapeBlue is a registered trademark.
Review Request 13992: (CLOUDSTACK-4405) additional patch for bridge name and firewall rules issues after KVM upgrade to 4.2
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13992/ --- Review request for cloudstack and edison su. Bugs: CLOUDSTACK-4405 Repository: cloudstack-git Description --- There still exist two issues after Edison's commits. (1) Migration from new hosts to old hosts failed. The bridge name on old host is set to cloudVirBr* if network.bridge.name.schema is set to 3.0 in /etc/cloudstack/agent/agent.properties, but the actual bridge name is breth*-* after running cloudstack-agent-upgrade. (2) all ports of vms (Basic zone, or Advanced zone with security groups) on old hosts are open, because the iptables rules are binding to device (bridge) name which is changed by cloudstack-agent-upgrade. Diffs - agent/bindir/cloudstack-agent-upgrade.in 4972d39 plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/BridgeVifDriver.java e3779a7 scripts/vm/network/security_group.py 0ac8b74 Diff: https://reviews.apache.org/r/13992/diff/ Testing --- tested ok on my environment. After this, the KVM upgrade steps : a. Install 4.2 cloudstack agent on each kvm host b. Run cloudstack-agent-upgrade. This script will upgrade all the existing bridge name to new bridge name, and update related firewall rules. c. install a libvirt hook: c1. mkdir /etc/libvirt/hooks c2. cp /usr/share/cloudstack-agent/lib/libvirtqemuhook /etc/libvirt/hooks/qemu c3. chmod +x /etc/libvirt/hooks/qemu c4. service libvirtd restart c5. service cloudstack-agent restart Thanks, Wei Zhou
Re: Password prompt like nine times
From my /etc/sudoers (btw, use visudo to edit with) cloud ALL =NOPASSWD : ALL In my /etc/sudoers.d/cloudstack cloud ALL =NOPASSWD : ALL Seems redundant but that was what the install script put in. T On Thu, Sep 5, 2013 at 3:05 AM, sebgoa run...@gmail.com wrote: On Sep 5, 2013, at 1:30 AM, Kelven Yang kelven.y...@citrix.com wrote: You need to make sudo-ers no-password required bad idea, someone compromises your account and he is root. Kelven On 9/4/13 4:10 PM, Tracy Phillips tracp...@mantoso.com wrote: On Wed, Sep 4, 2013 at 6:47 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Hi, I think this has been discussed a few times on the list before. When I start up the CSMS, I get prompted for my password about nine times. I thought updating the sudoers file would fix this, but it still happens. Any thoughts on this? Thanks! Mike, Do you mean the Cloudstack Management Server when you say CSMS? What do you have in your sudoers file? Tracy
Re: Password prompt like nine times
On Thu, Sep 5, 2013 at 3:05 AM, sebgoa run...@gmail.com wrote: On Sep 5, 2013, at 1:30 AM, Kelven Yang kelven.y...@citrix.com wrote: You need to make sudo-ers no-password required bad idea, someone compromises your account and he is root. True. But they are on your box unauthorized, its just a matter of time before it is cracked.
Re: [VOTE] Apache CloudStack 4.2.0 (fourth round)
+1 Testing done in OEL64*/KVM environment using a small foot print OEL64 VM image with userdata and reporting to central server to verify that VMs comes up properly. KVM hosts are 32 core / 256 GByte RAM / 1GBit / localstorage on fast RAID (8x600G;RAID5). Manager runs in VirtualBox with local MySQL. - up hypervisor VM limit via API to 100 and do single hypervisor deployment of 80 VMs through 5 cycles (deploy/start/stop/destroy). 2.2 seconds per startVM API call average. 100% hit ratio. - test concurrency of startVM operation, which was more or less non-existing in 41x; 8 KVM hosts, deploy 180 VMs in deploy/start/stop/destroy cycle. Depending on how many concurrent API calls are made the time per startVM call averages from 0.8 to 1.5 seconds during the start VM phase. During this test roughly one VM in 1000 fails to report back to logserver (they are reported as started in ACS MGR). This can be due to any reasons not cloud related but have not digged into this. ACS42 could only recently pass this test due to what seems to be a leakage of ROOT images. E.g., after running the test for a while, a large number of ROOT images are left in expunging state, impossible to remove and they are not cleaned up after 24 hours. I have not seen this in the latest 4.2. This issue was non-deterministic and did not leave any log traces that obviously gave away the cause (have not enabled debug logging since that generates too much data during this type of testing). ACS42 has significantly better concurrency compared to 411 and earlier. Roughly 10 times faster. The setup used in the testing is basic networking with security groups. For 411 the average startVM call times are in the 11+ seconds ball park, never lower. This figure does not change when adding more hosts. For this ACS42 test, the test image is small and quick to start so when adding more hosts, the gain eventually is zero since the java MGR process and MySQL on the manager head hits the CPU limits (for this test 4 cores and 8 GByte RAM was used, recent Intel Xeon CPUs). Upped the MGR java memory to 4G during this test. The default 2G is too little. When hitting the setup with 160 concurrent API operations the resident set size of the java mgr process goes to 2.6 Gbyte. The stopVM operation is run at max concurrency. Other operations were run at 40 concurrent operations active (via throttling). Standard ACS42 API binding was used under python 2.7.5 (python 2.6 is not thread safe and does not work) Having enough RAM in MGR for the test load is important. If RAM is not enough, odd things happens during test runs (failed API calls). No new features tested. Just plain basic churn of the VM deploy/start/stop/destroy cycle. /Ove * OEL64 environment using updates done roughly a week ago On 09/04/2013 06:43 AM, Animesh Chaturvedi wrote: I've created a 4.2.0 release, with the following artifacts up for a vote: Git Branch and Commit SH: https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;a=shortlog;h=refs/heads/4.2 Commit: e39a7d8e0d3f2fd3e326b1bdf4aaf9ba5d900b02 List of changes: https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;a=blob_plain;f=CHANGES;hb=4.2 Source release (checksums and signatures are available at the same location): https://dist.apache.org/repos/dist/dev/cloudstack/4.2.0/ PGP release keys (signed using 94BE0D7C): https://dist.apache.org/repos/dist/release/cloudstack/KEYS Testing instructions are here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/Release+test+procedure Vote will be open for 72 hours (Friday, 9/6 10:00 PM PST). For sanity in tallying the vote, can PMC members please be sure to indicate (binding) with their vote? [ ] +1 approve [ ] +0 no opinion [ ] -1 disapprove (and reason why) Thanks Animesh
Re: Review Request 13964: CLOUDSTACK-4203: Adding a test for migrating volumes of stopped vms to test_stopped_vm.py
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13964/#review25931 --- Commit dfee47e3b6c35d08cba7f28433b9e1bce12b4078 in branch refs/heads/4.2-forward from sanjeevneelarapu [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=dfee47e ] CLOUDSTACK-4203: Adding a test for migrating volumes of stopped vms to test_stopped_vm.py Signed-off-by: sanjeevneelarapu sanjeev.neelar...@citrix.com Signed-off-by: venkataswamybabu budumuru venkataswamybabu.budum...@citrix.com s - ASF Subversion and Git Services On Sept. 5, 2013, 6:32 a.m., sanjeev n wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13964/ --- (Updated Sept. 5, 2013, 6:32 a.m.) Review request for cloudstack, venkata swamy babu budumuru, SrikanteswaraRao Talluri, and Prasanna Santhanam. Bugs: CLOUDSTACK-4203 Repository: cloudstack-git Description --- Adding test for migrating volumes of stopped vm. Diffs - test/integration/component/test_stopped_vm.py 41eeb46 Diff: https://reviews.apache.org/r/13964/diff/ Testing --- Yes Thanks, sanjeev n
Re: Review Request 13964: CLOUDSTACK-4203: Adding a test for migrating volumes of stopped vms to test_stopped_vm.py
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13964/#review25932 --- Commit 883b9802d4a58633ae9bc75023e063751282783a in branch refs/heads/master from sanjeevneelarapu [ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=883b980 ] CLOUDSTACK-4203: Adding a test for migrating volumes of stopped vms to test_stopped_vm.py Signed-off-by: sanjeevneelarapu sanjeev.neelar...@citrix.com Signed-off-by: venkataswamybabu budumuru venkataswamybabu.budum...@citrix.com s (cherry picked from commit dfee47e3b6c35d08cba7f28433b9e1bce12b4078) - ASF Subversion and Git Services On Sept. 5, 2013, 6:32 a.m., sanjeev n wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13964/ --- (Updated Sept. 5, 2013, 6:32 a.m.) Review request for cloudstack, venkata swamy babu budumuru, SrikanteswaraRao Talluri, and Prasanna Santhanam. Bugs: CLOUDSTACK-4203 Repository: cloudstack-git Description --- Adding test for migrating volumes of stopped vm. Diffs - test/integration/component/test_stopped_vm.py 41eeb46 Diff: https://reviews.apache.org/r/13964/diff/ Testing --- Yes Thanks, sanjeev n
maven storage plugin project creation
Hi folks, I am in the process of creating a new storage plugin maven module need help in doing so. Do we follow any maven based archetype for such maven module creation ? Are there any links that prescribe the required steps ? I was looking at some of the pom files of default, sample solidfire modules present in plugins/storage/volume/ found neither of them have a group-id. Is this the desired convention for a cloudstack module ? Regards, Amit *CloudByte Inc.* http://www.cloudbyte.com/
Re: Password prompt like nine times
Thanks, everyone! On Thu, Sep 5, 2013 at 7:10 AM, Tracy Phillips tracp...@mantoso.com wrote: On Thu, Sep 5, 2013 at 3:05 AM, sebgoa run...@gmail.com wrote: On Sep 5, 2013, at 1:30 AM, Kelven Yang kelven.y...@citrix.com wrote: You need to make sudo-ers no-password required bad idea, someone compromises your account and he is root. True. But they are on your box unauthorized, its just a matter of time before it is cracked. -- *Mike Tutkowski* *Senior CloudStack Developer, SolidFire Inc.* e: mike.tutkow...@solidfire.com o: 303.746.7302 Advancing the way the world uses the cloudhttp://solidfire.com/solution/overview/?video=play *™*
RE: [DISCUSS] OOM killer and Routing/System VM's = :(
Well, you can't as far as I've looked in the source of panic.c. So I'm thinking of investigating of adding -1 as an option and seeing if I can push halt in, let's hope the guys that do kernel stuff find this useful too. So it seems the patch, I conjured up for panic.c, is seen as not so useful, there is however another way to achieve the same result. This would mean that we load a crash kernel with our own .sh script as init to do our bidding. Would that be a plan ? Cheers, Funs Sent from my iPhone On 4 sep. 2013, at 23:35, Marcus Sorensen shadow...@gmail.com wrote: What would work as a quick fix for this sort of situation would be if the machine could be configured to power off rather than rebooting on oom. Then the HA system would restart the VM, applying all configs. Anyone know how to do that? :-) On Wed, Sep 4, 2013 at 1:14 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: On 09/04/2013 11:37 AM, Roeland Kuipers wrote: Hi Darren, Thanks for your reply! Could you share a bit more on your plans/ideas? We also have been braining on other approaches of managing the systemvm's, especially small customizations for specific tenants. And maybe even leveraging a config mgmt tools like chef or puppet with the ability to integrate CS with that in some way. I'll have to send the full details later but here's a rough idea. The basic approach is this. Logical changes to the VRs (or system vms in general) get mapped to configuration items. So add a LB rule maps to iptables config and haproxy config. When you change a LB rule we then bump up the requested version of the configuration for iptables/haproxy. So the requested version will be 4 maybe. The applied version will be 3 as the VR still has the old configuration. Since 4 != 3, the VR will be signaled to pull the latest iptables/haproxy config. So it will pull the configuration. Say in the mean time somebody else adds four other LB rules. So the requested version is now at 8. So when the VR pulls the config it will get version 8, and then reply back saying it applied version 8. The applied version is now 8 which is greater than 4 (the version the first LB rule change was waiting for) so basically all async jobs waiting for the LB change will be done. To pull the configuration from the VR, the VR will be hitting a templating configuration system. So it pulls the full iptables and haproxy config. Not incremental changes. So if the VR ever reboots itself, it can easily just pull the latest config of everything and apply it. So it will be consistent. I'd be interested to hear what type of customizations you would like to add. It will definitely be an extensible system, but the problem is if your extensions wants to touch the same configuration files that ACS wants to manage. That gets a bit tricky as its really easy for each to break each other. But I can definitely add some hooks that users can use to mess up things and void the warranty. I've thought about chef and puppet for this, but basically it comes down to two things. I'm really interested in this being fast and light weight. Ruby is neither of those. So the core ACS stuff will probably remain as very simple shell scripts. Simple in that they really just need to download configuration and restart services. They know nothing about the nature of the changes. If, as an extension, you want to do something with puppet, chef, I'd be open to that. That's your deal. This approach has many other benefits. Like, for example, we can ensure that as we deploy a new ACS release existing system VMs can be updated (without a reboot, unless the kernel changes). Additionally, its fast and updates happen in near constant time. So most changes will be just a couple of seconds, even if you have 4000 LB rules. Darren
[GSoC] Can not access SSVM + CPVM with KVM + Openvswitch
Hi guys, I'm using an unofficial version ACS 4.1 (sdnextensions branch). I build my own cloud-agent package and install successfully into a kvm host and add this host success with two NICs: cloudbr0 (Management, Storage) and cloudbr1 (Guest, Public). I use openvswitch, follow Hugo's guide here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/KVM+with+OpenVSwitch SSVM and CPVM start normally but I can not ping or ssh into via local-link ( 169.254.0.0/16) although both of them are already allocated IP addresses. On MS, local-link nic is cloud0 (169.254.0.1). No error log happen. In DB, they are consistently in starting state, not running. Anyone can help? -- N.g.U.y.e.N.A.n.H.t.U
Re: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch
Ah. Finally error happens: 2013-09-05 22:10:05,748 WARN [storage.secondary.SecondaryStorageManagerImpl] (secstorage-1:null) Exception while trying to start secondary storage vm com.cloud.exception.AgentUnavailableException: Resource [Host:1] is unreachable: Host 1: Unable to start instance due to Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineManagerImpl.java:946) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerImpl.java:557) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerImpl.java:550) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.startSecStorageVm(SecondaryStorageManagerImpl.java:265) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.allocCapacity(SecondaryStorageManagerImpl.java:690) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.expandPool(SecondaryStorageManagerImpl.java:1297) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(PremiumSecondaryStorageManagerImpl.java:123) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(PremiumSecondaryStorageManagerImpl.java:50) at com.cloud.vm.SystemVmLoadScanner.loadScan(SystemVmLoadScanner.java:104) at com.cloud.vm.SystemVmLoadScanner.access$100(SystemVmLoadScanner.java:33) at com.cloud.vm.SystemVmLoadScanner$1.reallyRun(SystemVmLoadScanner.java:81) at com.cloud.vm.SystemVmLoadScanner$1.run(SystemVmLoadScanner.java:72) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) at java.util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:351) at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:722) Caused by: com.cloud.utils.exception.ExecutionException: Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineManagerImpl.java:913) ... 20 more Two system vms are destroyed and created new. 2013/9/5 Nguyen Anh Tu ng.t...@gmail.com Hi guys, I'm using an unofficial version ACS 4.1 (sdnextensions branch). I build my own cloud-agent package and install successfully into a kvm host and add this host success with two NICs: cloudbr0 (Management, Storage) and cloudbr1 (Guest, Public). I use openvswitch, follow Hugo's guide here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/KVM+with+OpenVSwitch SSVM and CPVM start normally but I can not ping or ssh into via local-link (169.254.0.0/16) although both of them are already allocated IP addresses. On MS, local-link nic is cloud0 (169.254.0.1). No error log happen. In DB, they are consistently in starting state, not running. Anyone can help? -- N.g.U.y.e.N.A.n.H.t.U -- N.g.U.y.e.N.A.n.H.t.U
Re: documentation/wiki is a mess
On Thu, Sep 5, 2013 at 7:00 AM, Daan Hoogland daan.hoogl...@gmail.com wrote: -1 on no docs no submits. Docs are important to people that need a contribution they didn't write themselves. The first ones are the ones to write docs where missing. You contribute because you need code, not because you need docs on it. If the developer who wrote the code for a feature can't tell me (or the rest of the userbase) how it works and how to use it, then I question exactly how useful the feature is.
Re: documentation/wiki is a mess
On 09/05/2013 07:56 AM, David Nalley wrote: On Thu, Sep 5, 2013 at 7:00 AM, Daan Hoogland daan.hoogl...@gmail.com wrote: -1 on no docs no submits. Docs are important to people that need a contribution they didn't write themselves. The first ones are the ones to write docs where missing. You contribute because you need code, not because you need docs on it. If the developer who wrote the code for a feature can't tell me (or the rest of the userbase) how it works and how to use it, then I question exactly how useful the feature is. Everyone should be on the hook to document in some fashion. The doc writer are usually just better at it. So as somebody who is more dev focused, I just don't know where to put things. I'm not about to touch the XML docs. That seems like a doc team's domain. So personally I'd like to see a bit more organization in the wiki and then a clear cut definition of when stuff goes to the XML. Additional proposals and design specs don't count as documenting functionality. Those things are just SDLC artifacts. Frankly I think the current design template should have the scope, impact, and QA notes and then link to another place in the wiki with the design. As the development is in progress the wiki page can be marked with WIP. We should be careful about no X, no commit policies. We don't want to discourage contributions. Documentation is useful and if somebody wants to contribute then I think they would be inclined to put some documentation such that it can be used by other people. If we make the process easy and obvious to do such, I think more documentation will exist. Darren
Re: Review Request 13992: (CLOUDSTACK-4405) additional patch for bridge name and firewall rules issues after KVM upgrade to 4.2
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13992/#review25939 --- Thanks for your fix on iptables, patch is OK for me. - edison su On Sept. 5, 2013, 12:32 p.m., Wei Zhou wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13992/ --- (Updated Sept. 5, 2013, 12:32 p.m.) Review request for cloudstack and edison su. Bugs: CLOUDSTACK-4405 Repository: cloudstack-git Description --- There still exist two issues after Edison's commits. (1) Migration from new hosts to old hosts failed. The bridge name on old host is set to cloudVirBr* if network.bridge.name.schema is set to 3.0 in /etc/cloudstack/agent/agent.properties, but the actual bridge name is breth*-* after running cloudstack-agent-upgrade. (2) all ports of vms (Basic zone, or Advanced zone with security groups) on old hosts are open, because the iptables rules are binding to device (bridge) name which is changed by cloudstack-agent-upgrade. Diffs - agent/bindir/cloudstack-agent-upgrade.in 4972d39 plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/BridgeVifDriver.java e3779a7 scripts/vm/network/security_group.py 0ac8b74 Diff: https://reviews.apache.org/r/13992/diff/ Testing --- tested ok on my environment. After this, the KVM upgrade steps : a. Install 4.2 cloudstack agent on each kvm host b. Run cloudstack-agent-upgrade. This script will upgrade all the existing bridge name to new bridge name, and update related firewall rules. c. install a libvirt hook: c1. mkdir /etc/libvirt/hooks c2. cp /usr/share/cloudstack-agent/lib/libvirtqemuhook /etc/libvirt/hooks/qemu c3. chmod +x /etc/libvirt/hooks/qemu c4. service libvirtd restart c5. service cloudstack-agent restart Thanks, Wei Zhou
Re: Review Request 13877: CIFS Secondary Storage
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13877/#review25940 --- Ok for me. - edison su On Aug. 28, 2013, 8:58 p.m., Donal Lafferty wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/13877/ --- (Updated Aug. 28, 2013, 8:58 p.m.) Review request for cloudstack, Chiradeep Vittal, Devdeep Singh, and edison su. Repository: cloudstack-git Description --- CIFS support for secondary storage is documented here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/CIFS+Support The feature was implemented by extending the NFS provider. Its validation was updated so that you can pass it a URL containing the details of a CIFS share. The code that mounts NFS shares was extended to allow it do the same for CIFS shares. Otherwise, the secondary storage code is left unchanged. The source is in the cifs_support branch of https://github.com/lafferty/cloudstack NB: not able to submit using git format-patch due to issue described here: http://stackoverflow.com/questions/14594051/upload-from-git-format-patch-to-reviewboard-fails-on-file-not-found-in-the-repo Diffs - engine/storage/datamotion/src/org/apache/cloudstack/storage/motion/AncientDataMotionStrategy.java d5e8a84516f4f37f23610a3d0315aa0a84991d46 engine/storage/src/org/apache/cloudstack/storage/datastore/protocol/DataStoreProtocol.java b27c96edbb72751c72c3963d6945eaeaeb42c74b plugins/storage/image/default/src/org/apache/cloudstack/storage/datastore/lifecycle/CloudStackImageStoreLifeCycleImpl.java 21a5e0a7194449d8b72f4537e14586c2b65b0bb5 plugins/storage/volume/default/src/org/apache/cloudstack/storage/datastore/lifecycle/CloudStackPrimaryDataStoreLifeCycleImpl.java b9b74243eddd2008227f8ea2b88832e6e276811d server/src/com/cloud/resource/ResourceManagerImpl.java 44ee8e3bd51be38acd6a90fd2fdd2383dad0fd15 server/src/com/cloud/storage/StorageManagerImpl.java b5d6ff953ef5d099fbe4dc56866e5948a46f1c8a server/src/com/cloud/test/DatabaseConfig.java 63f77b66520d5f27736d62a67eeb16af808c7dd5 services/secondary-storage/conf/agent.properties aab82b6337474b52ae26c472b321f2c93ac57e81 services/secondary-storage/conf/log4j.xml PRE-CREATION services/secondary-storage/pom.xml ea4bfcaf477af6be6a01113ac99ef262ae789071 services/secondary-storage/src/org/apache/cloudstack/storage/resource/LocalNfsSecondaryStorageResource.java c55c2361bd5551eb51ca1b14340ab4c2e1563865 services/secondary-storage/src/org/apache/cloudstack/storage/resource/NfsSecondaryStorageResource.java 8b9de39b70981b7fb4458b5a337b1b577c45aa3b services/secondary-storage/src/org/apache/cloudstack/storage/resource/SecondaryStorageDiscoverer.java 5d6d61f740ac69acc228b0712633190b654b8f27 services/secondary-storage/src/org/apache/cloudstack/storage/template/DownloadManagerImpl.java bf68b299c5c233f420ed1f78667f0f9522cf64de services/secondary-storage/test/org/apache/cloudstack/storage/resource/LocalNfsSecondaryStorageResourceTest.java 0c355ec884d50c530d2c77bec85e14550cc57352 services/secondary-storage/test/org/apache/cloudstack/storage/resource/NfsSecondaryStorageResourceTest.java PRE-CREATION utils/src/com/cloud/utils/UriUtils.java 1ff4d729cbf50154688bcea79bb8f48d1a972674 Diff: https://reviews.apache.org/r/13877/diff/ Testing --- Testing involved running the CIFS-specific and existing NFS tests for the secondary-storage service. E.g. cd ./services/secondary-storage mvn clean install -P developer,systemvm, -DskipTests=false Notice that the pom.xml was updated to allow tests to be turned on using a conditional. I ran the tests on the Citrix network, which allowed the existing NFS tests to run without modification. In addition, an integration test was done exercise the code in the SystemVM environment and to exercise updated validation code. The NFS secondary storage integration test was done using a XenServer deployment in which a template was registered and a VM created from the template. The CIFS secondary storage test was done on a Hyper-V hypervisor a template was registered. Thanks, Donal Lafferty
Re: Template Problem
Do you have contents of template_view and vm_templates? Thanks -min On 8/30/13 9:25 AM, Soheil Eizadi seiz...@infoblox.com wrote: It was in the original email, here it is by itself: mysql select * from template_store_ref; ++--+-+-+- ++--+++---+--- -+---++--- -+ +---+---+-+--- ---+-+-+ | id | store_id | template_id | created | last_updated | job_id | download_pct | size | store_role | physical_size | download_state | error_str | local_path | install_path | url | state | destroyed | is_copy | update_count | ref_cnt | updated | ++--+-+-+- ++--+++---+--- -+---++--- -+ +---+---+-+--- ---+-+-+ | 5 |1 | 1 | 2013-08-28 23:10:50 | 2013-08-29 21:01:19 | NULL | 100 | 2101252608 | Image |2101252608 | DOWNLOADED | NULL | NULL | template/tmpl/1/1/170c3425-6078-479e-8843-3f8fdc02aaf3.vhd | http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06-12-master -xen.vhd.bz2 | Ready | 0 | 0 |0 | 0 | NULL| | 6 |1 | 5 | 2013-08-29 18:16:53 | 2013-08-29 21:01:19 | NULL | 100 | 52428800 | Image | 50430464 | DOWNLOADED | NULL | NULL | template/tmpl/1/5/0911ae22-350f-38f5-ad3e-423e7e5d6539.vhd | http://people.apache.org/~bhaisaab/vms/ttylinux_pv.vhd | Ready | 0 | 0 |0 | 0 | NULL| ++--+-+-+- ++--+++---+--- -+---++--- -+ +---+---+-+--- ---+-+-+ 2 rows in set (0.00 sec) From: Chiradeep Vittal [chiradeep.vit...@citrix.com] Sent: Friday, August 30, 2013 9:22 AM To: dev@cloudstack.apache.org Subject: Re: Template Problem Just the contents of template_store_ref On 8/30/13 9:17 AM, Soheil Eizadi seiz...@infoblox.com wrote: Chiradeep, What do you want me to lookup, before I clean up my system? -Soheil From: Wei ZHOU [ustcweiz...@gmail.com] Sent: Thursday, August 29, 2013 3:16 PM To: dev@cloudstack.apache.org Subject: Re: Template Problem Chiradeep, Thanks. I cannot give more information as the environment has been cleaned. Maybe Soheil can. I think it is the same issue. 2013/8/29 Chiradeep Vittal chiradeep.vit...@citrix.com It should be template_store_ref On 8/29/13 2:34 PM, Wei ZHOU ustcweiz...@gmail.com wrote: It is the same issue to what I described in comment of CLOUDSTACK-4459: Wei Zhouhttps://issues.apache.org/jira/secure/ViewProfile.jspa?name=weizho u a dded a comment - 29/Aug/13 22:24 Edison, after I redeploy the environment (clean db, restore devcloud to original, reploydb), the system is up. Th systemvms (ssvm and cpvm) are running and the state are up, but I can not see tinylinux in template list. the template_host_ref table is empty. Wei Zhouhttps://issues.apache.org/jira/secure/ViewProfile.jspa?name=weizho u a dded a comment - 29/Aug/13 23:10 It works well on 4.2-forward branch. 2013/8/29 Soheil Eizadi seiz...@infoblox.com I migrated my development environment to the top of trunk which points to 4.3 now. I created a new CloudStack runtime environment running XenServer 6.2, for my Secondary Store I replicated another NFS Server that has images for preRC CloudStack version 4.2 SystemVM and TinyLinux on it. I have the CS running and it creates the SSVM and it is healthy, but I don't see any Templates in the UI, from the database I would expect to see the tiny Linux and SystemVM Template in the Templates view. I have restarted the management server and the problem is persistent. What is the next steps to debug this further? -Soheil Here is some logs below: From SSVM: /usr/local/cloud/systemvm/ssvm-check.sh First DNS server is 10.48.15.25 PING 10.48.15.25 (10.48.15.25): 56 data bytes 64 bytes from 10.48.15.25:
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Congrats Mr Musayev! On Wed, Sep 4, 2013 at 11:51 PM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: [DISCUSS] Remove async APIs from storage framework
+1 for removing complexity especially if the sync pattern is being used on top of the async pattern. I see this behavior in the AgentManager.send as well -- even thought the AM is capable of async, practically nobody uses it as such. But I guess the question will arise : what if I do want more than 10^n long-running storage jobs (cos my cloud is as successful as AWS :)) On 9/4/13 5:03 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: I've been reading over the storage code and have come to the conclusion that the async aspects of the storage framework should be removed. Whenever one introduces an async pattern you have to give a lot of consideration to its use, benefits, and impact. Within the context of ACS and given the current state of its code, I do not think it will be possible to realize any benefits of the current callback approach. Since nothing else in ACS uses callbacks, all of the async methods are essentially wrapped in synchronous calls. So nothing as it stands is actually async. Besides the current implementation, you need to conciser how you would expect an implementation of the storage framework to use the callback. The problem with callbacks is that they assume some in memory state. This means if the process/server crashes that state is lost. Many will say just serialize the callback to the DB, but that is very impractical. Since ACS doesn't actually stand in the data path, an async pattern won't really even allow it to have better performance. ACS is just waiting for some storage operation to happen. ACS can easily spawn 1000 threads and have them all wait. If you were to get to this point, you'd find that downstream you'll most likely have issues as you have 1000 create template operations so its killing your filer. So you will throttle storage operations to a level that won't kill your infrastructure and that level is no where near the scalability limits of threads. The callbacks pattern really complicates the code and I see no real benefit. Instead of spending a lot of effort trying to make all of ACS async to make it beneficial, I'd say that effort should be spent on making ACS idempotent and crash-only. The point being, there's more beneficial things we can do with our time. Given that only solidfire implements the new framework (and ACS legacy too), I would assume its a simple things for Edison to just go and quickly change it non-async. Darren
RE: Template Problem
Hi Min, Your timing is good, today I worked on this setup some more and figured out what is going on with it. I made some changes to it and so it is not exactly what I documented before but the problem is still there. I removed the Secondary Storage from my system and installed a new secondary storage with just the SystemVM template on it, to see if that would fix the problem, but I still had the problem :( If there is a Wiki page on this area that would be helpful, I am slowly coming up to speed looking at the code and becoming more familiar it. The GUI calls list templates with the templatefilter set to all. There is a bunch of logic searching the template based on lots of knobs but mostly it boils down to displaying what is in template_view which before I changed the secondary storage and now with the new secondary storage, only has the ISO in it and that is why it not showing any templates in the GUI. (mysql dump below.) Since I deleted the secondary storage, TemplateServiceImpl detected and reinstalled the tinyLinux image in it, and it is in the Ready state in template_store_ref (mysql dump below.) There was a change to template_view where it only displays items if vm_template.state='Active', the state of the tinyLinux is 'Allocated'. (mysql dump below.) I updated it manually to Active: mysql update vm_template set state='Active' where id=5; This fixed the template_view display to show the tinyLinux in template_view and fixed the GUI. Now the question is why is it not set to Active? May be it was never set to Active as the clause to display based on vm_template.state='Active'; came into the system from schema-420to430.sql. Hope this helps, -Soheil mysql select * from vm_template; +-+-+---+--++--+-+-+--+--++-+-++--+---+-+---+-+--+-+-+-+-++--+--+---+---+--+-+--+ | id | unique_name | name | uuid | public | featured | type| hvm | bits | url | format | created | removed | account_id | checksum | display_text | enable_password | enable_sshkey | guest_os_id | bootable | prepopulate | cross_zones | extractable | hypervisor_type | source_template_id | template_tag | sort_key | size | state | update_count | updated | dynamically_scalable | +-+-+---+--++--+-+-+--+--++-+-++--+---+-+---+-+--+-+-+-+-++--+--+---+---+--+-+--+ | 1 | routing-1 | SystemVM Template (XenServer) | db876372-11c1-11e3-b37b-005056aec95c | 0 |0 | SYSTEM | 0 | 32 | http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06-12-master-xen.vhd.bz2 | VHD| 2013-08-30 15:16:46 | NULL| 1 | fb1b6e032a160d86f2c28feb5add6d83 | SystemVM Template (XenServer) | 0 | 0 | 183 |1 | 0 | 1 | 0 | XenServer | NULL | NULL |0 | 276162048 | Allocated |0 | NULL|0 | | 2 | centos53-x86_64 | CentOS 5.3(64-bit) no GUI (XenServer) | db876c46-11c1-11e3-b37b-005056aec95c | 1 |1 | BUILTIN | 0 | 64 | http://download.cloud.com/templates/builtin/f59f18fb-ae94-4f97-afd2-f84755767aca.vhd.bz2 | VHD| 2013-08-30 15:16:46 | 2013-08-30 15:16:46 | 1 | b63d854a9560c013142567bbae8d98cf | CentOS 5.3(64-bit) no GUI (XenServer) | 0 | 0 | 12 |1 | 0 | 1 | 1 | XenServer | NULL | NULL |0 | NULL | Allocated |0 | NULL|0 | | 3 | routing-3 | SystemVM Template (KVM) | db87748e-11c1-11e3-b37b-005056aec95c | 0 |0 | SYSTEM | 0 | 64
RE: Questions about 4.2 RC
Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? I suggested this and got no answers, meaning +1 on this one. Animesh Yes that is the plan
RE: Questions about 4.2 RC
-Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 1:22 AM To: dev@cloudstack.apache.org; Animesh Chaturvedi Subject: Questions about 4.2 RC Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. The command is only available in non-oss build, but is still listed in command.properties, as there is no separate file for commands non-oss build. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. It's by design, the vm snapshot is listed under vm-vm snapshot view, while the volume snapshot listed under storage-snapshot view. -Sebastien
RE: Could not find corresponding resource manager for VMware
Ilya,
I was testing out the cloudsand version, and I noticed the following error:
When creating an instance from a template, the instance ends with an error
state Error Core:530 Unable to start instance 'name' {GUID}, see mgmt server
for details. This error is also logged in the mgmt servers log file. I also
noticed on the hypervisor side {vmware} that the vm gets created without a
disk. The system then tries to boot the vm and of course it errors out. So it
seems the root cause is when the system creates the instance from the template
it is not creating the vmdk disk. I have tried this with multiple templates,
and even the default centos 5.3 template. If I deploy the template direct
using vCenter it works. Thoughts?
-Michael
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Wed, 4 Sep 2013 16:12:21 +
Michael,
Yes, if I see features that would be beneficial now rather than later when
4.3 comes out. What will happen then is, I take the feature from 4.3 (master)
and backport it to CloudSand 4.2, test it, and if goes well, I push it to
github, release rpms and virt appliance. Think of it as hybrid beta release
of ACS.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 11:50 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Sweet, I didn't even know this existed, I will check it out. So once ACS 4.2
comes out, do you still plan on adding onto cloudsand?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Tue, 3 Sep 2013 22:19:35 +
Michael,
ACS does support vSphere 5.1 as I run it now.
Customized edition of CloudStack known as CloudSand, has extra VMware
support built in (primary dvswitch support and ACS hostnames - displayed in
VC). Its being ran in production by several companies. CloudSand edition can
be upgraded to official ACS 4.2 when it comes out.
Check out the code from http://www.github.com/serverchief/cloudsand
Homepage on www.cloudsand.com
I will try to release CloudSand edition of ACS as virtual appliance this
week.
Since you've squared away build process, you can pull the source code and
build your own rpms.
NOTE: you need to enable dvswitch support and vm.name support in
settings.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 1:32 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
So adding the -p nonoss to ./package.sh fixed the issue. Do I need
to do something to request the documentation update, or will someone
from the dev team take care of that?
Also does CS 4.1 support vsphere 5.1? I've been doing some reading
at it seems like you can use the vim25.jar from the 5.1 SDK kit, but
you have to rename it to vim25_51.jar, however I think renaming the
jar file to vim25_51 would kill the install-nonoss script because it
looks for the file to be named vim25.jar.
From: mphilli7...@hotmail.com
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Mon, 2 Sep 2013 22:32:45 -0500
Let me try that, I will report back my findings.
How do I file a report bug if it fixes the issue?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Tue, 3 Sep 2013 02:46:35 +
I think its documented (but I can't recall for certain). You
need to use -p
nonoss to package nonoss packages.
Also you don't need to do mvn install -Dnonoss. Its good to do
it initially
to confirm all builds well, but going forward the packaging process
will take care of It.
Regards
Ilya
- All mistakes in this message are not mine but Android's.
Original message
From: Michael Phillips mphilli7...@hotmail.com
Date: 09/02/2013 10:33 PM (GMT-05:00)
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
I did not..
I added the jars to cloudstack using the mvn install -Dnonoss
command
then packaged it just using package.sh
Guessing I was supposed to add the -p nonoss to that? Is that
documented anywhere?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Tue, 3 Sep 2013
RE: Could not find corresponding resource manager for VMware
So not an Appliance issue, but general cloudstack issue if I read this right.
Can you paste the management-server.log in paste.cloudstack.org and respond
with a link?
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Thursday, September 05, 2013 5:18 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Ilya,
I was testing out the cloudsand version, and I noticed the following error:
When creating an instance from a template, the instance ends with an error
state Error Core:530 Unable to start instance 'name' {GUID}, see mgmt
server for details. This error is also logged in the mgmt servers log file. I
also
noticed on the hypervisor side {vmware} that the vm gets created without a
disk. The system then tries to boot the vm and of course it errors out. So it
seems the root cause is when the system creates the instance from the
template it is not creating the vmdk disk. I have tried this with multiple
templates, and even the default centos 5.3 template. If I deploy the
template direct using vCenter it works. Thoughts?
-Michael
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Wed, 4 Sep 2013 16:12:21 +
Michael,
Yes, if I see features that would be beneficial now rather than later when
4.3 comes out. What will happen then is, I take the feature from 4.3 (master)
and backport it to CloudSand 4.2, test it, and if goes well, I push it to
github,
release rpms and virt appliance. Think of it as hybrid beta release of ACS.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 11:50 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Sweet, I didn't even know this existed, I will check it out. So once
ACS 4.2 comes out, do you still plan on adding onto cloudsand?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Tue, 3 Sep 2013 22:19:35 +
Michael,
ACS does support vSphere 5.1 as I run it now.
Customized edition of CloudStack known as CloudSand, has extra
VMware
support built in (primary dvswitch support and ACS hostnames -
displayed in VC). Its being ran in production by several companies.
CloudSand edition can be upgraded to official ACS 4.2 when it comes out.
Check out the code from
http://www.github.com/serverchief/cloudsand
Homepage on www.cloudsand.com
I will try to release CloudSand edition of ACS as virtual appliance this
week.
Since you've squared away build process, you can pull the source
code and build your own rpms.
NOTE: you need to enable dvswitch support and vm.name support in
settings.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 1:32 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
So adding the -p nonoss to ./package.sh fixed the issue. Do I
need to do something to request the documentation update, or
will someone from the dev team take care of that?
Also does CS 4.1 support vsphere 5.1? I've been doing some
reading at it seems like you can use the vim25.jar from the 5.1
SDK kit, but you have to rename it to vim25_51.jar, however I
think renaming the jar file to vim25_51 would kill the
install-nonoss script because it looks for the file to be named
vim25.jar.
From: mphilli7...@hotmail.com
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Mon, 2 Sep 2013 22:32:45 -0500
Let me try that, I will report back my findings.
How do I file a report bug if it fixes the issue?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager
for VMware
Date: Tue, 3 Sep 2013 02:46:35 +
I think its documented (but I can't recall for certain). You
need to use -p
nonoss to package nonoss packages.
Also you don't need to do mvn install -Dnonoss. Its good
to do it initially
to confirm all builds well, but going forward the packaging
process will take care of It.
Regards
Ilya
- All mistakes in this message are not mine but Android's.
Original message
From: Michael Phillips mphilli7...@hotmail.com
Date: 09/02/2013 10:33 PM (GMT-05:00)
To: dev@cloudstack.apache.org
RE: [ANNOUNCE] New PMC member: Ilya Musayev
I'll cut the speech short, but I do evangelize it in NYC, we will add a new big user soon, cant disclose the name just yet, but its well known :) -Original Message- From: Daan Hoogland [mailto:daan.hoogl...@gmail.com] Sent: Thursday, September 05, 2013 6:54 AM To: dev Subject: Re: [ANNOUNCE] New PMC member: Ilya Musayev At it Ilya, how about a 'speech' on your take on the future of cloudstack? congrats, Daan On Thu, Sep 5, 2013 at 11:15 AM, Saksham Srivastava saksham.srivast...@citrix.com wrote: Congratulations Ilya. Regards, Saksham -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 12:22 PM To: dev@cloudstack.apache.org Subject: [ANNOUNCE] New PMC member: Ilya Musayev The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: [DISCUSS] OOM killer and Routing/System VM's = :(
Maintaining a custom kernel is a big hassle, even if it is a few lines of code change. Can we do something in userspace? What about the software watchdog that is available? Along the lines of: http://goo.gl/oO3Lzr http://linux.die.net/man/8/watchdog On 9/5/13 7:13 AM, Funs Kessen fkes...@schubergphilis.com wrote: Well, you can't as far as I've looked in the source of panic.c. So I'm thinking of investigating of adding -1 as an option and seeing if I can push halt in, let's hope the guys that do kernel stuff find this useful too. So it seems the patch, I conjured up for panic.c, is seen as not so useful, there is however another way to achieve the same result. This would mean that we load a crash kernel with our own .sh script as init to do our bidding. Would that be a plan ? Cheers, Funs Sent from my iPhone On 4 sep. 2013, at 23:35, Marcus Sorensen shadow...@gmail.com wrote: What would work as a quick fix for this sort of situation would be if the machine could be configured to power off rather than rebooting on oom. Then the HA system would restart the VM, applying all configs. Anyone know how to do that? :-) On Wed, Sep 4, 2013 at 1:14 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: On 09/04/2013 11:37 AM, Roeland Kuipers wrote: Hi Darren, Thanks for your reply! Could you share a bit more on your plans/ideas? We also have been braining on other approaches of managing the systemvm's, especially small customizations for specific tenants. And maybe even leveraging a config mgmt tools like chef or puppet with the ability to integrate CS with that in some way. I'll have to send the full details later but here's a rough idea. The basic approach is this. Logical changes to the VRs (or system vms in general) get mapped to configuration items. So add a LB rule maps to iptables config and haproxy config. When you change a LB rule we then bump up the requested version of the configuration for iptables/haproxy. So the requested version will be 4 maybe. The applied version will be 3 as the VR still has the old configuration. Since 4 != 3, the VR will be signaled to pull the latest iptables/haproxy config. So it will pull the configuration. Say in the mean time somebody else adds four other LB rules. So the requested version is now at 8. So when the VR pulls the config it will get version 8, and then reply back saying it applied version 8. The applied version is now 8 which is greater than 4 (the version the first LB rule change was waiting for) so basically all async jobs waiting for the LB change will be done. To pull the configuration from the VR, the VR will be hitting a templating configuration system. So it pulls the full iptables and haproxy config. Not incremental changes. So if the VR ever reboots itself, it can easily just pull the latest config of everything and apply it. So it will be consistent. I'd be interested to hear what type of customizations you would like to add. It will definitely be an extensible system, but the problem is if your extensions wants to touch the same configuration files that ACS wants to manage. That gets a bit tricky as its really easy for each to break each other. But I can definitely add some hooks that users can use to mess up things and void the warranty. I've thought about chef and puppet for this, but basically it comes down to two things. I'm really interested in this being fast and light weight. Ruby is neither of those. So the core ACS stuff will probably remain as very simple shell scripts. Simple in that they really just need to download configuration and restart services. They know nothing about the nature of the changes. If, as an extension, you want to do something with puppet, chef, I'd be open to that. That's your deal. This approach has many other benefits. Like, for example, we can ensure that as we deploy a new ACS release existing system VMs can be updated (without a reboot, unless the kernel changes). Additionally, its fast and updates happen in near constant time. So most changes will be just a couple of seconds, even if you have 4000 LB rules. Darren
create ubuntu 12.04 template from snapshot,then create new vm from this template and the new vm failed to start!!!
Hi, I use cloudstack 4.1.1 and xenserver 6.1. The issue scenario as following: 1.set up concurring snapshots for a Ubuntu 12.04 VM (hourly snapshot for testing) 2. Create a template from generated snapshot 3. create a new instance from the template 4. start the vm and getting blackscreen, the vm failed to boot. 5. if I changed following parameters, I can start the vm successfully. --xe vm-param-clear uuid= param-name=HVM-boot-policy (it value was BIOS order) --xe vm-param-clear uuid= param-name= HVM-boot-params=(it value was order: dc) -- xe vm-param-set uuid= PV-bootloader=pygrub (it value was null) 6. But after stop/start it again, the vm will lost those changes. You will have to manually change it again. My questions is: Do we have any way to let the vm keep those changed parameters in step5? I tried to find some place I can changed in DB but get nothing. Great Thanks, William This e-mail may be privileged and/or confidential, and the sender does not waive any related rights and obligations. Any distribution, use or copying of this e-mail or the information it contains by other than an intended recipient is unauthorized. If you received this e-mail in error, please advise me (by return e-mail or otherwise) immediately. Ce courrier électronique est confidentiel et protégé. L'expéditeur ne renonce pas aux droits et obligations qui s'y rapportent. Toute diffusion, utilisation ou copie de ce message ou des renseignements qu'il contient par une personne autre que le (les) destinataire(s) désigné(s) est interdite. Si vous recevez ce courrier électronique par erreur, veuillez m'en aviser immédiatement, par retour de courrier électronique ou par un autre moyen.
RE: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch
Tuna, Instead of SSH, have you considered launching a console and logging in as root? Password for root is 6m1ll10n In the future, when posting errors, try including 20 -30 lines above the actual exception as there maybe (and usually are) the pointers to the issue. Regards ilya -Original Message- From: Nguyen Anh Tu [mailto:ng.t...@gmail.com] Sent: Thursday, September 05, 2013 11:13 AM To: dev@cloudstack.apache.org Cc: Hugo Trippaers Subject: Re: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch Ah. Finally error happens: 2013-09-05 22:10:05,748 WARN [storage.secondary.SecondaryStorageManagerImpl] (secstorage-1:null) Exception while trying to start secondary storage vm com.cloud.exception.AgentUnavailableException: Resource [Host:1] is unreachable: Host 1: Unable to start instance due to Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineMa nagerImpl.java:946) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerIm pl.java:557) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerIm pl.java:550) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.startSecStorag eVm(SecondaryStorageManagerImpl.java:265) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.allocCapacity( SecondaryStorageManagerImpl.java:690) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.expandPool(S econdaryStorageManagerImpl.java:1297) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(Pre miumSecondaryStorageManagerImpl.java:123) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(Pre miumSecondaryStorageManagerImpl.java:50) at com.cloud.vm.SystemVmLoadScanner.loadScan(SystemVmLoadScanner.java :104) at com.cloud.vm.SystemVmLoadScanner.access$100(SystemVmLoadScanner.ja va:33) at com.cloud.vm.SystemVmLoadScanner$1.reallyRun(SystemVmLoadScanner.j ava:81) at com.cloud.vm.SystemVmLoadScanner$1.run(SystemVmLoadScanner.java:72 ) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) at java.util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:35 1) at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.ac cess$301(ScheduledThreadPoolExecutor.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.ru n(ScheduledThreadPoolExecutor.java:293) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.jav a:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.ja va:615) at java.lang.Thread.run(Thread.java:722) Caused by: com.cloud.utils.exception.ExecutionException: Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineMa nagerImpl.java:913) ... 20 more Two system vms are destroyed and created new. 2013/9/5 Nguyen Anh Tu ng.t...@gmail.com Hi guys, I'm using an unofficial version ACS 4.1 (sdnextensions branch). I build my own cloud-agent package and install successfully into a kvm host and add this host success with two NICs: cloudbr0 (Management, Storage) and cloudbr1 (Guest, Public). I use openvswitch, follow Hugo's guide here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/KVM+with+Ope nVS witch SSVM and CPVM start normally but I can not ping or ssh into via local-link (169.254.0.0/16) although both of them are already allocated IP addresses. On MS, local-link nic is cloud0 (169.254.0.1). No error log happen. In DB, they are consistently in starting state, not running. Anyone can help? -- N.g.U.y.e.N.A.n.H.t.U -- N.g.U.y.e.N.A.n.H.t.U
RE: Could not find corresponding resource manager for VMware
Pretty sure it's not an appliance issue. BTW, I am not using the appliance
version of cloudsand, I built from source which was on github. Looks like
paste.cloudstack.org might be down...
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Thu, 5 Sep 2013 21:32:40 +
So not an Appliance issue, but general cloudstack issue if I read this right.
Can you paste the management-server.log in paste.cloudstack.org and respond
with a link?
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Thursday, September 05, 2013 5:18 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Ilya,
I was testing out the cloudsand version, and I noticed the following
error:
When creating an instance from a template, the instance ends with an error
state Error Core:530 Unable to start instance 'name' {GUID}, see mgmt
server for details. This error is also logged in the mgmt servers log file.
I also
noticed on the hypervisor side {vmware} that the vm gets created without a
disk. The system then tries to boot the vm and of course it errors out. So
it
seems the root cause is when the system creates the instance from the
template it is not creating the vmdk disk. I have tried this with multiple
templates, and even the default centos 5.3 template. If I deploy the
template direct using vCenter it works. Thoughts?
-Michael
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Wed, 4 Sep 2013 16:12:21 +
Michael,
Yes, if I see features that would be beneficial now rather than later
when
4.3 comes out. What will happen then is, I take the feature from 4.3
(master)
and backport it to CloudSand 4.2, test it, and if goes well, I push it to
github,
release rpms and virt appliance. Think of it as hybrid beta release of ACS.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 11:50 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Sweet, I didn't even know this existed, I will check it out. So once
ACS 4.2 comes out, do you still plan on adding onto cloudsand?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Tue, 3 Sep 2013 22:19:35 +
Michael,
ACS does support vSphere 5.1 as I run it now.
Customized edition of CloudStack known as CloudSand, has extra
VMware
support built in (primary dvswitch support and ACS hostnames -
displayed in VC). Its being ran in production by several companies.
CloudSand edition can be upgraded to official ACS 4.2 when it comes out.
Check out the code from
http://www.github.com/serverchief/cloudsand
Homepage on www.cloudsand.com
I will try to release CloudSand edition of ACS as virtual appliance
this
week.
Since you've squared away build process, you can pull the source
code and build your own rpms.
NOTE: you need to enable dvswitch support and vm.name support in
settings.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 1:32 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
So adding the -p nonoss to ./package.sh fixed the issue. Do I
need to do something to request the documentation update, or
will someone from the dev team take care of that?
Also does CS 4.1 support vsphere 5.1? I've been doing some
reading at it seems like you can use the vim25.jar from the 5.1
SDK kit, but you have to rename it to vim25_51.jar, however I
think renaming the jar file to vim25_51 would kill the
install-nonoss script because it looks for the file to be named
vim25.jar.
From: mphilli7...@hotmail.com
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Mon, 2 Sep 2013 22:32:45 -0500
Let me try that, I will report back my findings.
How do I file a report bug if it fixes the issue?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager
for VMware
Date: Tue, 3 Sep 2013 02:46:35 +
I think its documented (but I can't recall for certain). You
need to use -p
nonoss to package nonoss packages.
Re: Template Problem
Thanks Soheil. From your description, I just realized that you are working on master branch. You may have mentioned in your earlier email, but I have missed that. In master branch, we reused state column of db table vm_template to indicate whether a template is active or not active, but we missed one place to update this column value for existing template, thus your problem. I have made a commit 072013911703c906ba6af0374d25afb494e74737 to address this issue, verified on my local and worked fine for fresh install env. Can you try again with latest code? -min On 9/5/13 11:43 AM, Soheil Eizadi seiz...@infoblox.com wrote: Hi Min, Your timing is good, today I worked on this setup some more and figured out what is going on with it. I made some changes to it and so it is not exactly what I documented before but the problem is still there. I removed the Secondary Storage from my system and installed a new secondary storage with just the SystemVM template on it, to see if that would fix the problem, but I still had the problem :( If there is a Wiki page on this area that would be helpful, I am slowly coming up to speed looking at the code and becoming more familiar it. The GUI calls list templates with the templatefilter set to all. There is a bunch of logic searching the template based on lots of knobs but mostly it boils down to displaying what is in template_view which before I changed the secondary storage and now with the new secondary storage, only has the ISO in it and that is why it not showing any templates in the GUI. (mysql dump below.) Since I deleted the secondary storage, TemplateServiceImpl detected and reinstalled the tinyLinux image in it, and it is in the Ready state in template_store_ref (mysql dump below.) There was a change to template_view where it only displays items if vm_template.state='Active', the state of the tinyLinux is 'Allocated'. (mysql dump below.) I updated it manually to Active: mysql update vm_template set state='Active' where id=5; This fixed the template_view display to show the tinyLinux in template_view and fixed the GUI. Now the question is why is it not set to Active? May be it was never set to Active as the clause to display based on vm_template.state='Active'; came into the system from schema-420to430.sql. Hope this helps, -Soheil mysql select * from vm_template; +-+-+---+- -++--+-+-+--+- -- ---++-+-+- ---+--+--- +-+---+-+--+-- ---+-+-+-+ +--+--+---+---+--+ -+--+ | id | unique_name | name | uuid | public | featured | type| hvm | bits | url | format | created | removed | account_id | checksum | display_text | enable_password | enable_sshkey | guest_os_id | bootable | prepopulate | cross_zones | extractable | hypervisor_type | source_template_id | template_tag | sort_key | size | state | update_count | updated | dynamically_scalable | +-+-+---+- -++--+-+-+--+- -- ---++-+-+- ---+--+--- +-+---+-+--+-- ---+-+-+-+ +--+--+---+---+--+ -+--+ | 1 | routing-1 | SystemVM Template (XenServer) | db876372-11c1-11e3-b37b-005056aec95c | 0 |0 | SYSTEM | 0 | 32 | http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06-12-master -xen.vhd.bz2 | VHD| 2013-08-30 15:16:46 | NULL| 1 | fb1b6e032a160d86f2c28feb5add6d83 | SystemVM Template (XenServer) | 0 | 0 | 183 |1 | 0 | 1 | 0 | XenServer | NULL | NULL |0 | 276162048 | Allocated |0 | NULL |0 | | 2 | centos53-x86_64 | CentOS 5.3(64-bit) no GUI (XenServer) | db876c46-11c1-11e3-b37b-005056aec95c | 1 |1 | BUILTIN | 0 | 64 | http://download.cloud.com/templates/builtin/f59f18fb-ae94-4f97-afd2-f84755 767aca.vhd.bz2
Re: [DISCUSS] Remove async APIs from storage framework
Kelvin, I really have some concerns about the entire async framework that was put into place as part of the ipc framework. I don't know what VM sync is so I can't really comment too much on that. Basically my concern with the async framework is that it's yet another custom ACS thing. The use of cglib in the callbacks (or futures?) is to obscure and is not clear to a new person coming into ACS (for example, me). If we wish to do async stuff it should be based off of java 5 futures or guava's ListenableFutures. Those APIs and libraries are far more natural to java devs in the wild. Or we can get inspiration from scala and akka on how to do async well in java. In short, if we have a real use case for async, then I'd prefer we use an off the shelf library. In general I will lean towards most code being synchronous. To answer Chiradeeps comment on how do we do insane scale. That just requires true horizontal scalability of the management stack (which I'm not too sure if that's been fully proven beyond 3 servers and 30k hypervisors, but that's something we can address) and a really big database and somebody who enjoys really large failure domains. Darren On Sep 5, 2013, at 2:51 PM, Kelven Yang kelven.y...@citrix.com wrote: New VM sync and job flow handling are moving towards event-driven model which is async natural. Not sure about the details on how storage framework is using async mechanism, but argument on callback to assume memory state does not stand as strong as it might be, since arguments in any regular call pass information in memory the same way. Excessive callbacks (async method) usage pattern does encourage people to program in a wrong direction of async programming, I would prefer to substitute callback with event signaling and maintain/access flow-context through separated facility with well defined API, this can prevent developers to pass too much information directly and create a tight-coupling pattern. However, I'm a little concerning about yet another storage refactoring right after its previous refactoring work is barely done. Kelven On 9/5/13 12:57 PM, Chiradeep Vittal chiradeep.vit...@citrix.com wrote: +1 for removing complexity especially if the sync pattern is being used on top of the async pattern. I see this behavior in the AgentManager.send as well -- even thought the AM is capable of async, practically nobody uses it as such. But I guess the question will arise : what if I do want more than 10^n long-running storage jobs (cos my cloud is as successful as AWS :)) On 9/4/13 5:03 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: I've been reading over the storage code and have come to the conclusion that the async aspects of the storage framework should be removed. Whenever one introduces an async pattern you have to give a lot of consideration to its use, benefits, and impact. Within the context of ACS and given the current state of its code, I do not think it will be possible to realize any benefits of the current callback approach. Since nothing else in ACS uses callbacks, all of the async methods are essentially wrapped in synchronous calls. So nothing as it stands is actually async. Besides the current implementation, you need to conciser how you would expect an implementation of the storage framework to use the callback. The problem with callbacks is that they assume some in memory state. This means if the process/server crashes that state is lost. Many will say just serialize the callback to the DB, but that is very impractical. Since ACS doesn't actually stand in the data path, an async pattern won't really even allow it to have better performance. ACS is just waiting for some storage operation to happen. ACS can easily spawn 1000 threads and have them all wait. If you were to get to this point, you'd find that downstream you'll most likely have issues as you have 1000 create template operations so its killing your filer. So you will throttle storage operations to a level that won't kill your infrastructure and that level is no where near the scalability limits of threads. The callbacks pattern really complicates the code and I see no real benefit. Instead of spending a lot of effort trying to make all of ACS async to make it beneficial, I'd say that effort should be spent on making ACS idempotent and crash-only. The point being, there's more beneficial things we can do with our time. Given that only solidfire implements the new framework (and ACS legacy too), I would assume its a simple things for Edison to just go and quickly change it non-async. Darren
RE: Could not find corresponding resource manager for VMware
Maybe try pastebin.com then?
Whats is your vmware.root.disk.controller set to in global settings?
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Thursday, September 05, 2013 5:55 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Pretty sure it's not an appliance issue. BTW, I am not using the appliance
version of cloudsand, I built from source which was on github. Looks like
paste.cloudstack.org might be down...
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for VMware
Date: Thu, 5 Sep 2013 21:32:40 +
So not an Appliance issue, but general cloudstack issue if I read this
right.
Can you paste the management-server.log in paste.cloudstack.org and
respond with a link?
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Thursday, September 05, 2013 5:18 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Ilya,
I was testing out the cloudsand version, and I noticed the following
error:
When creating an instance from a template, the instance ends with an
error state Error Core:530 Unable to start instance 'name' {GUID},
see mgmt server for details. This error is also logged in the mgmt
servers log file. I also noticed on the hypervisor side {vmware}
that the vm gets created without a disk. The system then tries to
boot the vm and of course it errors out. So it seems the root cause
is when the system creates the instance from the template it is not
creating the vmdk disk. I have tried this with multiple templates,
and even the default centos 5.3 template. If I deploy the template
direct using vCenter it works. Thoughts?
-Michael
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Wed, 4 Sep 2013 16:12:21 +
Michael,
Yes, if I see features that would be beneficial now rather than
later when
4.3 comes out. What will happen then is, I take the feature from 4.3
(master) and backport it to CloudSand 4.2, test it, and if goes
well, I push it to github, release rpms and virt appliance. Think of it as
hybrid beta release of ACS.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 11:50 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Sweet, I didn't even know this existed, I will check it out. So
once ACS 4.2 comes out, do you still plan on adding onto cloudsand?
From: imusa...@webmd.net
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager for
VMware
Date: Tue, 3 Sep 2013 22:19:35 +
Michael,
ACS does support vSphere 5.1 as I run it now.
Customized edition of CloudStack known as CloudSand, has extra
VMware
support built in (primary dvswitch support and ACS hostnames -
displayed in VC). Its being ran in production by several companies.
CloudSand edition can be upgraded to official ACS 4.2 when it comes
out.
Check out the code from
http://www.github.com/serverchief/cloudsand
Homepage on www.cloudsand.com
I will try to release CloudSand edition of ACS as virtual
appliance this
week.
Since you've squared away build process, you can pull the source
code and build your own rpms.
NOTE: you need to enable dvswitch support and vm.name support
in
settings.
Regards
ilya
-Original Message-
From: Michael Phillips [mailto:mphilli7...@hotmail.com]
Sent: Tuesday, September 03, 2013 1:32 PM
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager
for VMware
So adding the -p nonoss to ./package.sh fixed the issue.
Do I need to do something to request the documentation
update, or will someone from the dev team take care of that?
Also does CS 4.1 support vsphere 5.1? I've been doing some
reading at it seems like you can use the vim25.jar from the
5.1 SDK kit, but you have to rename it to vim25_51.jar,
however I think renaming the jar file to vim25_51 would kill
the install-nonoss script because it looks for the file to
be named
vim25.jar.
From: mphilli7...@hotmail.com
To: dev@cloudstack.apache.org
Subject: RE: Could not find corresponding resource manager
for VMware
Date: Mon, 2 Sep 2013 22:32:45 -0500
Let me try that, I
Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers (2)
Marcus, Internal load balancing is NOT supported in Basic zone. Therefore, if you create a Basic zone through cloudstack UI, cloudstack UI will NOT do them automatically: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf enable the InternalLbVM provider Jessica -Original Message- From: Jessica Wang [mailto:jessica.w...@citrix.com] Sent: Thursday, September 05, 2013 2:49 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi; Brian Federle Subject: RE: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers Marcus, list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf enable the InternalLbVM provider, If you create an Advanced zone through cloudStack UI, you don't need to do anything above. cloudstack UI will do them automatically. To create an Advanced zone through cloudStack UI: Infrastructure menu zones click Add zone button Add zone wizard pops up click Advanced Jessica -Original Message- From: Marcus Sorensen [mailto:shadow...@gmail.com] Sent: Thursday, September 05, 2013 2:15 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi Subject: Re: Questions about 4.2 RC More questions... I've been going down the rabbithole in figuring out how to deploy VPCs. Part of this might be a missing marvin config in my zone deploy, I still have no idea yet, but after reviewing https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-vpc-tiers.html, I found that I have to do this: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf Now I can go into the UI and enable the InternalLbVM provider, and finally create a VPC. Is this how we expect users to do it, or is there something I'm missing in the UI. On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote: -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 1:22 AM To: dev@cloudstack.apache.org; Animesh Chaturvedi Subject: Questions about 4.2 RC Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. The command is only available in non-oss build, but is still listed in command.properties, as there is no separate file for commands non-oss build. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. It's by design, the vm snapshot is listed under vm-vm snapshot view, while the volume snapshot listed under storage-snapshot view. -Sebastien
Re: HA redundant virtual router
Here is the doc. https://cwiki.apache.org/confluence/display/CLOUDSTACK/Redundant+Virtual+Router+Functional+Spec It's not extremely detail, but describe today's design generally. --Sheng On Thu, Aug 29, 2013 at 8:17 AM, Daan Hoogland daan.hoogl...@gmail.comwrote: ok, let's postpone the discussion till you are at least halve done. We will of course continue to deliberate on what we need internally. Daan On Thu, Aug 29, 2013 at 5:08 PM, Sheng Yang sh...@yasker.org wrote: Hi Daan, As I said, I am writing a design doc to describe the current redundant router policy, to help understanding redundant router. Current it doesn't support VPC, so how to implement it in VPC is still open to discuss. --Sheng On Thu, Aug 29, 2013 at 4:26 AM, Daan Hoogland daan.hoogl...@gmail.com wrote: Sheng, just to make sure; You are going to write this document? I see Roeland understood your mail like this. When you do, I'd like you to keep in mind that we also want redundant routers within a VPC to ensure ACS upgrades are more seamless for customer application groups and - dtap streets. If you need any help on writing such a doc, let me know. kind regards, Daan On Thu, Aug 29, 2013 at 1:13 PM, Roeland Kuipers rkuip...@schubergphilis.com wrote: Hi Sheng, Thanks for the info. Looking forward to the design doc, I trust this will make things clearer. In the meantime will be doing some research and thinking too, to see how we can improve things to also have HA on the RvR in a safe way. We will share this once ready. Thanks, Roeland From: Sheng Yang [mailto:sh...@yasker.org] Sent: donderdag 29 augustus 2013 0:19 To: dev@cloudstack.apache.org Cc: int-cloud; Daan Hoogland Subject: Re: HA redundant virtual router Hi Roeland, I would write a design doc to explain how redundant router works currently. For example, for the point 2, we have to force BACKUP become MASTER because: 1. CS cannot communicate with MASTER at the time 2. CS can communicate with BACKUP. 3. Rule has to be programmed immediately. 4. In case old MASTER come back, it should yield to the VR with updated rule, rather than preempt the updated VR. In this case, CS need to communicate with RvR to program the new rule, thus it need to intervene the RvR to ensure that if there is only one VR got the rule, it should become MASTER. Still, I would write a doc later to try to cover every concern of RvR design. --Sheng On Tue, Aug 27, 2013 at 3:40 AM, Roeland Kuipers rkuip...@schubergphilis.commailto:rkuip...@schubergphilis.com wrote: Hi Sheng, Thanks for your reply. I'll see if we can replay this scenario. With respect to point 1: a good principal IMHO. Point 2: Why do we force a keepalived node to become master and not wait for keepalived to become master? This way there is less reason to intervene and less risk of multiple masters? As we have seen this behavior with RvR without HA in the past. The downside that updates to rules do not function until backup becomes master. But maybe this is wise anyways since there is something wrong. This conflicts a bit with point 2 as we do intervene here. Point 3: In my opinion keepalived is solid enough to leave this responsibility with keepalived and that CS just should check the state and not fiddle with priorities to force masters. Because there is obviously a reason why BACKUP refuses to become master. I think we should let keepalived prevent multiple master as is designed to prevent this. Or do I miss something here? Actually in the scenario you described, with a functioning guest network, keepalived should be able to handle this situation if we make sure all routers have different prios. I still have the opinion HA and RvR are different mechanisms. So what do you think is necessary to have the possibility of HA icw RvR? We have a clear business requirement to have this implement on CS. And we have Developers willing to create these changes to make this possible. We also like to see RvR on VPC's and are also willing to contribute this functionality. Thanks for your feedback! Cheers, Roeland -Original Message- From: Sheng Yang [mailto:sh...@yasker.orgmailto:sh...@yasker.org] Sent: vrijdag 23 augustus 2013 23:25 To: dev@cloudstack.apache.orgmailto:dev@cloudstack.apache.org Subject: Re: HA redundant virtual router Hi Roeland, Thank you for your testing! Power off is not an concern right now, because at that time the VM would disappear anyway. Our concern is more about if VM is still alive but we cannot detect it for a while. For example, a network glitch happened, CS lost connection to the host temporarily(control network), but the guest network is still working. HA
RE: [ANNOUNCE] New PMC member: Ilya Musayev
Thank you all.. -Original Message- From: Ahmad Emneina [mailto:aemne...@gmail.com] Sent: Thursday, September 05, 2013 2:52 PM To: dev@cloudstack.apache.org Subject: Re: [ANNOUNCE] New PMC member: Ilya Musayev Congrats Mr Musayev! On Wed, Sep 4, 2013 at 11:51 PM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Congrats! - Original Message - From: Ilya Musayev imusa...@webmd.net To: dev@cloudstack.apache.org, aemne...@gmail.com Sent: Thursday, September 5, 2013 1:42:43 PM Subject: RE: [ANNOUNCE] New PMC member: Ilya Musayev Thank you all.. -Original Message- From: Ahmad Emneina [mailto:aemne...@gmail.com] Sent: Thursday, September 05, 2013 2:52 PM To: dev@cloudstack.apache.org Subject: Re: [ANNOUNCE] New PMC member: Ilya Musayev Congrats Mr Musayev! On Wed, Sep 4, 2013 at 11:51 PM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
Re: Questions about 4.2 RC
So I guess I need to figure out of that is broken, or if it's marvin,
and if it's marvin, how to fix it.
Now that I've gotten past that, I see the storage refactor has broken
CLVM. I'll file a bug, but it looks like the process is doing
something like:
copy template to primary storage, then create copy of primary storage
template as new volume
This breaks CLVM, because it used to just do:
copy template to primary storage as new volume
Since we can't efficiently clone in CLVM, it expects to always copy
the template from secondary storage, rather than copying to primary
first and then copying the whole template from the primary back to the
same disks. 1) because it thrashes the disks, and 2) copying the
template is usually much faster because the template is sparse, and
the logical volume is not, so copying a 10G template with a real size
of 500M is much faster than copying a 10G logical volume to another
10G logical volume.
in KVMStorageProcessor.java cloneVolumeFromBaseTemplate:
if (primaryPool.getType() == StoragePoolType.CLVM) {
vol = templateToPrimaryDownload(templatePath, primaryPool);
}
This will never work, because templateToPrimaryDownload expects
secondary storage, and we have copied the template to primary storage
and are passing that. e.g.:
{
org.apache.cloudstack.storage.command.CopyCommand: {
destTO: {
org.apache.cloudstack.storage.to.VolumeObjectTO: {
accountId: 2,
dataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
format: QCOW2,
hypervisorType: KVM,
id: 9,
name: ROOT-9,
size: 1073741824,
uuid: d73f3a2b-9e63-4faf-a45b-d6fcf7633793,
vmName: i-2-9-VM,
volumeId: 9,
volumeType: ROOT
}
},
executeInSequence: true,
srcTO: {
org.apache.cloudstack.storage.to.TemplateObjectTO: {
accountId: 2,
checksum: 44cd0e6330a59f031460bc18a40c95a2,
displayText: tiny,
format: QCOW2,
hvm: true,
hypervisorType: KVM,
id: 201,
imageDataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
name: 201-2-a04f958e-0aed-3642-960f-a675a2ee1c44,
origUrl:
http://mirrors.betterservers.com/template/tiny-centos-63.qcow2;,
path: c8da0364-6f94-4c71-9c1d-74078e55bbb8,
uuid: 7dcdb1fb-e7e3-4de0-bf93-13d3e6c4ade5
}
},
wait: 0
}
}
Also, format should be 'RAW', I believe, not 'QCOW2'.
On Thu, Sep 5, 2013 at 3:40 PM, Animesh Chaturvedi
animesh.chaturv...@citrix.com wrote:
-Original Message-
From: Marcus Sorensen [mailto:shadow...@gmail.com]
Sent: Thursday, September 05, 2013 2:15 PM
To: dev@cloudstack.apache.org
Cc: Animesh Chaturvedi
Subject: Re: Questions about 4.2 RC
More questions... I've been going down the rabbithole in figuring out
how to deploy VPCs. Part of this might be a missing marvin config in my
zone deploy, I still have no idea yet, but after reviewing
https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-vpc-
tiers.html,
I found that I have to do this:
list internalloadbalancerelements
configure internalloadbalancerelement enabled=true
id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf
Now I can go into the UI and enable the InternalLbVM provider, and
finally create a VPC. Is this how we expect users to do it, or is there
something I'm missing in the UI.
[Animesh] Marcus from the spec The cloudStack UI will enable the
element/provider automatically as a part of physical network creation.
On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote:
-Original Message-
From: sebgoa [mailto:run...@gmail.com]
Sent: Thursday, September 05, 2013 1:22 AM
To: dev@cloudstack.apache.org; Animesh Chaturvedi
Subject: Questions about 4.2 RC
Hi,
I followed the testing procedure for the 4.2 RC, I am ready to vote
+1 but I have questions (hence the new thread, don't want to hijack
the VOTE).
1. the CHANGES file points to the release notes, but where are the
release notes ? I know that no-one really stepped
Re: [ANNOUNCE] New PMC member: Ilya Musayev
Well done Ilya! :-) Thank you for all your help. On 5 September 2013 21:42, Musayev, Ilya imusa...@webmd.net wrote: Thank you all.. -Original Message- From: Ahmad Emneina [mailto:aemne...@gmail.com] Sent: Thursday, September 05, 2013 2:52 PM To: dev@cloudstack.apache.org Subject: Re: [ANNOUNCE] New PMC member: Ilya Musayev Congrats Mr Musayev! On Wed, Sep 4, 2013 at 11:51 PM, sebgoa run...@gmail.com wrote: The Project Management Committee (PMC) for Apache CloudStack has asked Ilya Musayev to join the PMC and we are pleased to announce that they have accepted. Join me in congratulating Ilya, -The CloudStack PMC
RE: Questions about 4.2 RC
-Original Message- From: Marcus Sorensen [mailto:shadow...@gmail.com] Sent: Thursday, September 05, 2013 2:15 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi Subject: Re: Questions about 4.2 RC More questions... I've been going down the rabbithole in figuring out how to deploy VPCs. Part of this might be a missing marvin config in my zone deploy, I still have no idea yet, but after reviewing https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-vpc- tiers.html, I found that I have to do this: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf Now I can go into the UI and enable the InternalLbVM provider, and finally create a VPC. Is this how we expect users to do it, or is there something I'm missing in the UI. [Animesh] Marcus from the spec The cloudStack UI will enable the element/provider automatically as a part of physical network creation. On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote: -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 1:22 AM To: dev@cloudstack.apache.org; Animesh Chaturvedi Subject: Questions about 4.2 RC Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. The command is only available in non-oss build, but is still listed in command.properties, as there is no separate file for commands non-oss build. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. It's by design, the vm snapshot is listed under vm-vm snapshot view, while the volume snapshot listed under storage-snapshot view. -Sebastien
RE: Template Problem
Great, I'm glad you were able to fix it. Next time I sync up with master I will retry it and report on this thread. Thanks, -Soheil From: Min Chen [min.c...@citrix.com] Sent: Thursday, September 05, 2013 2:58 PM To: dev@cloudstack.apache.org Subject: Re: Template Problem Thanks Soheil. From your description, I just realized that you are working on master branch. You may have mentioned in your earlier email, but I have missed that. In master branch, we reused state column of db table vm_template to indicate whether a template is active or not active, but we missed one place to update this column value for existing template, thus your problem. I have made a commit 072013911703c906ba6af0374d25afb494e74737 to address this issue, verified on my local and worked fine for fresh install env. Can you try again with latest code? -min On 9/5/13 11:43 AM, Soheil Eizadi seiz...@infoblox.com wrote: Hi Min, Your timing is good, today I worked on this setup some more and figured out what is going on with it. I made some changes to it and so it is not exactly what I documented before but the problem is still there. I removed the Secondary Storage from my system and installed a new secondary storage with just the SystemVM template on it, to see if that would fix the problem, but I still had the problem :( If there is a Wiki page on this area that would be helpful, I am slowly coming up to speed looking at the code and becoming more familiar it. The GUI calls list templates with the templatefilter set to all. There is a bunch of logic searching the template based on lots of knobs but mostly it boils down to displaying what is in template_view which before I changed the secondary storage and now with the new secondary storage, only has the ISO in it and that is why it not showing any templates in the GUI. (mysql dump below.) Since I deleted the secondary storage, TemplateServiceImpl detected and reinstalled the tinyLinux image in it, and it is in the Ready state in template_store_ref (mysql dump below.) There was a change to template_view where it only displays items if vm_template.state='Active', the state of the tinyLinux is 'Allocated'. (mysql dump below.) I updated it manually to Active: mysql update vm_template set state='Active' where id=5; This fixed the template_view display to show the tinyLinux in template_view and fixed the GUI. Now the question is why is it not set to Active? May be it was never set to Active as the clause to display based on vm_template.state='Active'; came into the system from schema-420to430.sql. Hope this helps, -Soheil mysql select * from vm_template; +-+-+---+- -++--+-+-+--+- -- ---++-+-+- ---+--+--- +-+---+-+--+-- ---+-+-+-+ +--+--+---+---+--+ -+--+ | id | unique_name | name | uuid | public | featured | type| hvm | bits | url | format | created | removed | account_id | checksum | display_text | enable_password | enable_sshkey | guest_os_id | bootable | prepopulate | cross_zones | extractable | hypervisor_type | source_template_id | template_tag | sort_key | size | state | update_count | updated | dynamically_scalable | +-+-+---+- -++--+-+-+--+- -- ---++-+-+- ---+--+--- +-+---+-+--+-- ---+-+-+-+ +--+--+---+---+--+ -+--+ | 1 | routing-1 | SystemVM Template (XenServer) | db876372-11c1-11e3-b37b-005056aec95c | 0 |0 | SYSTEM | 0 | 32 | http://download.cloud.com/templates/4.2/systemvmtemplate-2013-06-12-master -xen.vhd.bz2 | VHD| 2013-08-30 15:16:46 | NULL| 1 | fb1b6e032a160d86f2c28feb5add6d83 | SystemVM Template (XenServer) | 0 | 0 | 183 |1 | 0 | 1 | 0 | XenServer | NULL | NULL |0 | 276162048
which are the managed JVMs
Since I'm changing the nature of the @DB annotation (removing it really). I want to know besides the management server, what other runtimes should I be looking at to ensure everything is working properly. I'm thinking it is just AWSAPI and UsageServer only. Are there an other java applications in ACS that is spring managed? Darren
Re: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers (2)
Thanks. It must be the marvin deployment then that's causing me grief. I haven't manually filled out the UI in awhile. On Thu, Sep 5, 2013 at 4:01 PM, Jessica Wang jessica.w...@citrix.com wrote: Marcus, Internal load balancing is NOT supported in Basic zone. Therefore, if you create a Basic zone through cloudstack UI, cloudstack UI will NOT do them automatically: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf enable the InternalLbVM provider Jessica -Original Message- From: Jessica Wang [mailto:jessica.w...@citrix.com] Sent: Thursday, September 05, 2013 2:49 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi; Brian Federle Subject: RE: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers Marcus, list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf enable the InternalLbVM provider, If you create an Advanced zone through cloudStack UI, you don't need to do anything above. cloudstack UI will do them automatically. To create an Advanced zone through cloudStack UI: Infrastructure menu zones click Add zone button Add zone wizard pops up click Advanced Jessica -Original Message- From: Marcus Sorensen [mailto:shadow...@gmail.com] Sent: Thursday, September 05, 2013 2:15 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi Subject: Re: Questions about 4.2 RC More questions... I've been going down the rabbithole in figuring out how to deploy VPCs. Part of this might be a missing marvin config in my zone deploy, I still have no idea yet, but after reviewing https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-vpc-tiers.html, I found that I have to do this: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf Now I can go into the UI and enable the InternalLbVM provider, and finally create a VPC. Is this how we expect users to do it, or is there something I'm missing in the UI. On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote: -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 1:22 AM To: dev@cloudstack.apache.org; Animesh Chaturvedi Subject: Questions about 4.2 RC Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. The command is only available in non-oss build, but is still listed in command.properties, as there is no separate file for commands non-oss build. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. It's by design, the vm snapshot is listed under vm-vm snapshot view, while the volume snapshot listed under storage-snapshot view. -Sebastien
RE: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers
Marcus, list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf enable the InternalLbVM provider, If you create an Advanced zone through cloudStack UI, you don't need to do anything above. cloudstack UI will do them automatically. To create an Advanced zone through cloudStack UI: Infrastructure menu zones click Add zone button Add zone wizard pops up click Advanced Jessica -Original Message- From: Marcus Sorensen [mailto:shadow...@gmail.com] Sent: Thursday, September 05, 2013 2:15 PM To: dev@cloudstack.apache.org Cc: Animesh Chaturvedi Subject: Re: Questions about 4.2 RC More questions... I've been going down the rabbithole in figuring out how to deploy VPCs. Part of this might be a missing marvin config in my zone deploy, I still have no idea yet, but after reviewing https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-vpc-tiers.html, I found that I have to do this: list internalloadbalancerelements configure internalloadbalancerelement enabled=true id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf Now I can go into the UI and enable the InternalLbVM provider, and finally create a VPC. Is this how we expect users to do it, or is there something I'm missing in the UI. On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote: -Original Message- From: sebgoa [mailto:run...@gmail.com] Sent: Thursday, September 05, 2013 1:22 AM To: dev@cloudstack.apache.org; Animesh Chaturvedi Subject: Questions about 4.2 RC Hi, I followed the testing procedure for the 4.2 RC, I am ready to vote +1 but I have questions (hence the new thread, don't want to hijack the VOTE). 1. the CHANGES file points to the release notes, but where are the release notes ? I know that no-one really stepped up on the RN but can't we at a minimum do a JIRA query that lists the bugs fixed in 4.2 and the new features ? 2. Under infrastructure when I view the DevCloud Zone, I get this in the log: WARN [cloud.api.ApiServer] (642225853@qtp-111037821-10:) Unknown API command: listVmwareDcs This is not a deal breaker but why is our UI calling a non-implemented API ? This sounds like introducing known issues. If it's not implemented yet, remove it. The command is only available in non-oss build, but is still listed in command.properties, as there is no separate file for commands non-oss build. 3. Under instances, the new tooltip shows a 'vm snapshot' icon. When you use it, it snapshots but the snapshot is not shown under storageview snapshots ? Instead you see the snapshot under the instance view. Proceeding to destroy the instance will destroy the snapshot, or at a minimum we don't have access to it via the UI. It seems to be the difference between listVMSnapshot and listSnapshots. It's by design, the vm snapshot is listed under vm-vm snapshot view, while the volume snapshot listed under storage-snapshot view. -Sebastien
RE: Questions about 4.2 RC
-Original Message-
From: Marcus Sorensen [mailto:shadow...@gmail.com]
Sent: Thursday, September 05, 2013 3:34 PM
To: Animesh Chaturvedi
Cc: dev@cloudstack.apache.org
Subject: Re: Questions about 4.2 RC
So I guess I need to figure out of that is broken, or if it's marvin, and if
it's
marvin, how to fix it.
Now that I've gotten past that, I see the storage refactor has broken CLVM.
I'll file a bug, but it looks like the process is doing something like:
copy template to primary storage, then create copy of primary storage
template as new volume
This breaks CLVM, because it used to just do:
copy template to primary storage as new volume
Since we can't efficiently clone in CLVM, it expects to always copy the
template from secondary storage, rather than copying to primary first and
then copying the whole template from the primary back to the same disks. 1)
because it thrashes the disks, and 2) copying the template is usually much
faster because the template is sparse, and the logical volume is not, so
copying a 10G template with a real size of 500M is much faster than copying a
10G logical volume to another 10G logical volume.
in KVMStorageProcessor.java cloneVolumeFromBaseTemplate:
if (primaryPool.getType() == StoragePoolType.CLVM) {
vol = templateToPrimaryDownload(templatePath, primaryPool);
}
If the copycommand passes the template path on secondary storage will solve the
issue, right?
Let me figure it out.
This will never work, because templateToPrimaryDownload expects
secondary storage, and we have copied the template to primary storage and
are passing that. e.g.:
{
org.apache.cloudstack.storage.command.CopyCommand: {
destTO: {
org.apache.cloudstack.storage.to.VolumeObjectTO: {
accountId: 2,
dataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
format: QCOW2,
hypervisorType: KVM,
id: 9,
name: ROOT-9,
size: 1073741824,
uuid: d73f3a2b-9e63-4faf-a45b-d6fcf7633793,
vmName: i-2-9-VM,
volumeId: 9,
volumeType: ROOT
}
},
executeInSequence: true,
srcTO: {
org.apache.cloudstack.storage.to.TemplateObjectTO: {
accountId: 2,
checksum: 44cd0e6330a59f031460bc18a40c95a2,
displayText: tiny,
format: QCOW2,
hvm: true,
hypervisorType: KVM,
id: 201,
imageDataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
name: 201-2-a04f958e-0aed-3642-960f-a675a2ee1c44,
origUrl:
http://mirrors.betterservers.com/template/tiny-centos-63.qcow2;,
path: c8da0364-6f94-4c71-9c1d-74078e55bbb8,
uuid: 7dcdb1fb-e7e3-4de0-bf93-13d3e6c4ade5
}
},
wait: 0
}
}
Also, format should be 'RAW', I believe, not 'QCOW2'.
On Thu, Sep 5, 2013 at 3:40 PM, Animesh Chaturvedi
animesh.chaturv...@citrix.com wrote:
-Original Message-
From: Marcus Sorensen [mailto:shadow...@gmail.com]
Sent: Thursday, September 05, 2013 2:15 PM
To: dev@cloudstack.apache.org
Cc: Animesh Chaturvedi
Subject: Re: Questions about 4.2 RC
More questions... I've been going down the rabbithole in figuring out
how to deploy VPCs. Part of this might be a missing marvin config in
my zone deploy, I still have no idea yet, but after reviewing
https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-
v
pc-
tiers.html,
I found that I have to do this:
list internalloadbalancerelements
configure internalloadbalancerelement enabled=true
id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf
Now I can go into the UI and enable the InternalLbVM provider, and
finally create a VPC. Is this how we expect users to do it, or is
there something I'm missing in the UI.
[Animesh] Marcus from the spec The cloudStack UI will enable the
element/provider automatically as a part of physical network creation.
On Thu, Sep 5, 2013 at 2:37 PM, Edison Su edison...@citrix.com wrote:
-Original
Re: [DISCUSS] Remove async APIs from storage framework
To tell the truth, I don't like the use of cglib in this case either, it is per-request of making callback input available to IDE's intelli-sense editing and be friendly to automatic code-refactoring through IDE (Eclipse) which we do all the time. I'm fine with us to go with other standard approaches, my concern is the timing. Kelven On 9/5/13 3:25 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: Kelvin, I really have some concerns about the entire async framework that was put into place as part of the ipc framework. I don't know what VM sync is so I can't really comment too much on that. Basically my concern with the async framework is that it's yet another custom ACS thing. The use of cglib in the callbacks (or futures?) is to obscure and is not clear to a new person coming into ACS (for example, me). If we wish to do async stuff it should be based off of java 5 futures or guava's ListenableFutures. Those APIs and libraries are far more natural to java devs in the wild. Or we can get inspiration from scala and akka on how to do async well in java. In short, if we have a real use case for async, then I'd prefer we use an off the shelf library. In general I will lean towards most code being synchronous. To answer Chiradeeps comment on how do we do insane scale. That just requires true horizontal scalability of the management stack (which I'm not too sure if that's been fully proven beyond 3 servers and 30k hypervisors, but that's something we can address) and a really big database and somebody who enjoys really large failure domains. Darren On Sep 5, 2013, at 2:51 PM, Kelven Yang kelven.y...@citrix.com wrote: New VM sync and job flow handling are moving towards event-driven model which is async natural. Not sure about the details on how storage framework is using async mechanism, but argument on callback to assume memory state does not stand as strong as it might be, since arguments in any regular call pass information in memory the same way. Excessive callbacks (async method) usage pattern does encourage people to program in a wrong direction of async programming, I would prefer to substitute callback with event signaling and maintain/access flow-context through separated facility with well defined API, this can prevent developers to pass too much information directly and create a tight-coupling pattern. However, I'm a little concerning about yet another storage refactoring right after its previous refactoring work is barely done. Kelven On 9/5/13 12:57 PM, Chiradeep Vittal chiradeep.vit...@citrix.com wrote: +1 for removing complexity especially if the sync pattern is being used on top of the async pattern. I see this behavior in the AgentManager.send as well -- even thought the AM is capable of async, practically nobody uses it as such. But I guess the question will arise : what if I do want more than 10^n long-running storage jobs (cos my cloud is as successful as AWS :)) On 9/4/13 5:03 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: I've been reading over the storage code and have come to the conclusion that the async aspects of the storage framework should be removed. Whenever one introduces an async pattern you have to give a lot of consideration to its use, benefits, and impact. Within the context of ACS and given the current state of its code, I do not think it will be possible to realize any benefits of the current callback approach. Since nothing else in ACS uses callbacks, all of the async methods are essentially wrapped in synchronous calls. So nothing as it stands is actually async. Besides the current implementation, you need to conciser how you would expect an implementation of the storage framework to use the callback. The problem with callbacks is that they assume some in memory state. This means if the process/server crashes that state is lost. Many will say just serialize the callback to the DB, but that is very impractical. Since ACS doesn't actually stand in the data path, an async pattern won't really even allow it to have better performance. ACS is just waiting for some storage operation to happen. ACS can easily spawn 1000 threads and have them all wait. If you were to get to this point, you'd find that downstream you'll most likely have issues as you have 1000 create template operations so its killing your filer. So you will throttle storage operations to a level that won't kill your infrastructure and that level is no where near the scalability limits of threads. The callbacks pattern really complicates the code and I see no real benefit. Instead of spending a lot of effort trying to make all of ACS async to make it beneficial, I'd say that effort should be spent on making ACS idempotent and crash-only. The point being, there's more beneficial things we can do with our time. Given that only solidfire implements the new framework (and ACS legacy too), I
RE: [DISCUSS] OOM killer and Routing/System VM's = :(
If I recall correctly, oom actually prints something into syslog so a cron job that watches syslog and simply just shuts down the vm should work. --Alex -Original Message- From: Chiradeep Vittal [mailto:chiradeep.vit...@citrix.com] Sent: Thursday, September 5, 2013 12:48 PM To: dev@cloudstack.apache.org; Marcus Sorensen Cc: Roeland Kuipers; int-cloud Subject: Re: [DISCUSS] OOM killer and Routing/System VM's = :( Maintaining a custom kernel is a big hassle, even if it is a few lines of code change. Can we do something in userspace? What about the software watchdog that is available? Along the lines of: http://goo.gl/oO3Lzr http://linux.die.net/man/8/watchdog On 9/5/13 7:13 AM, Funs Kessen fkes...@schubergphilis.com wrote: Well, you can't as far as I've looked in the source of panic.c. So I'm thinking of investigating of adding -1 as an option and seeing if I can push halt in, let's hope the guys that do kernel stuff find this useful too. So it seems the patch, I conjured up for panic.c, is seen as not so useful, there is however another way to achieve the same result. This would mean that we load a crash kernel with our own .sh script as init to do our bidding. Would that be a plan ? Cheers, Funs Sent from my iPhone On 4 sep. 2013, at 23:35, Marcus Sorensen shadow...@gmail.com wrote: What would work as a quick fix for this sort of situation would be if the machine could be configured to power off rather than rebooting on oom. Then the HA system would restart the VM, applying all configs. Anyone know how to do that? :-) On Wed, Sep 4, 2013 at 1:14 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: On 09/04/2013 11:37 AM, Roeland Kuipers wrote: Hi Darren, Thanks for your reply! Could you share a bit more on your plans/ideas? We also have been braining on other approaches of managing the systemvm's, especially small customizations for specific tenants. And maybe even leveraging a config mgmt tools like chef or puppet with the ability to integrate CS with that in some way. I'll have to send the full details later but here's a rough idea. The basic approach is this. Logical changes to the VRs (or system vms in general) get mapped to configuration items. So add a LB rule maps to iptables config and haproxy config. When you change a LB rule we then bump up the requested version of the configuration for iptables/haproxy. So the requested version will be 4 maybe. The applied version will be 3 as the VR still has the old configuration. Since 4 != 3, the VR will be signaled to pull the latest iptables/haproxy config. So it will pull the configuration. Say in the mean time somebody else adds four other LB rules. So the requested version is now at 8. So when the VR pulls the config it will get version 8, and then reply back saying it applied version 8. The applied version is now 8 which is greater than 4 (the version the first LB rule change was waiting for) so basically all async jobs waiting for the LB change will be done. To pull the configuration from the VR, the VR will be hitting a templating configuration system. So it pulls the full iptables and haproxy config. Not incremental changes. So if the VR ever reboots itself, it can easily just pull the latest config of everything and apply it. So it will be consistent. I'd be interested to hear what type of customizations you would like to add. It will definitely be an extensible system, but the problem is if your extensions wants to touch the same configuration files that ACS wants to manage. That gets a bit tricky as its really easy for each to break each other. But I can definitely add some hooks that users can use to mess up things and void the warranty. I've thought about chef and puppet for this, but basically it comes down to two things. I'm really interested in this being fast and light weight. Ruby is neither of those. So the core ACS stuff will probably remain as very simple shell scripts. Simple in that they really just need to download configuration and restart services. They know nothing about the nature of the changes. If, as an extension, you want to do something with puppet, chef, I'd be open to that. That's your deal. This approach has many other benefits. Like, for example, we can ensure that as we deploy a new ACS release existing system VMs can be updated (without a reboot, unless the kernel changes). Additionally, its fast and updates happen in near constant time. So most changes will be just a couple of seconds, even if you have 4000 LB rules. Darren
Re: which are the managed JVMs
For removing @DB, that's pretty much it. The only other place is those JUnit tests that have physical DB calls in their test-runs Kelven On 9/5/13 4:28 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: Since I'm changing the nature of the @DB annotation (removing it really). I want to know besides the management server, what other runtimes should I be looking at to ensure everything is working properly. I'm thinking it is just AWSAPI and UsageServer only. Are there an other java applications in ACS that is spring managed? Darren
RE: Questions about 4.2 RC
Yes, preferably we skip the initial template copy altogether and go
straight to the clone with the template info and new destination disk.
Sorry, shouldn't have hijacked the thread :)
On Sep 5, 2013 5:38 PM, Edison Su edison...@citrix.com wrote:
-Original Message-
From: Marcus Sorensen [mailto:shadow...@gmail.com]
Sent: Thursday, September 05, 2013 3:34 PM
To: Animesh Chaturvedi
Cc: dev@cloudstack.apache.org
Subject: Re: Questions about 4.2 RC
So I guess I need to figure out of that is broken, or if it's marvin,
and if it's
marvin, how to fix it.
Now that I've gotten past that, I see the storage refactor has broken
CLVM.
I'll file a bug, but it looks like the process is doing something like:
copy template to primary storage, then create copy of primary storage
template as new volume
This breaks CLVM, because it used to just do:
copy template to primary storage as new volume
Since we can't efficiently clone in CLVM, it expects to always copy the
template from secondary storage, rather than copying to primary first and
then copying the whole template from the primary back to the same disks.
1)
because it thrashes the disks, and 2) copying the template is usually
much
faster because the template is sparse, and the logical volume is not, so
copying a 10G template with a real size of 500M is much faster than
copying a
10G logical volume to another 10G logical volume.
in KVMStorageProcessor.java cloneVolumeFromBaseTemplate:
if (primaryPool.getType() == StoragePoolType.CLVM) {
vol = templateToPrimaryDownload(templatePath,
primaryPool);
}
If the copycommand passes the template path on secondary storage will
solve the issue, right?
Let me figure it out.
This will never work, because templateToPrimaryDownload expects
secondary storage, and we have copied the template to primary storage and
are passing that. e.g.:
{
org.apache.cloudstack.storage.command.CopyCommand: {
destTO: {
org.apache.cloudstack.storage.to.VolumeObjectTO: {
accountId: 2,
dataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
format: QCOW2,
hypervisorType: KVM,
id: 9,
name: ROOT-9,
size: 1073741824,
uuid: d73f3a2b-9e63-4faf-a45b-d6fcf7633793,
vmName: i-2-9-VM,
volumeId: 9,
volumeType: ROOT
}
},
executeInSequence: true,
srcTO: {
org.apache.cloudstack.storage.to.TemplateObjectTO: {
accountId: 2,
checksum: 44cd0e6330a59f031460bc18a40c95a2,
displayText: tiny,
format: QCOW2,
hvm: true,
hypervisorType: KVM,
id: 201,
imageDataStore: {
org.apache.cloudstack.storage.to.PrimaryDataStoreTO: {
host: localhost,
id: 2,
path: /vg0,
poolType: CLVM,
port: 0,
uuid: 4e00fe65-c47e-4b85-afe8-4f97fb8689d0
}
},
name: 201-2-a04f958e-0aed-3642-960f-a675a2ee1c44,
origUrl:
http://mirrors.betterservers.com/template/tiny-centos-63.qcow2;,
path: c8da0364-6f94-4c71-9c1d-74078e55bbb8,
uuid: 7dcdb1fb-e7e3-4de0-bf93-13d3e6c4ade5
}
},
wait: 0
}
}
Also, format should be 'RAW', I believe, not 'QCOW2'.
On Thu, Sep 5, 2013 at 3:40 PM, Animesh Chaturvedi
animesh.chaturv...@citrix.com wrote:
-Original Message-
From: Marcus Sorensen [mailto:shadow...@gmail.com]
Sent: Thursday, September 05, 2013 2:15 PM
To: dev@cloudstack.apache.org
Cc: Animesh Chaturvedi
Subject: Re: Questions about 4.2 RC
More questions... I've been going down the rabbithole in figuring out
how to deploy VPCs. Part of this might be a missing marvin config in
my zone deploy, I still have no idea yet, but after reviewing
https://cwiki.apache.org/CLOUDSTACK/internal-load-balancing-between-
v
pc-
tiers.html,
I found that I have to do this:
list internalloadbalancerelements
configure internalloadbalancerelement enabled=true
id=de5f3495-6a48-4451-bd2f-e9cb9a7f6fbf
Now I can go into the UI and enable the InternalLbVM provider, and
finally create a VPC.
CallContext and others
In order to manage the database context at the bottom of the stack I'm
having to create a small framework that does stuff as the code enters
the managed context and leaves. So the easiest way to use the framework
is if you have
new Runnable() {
public void run() {}
}
in your code somewhere you just instead just do
new ManagedContextRunnable() {
public void doRun {...}
}
Additionally for places where there is no runnable at the bottom of the
stack (like ApiServer), there is a simple API as below.
public interface ManagedContext {
public void runWithContext(Runnable run);
public T T callWithContext(CallableT callable) throws
Exception;
public void registerListener(ManagedContextListener? listener);
public void unregisterListener(ManagedContextListener? listener);
}
The (un)register methods are for registering listeners to do stuff as
things enter/leave the context (like setup the transaction, etc).
Now this gets to my question. So this is yet another context thread
local thing that needs to be setup. So I'm somewhat trying to tie the
CallContext, ServerContext, AsyncJobExecutionContext, and this together.
So the managed context framework is the central place to wire this
all up. Here's what I'm doing, tell me if its terribly bad.
CallContext: One thing I see with the CallContext is that its really
easy to register twice and then leak the NDC stack. What I propose is
that on entering the ManagedContext I will setup the CallContext with a
context that has the user as system and the context id as random. As
people call register/unregister it will push/pop a CallContext on some
stack and additionally push/pop the NDC. When leaving the
ManagedContext I will ensure that all CallContexts are pop'd off the
stack, thus clearing the NDC. This approach has the added benefit that
by default the context will be setup as system and then all background
tasks don't explicitly have to call register(...)
ServerContexts: Just delete it? Basically it seems to just open/close
transaction and I'll be doing that anyhow. It doesn't seem to really be
used yet. Maybe my Manage Context framework essentially is in the
spirit of what that was going to be?
AsyncJobExecutionContext: Not too much, I can just ensure that the
AsyncJobExecutionContext is cleared on exit.
On last thought (sorry this is long). I also created the concept of a
ManagedThreadLocal. Its like one thread local to rule them all. Thread
locals are fine, but people forget to clear them. So what you do is
have a managed thread local that looks and acts just like a TL but is
really tied to a master TL. When the main context of the thread clears,
the master TL is cleared and all associated managed TLs will be cleared
too. Then you don't ever leak TLs.
Darren
tip for applicationContext.xml in eclipse
So if you have the Spring IDE (or STS) installed in your Eclipse you get really nice autocompletion in spring config files. So ctrl+space in the class= will pop up all the classes. Or name= on the property / will autocomplete bean properties. Really really useful, as typing class names by hand is slow and painful. But... none of this was working with ACS for me. So this is what I found. First you really should be editing applicationContext.xml.in. But that isn't an xml file, so it doesn't like that. If you look for applicationContext.xml, its in the cloudstack project which is not a java project and has no classpath and thus autocompletion can't find your classes. So what I did was in client/ ln -s tomcatconf/applicationContext.xml.in applicationContext.xml Now in cloud-client-ui you have applicationContext.xml at the root and when you open it you get autocompletion magic. Now if your on one of those operation systems that doesn't understand symlinks, I dunno. But don't fret, once I modularize spring (coming real soon) the spring XML files will all be in the right place so you get this behaviour by default. Darren
Re: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch
2013/9/6 Musayev, Ilya imusa...@webmd.net Instead of SSH, have you considered launching a console and logging in as root? Password for root is 6m1ll10n Thanks Musayev. It's really helpful. I didn't know about the root password In the future, when posting errors, try including 20 -30 lines above the actual exception as there maybe (and usually are) the pointers to the issue. Yah. will do -- N.g.U.y.e.N.A.n.H.t.U
Re: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch
Hi guys, The problem is clearer now. I logged to SSVM and CPVM and saw that they didn't receive IP addresses. Log on MS show that: http://pastebin.com/n2H4Givm Any hint? Thanks for any help :-) 2013/9/6 Musayev, Ilya imusa...@webmd.net Tuna, Instead of SSH, have you considered launching a console and logging in as root? Password for root is 6m1ll10n In the future, when posting errors, try including 20 -30 lines above the actual exception as there maybe (and usually are) the pointers to the issue. Regards ilya -Original Message- From: Nguyen Anh Tu [mailto:ng.t...@gmail.com] Sent: Thursday, September 05, 2013 11:13 AM To: dev@cloudstack.apache.org Cc: Hugo Trippaers Subject: Re: [GSoC] Can not access SSVM + CPVM with KVM + Openvswitch Ah. Finally error happens: 2013-09-05 22:10:05,748 WARN [storage.secondary.SecondaryStorageManagerImpl] (secstorage-1:null) Exception while trying to start secondary storage vm com.cloud.exception.AgentUnavailableException: Resource [Host:1] is unreachable: Host 1: Unable to start instance due to Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineMa nagerImpl.java:946) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerIm pl.java:557) at com.cloud.vm.VirtualMachineManagerImpl.start(VirtualMachineManagerIm pl.java:550) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.startSecStorag eVm(SecondaryStorageManagerImpl.java:265) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.allocCapacity( SecondaryStorageManagerImpl.java:690) at com.cloud.storage.secondary.SecondaryStorageManagerImpl.expandPool(S econdaryStorageManagerImpl.java:1297) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(Pre miumSecondaryStorageManagerImpl.java:123) at com.cloud.secstorage.PremiumSecondaryStorageManagerImpl.scanPool(Pre miumSecondaryStorageManagerImpl.java:50) at com.cloud.vm.SystemVmLoadScanner.loadScan(SystemVmLoadScanner.java :104) at com.cloud.vm.SystemVmLoadScanner.access$100(SystemVmLoadScanner.ja va:33) at com.cloud.vm.SystemVmLoadScanner$1.reallyRun(SystemVmLoadScanner.j ava:81) at com.cloud.vm.SystemVmLoadScanner$1.run(SystemVmLoadScanner.java:72 ) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) at java.util.concurrent.FutureTask$Sync.innerRunAndReset(FutureTask.java:35 1) at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.ac cess$301(ScheduledThreadPoolExecutor.java:178) at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.ru n(ScheduledThreadPoolExecutor.java:293) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.jav a:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.ja va:615) at java.lang.Thread.run(Thread.java:722) Caused by: com.cloud.utils.exception.ExecutionException: Unable to start VM[SecondaryStorageVm|s-1-VM] due to error in finalizeStart, not retrying at com.cloud.vm.VirtualMachineManagerImpl.advanceStart(VirtualMachineMa nagerImpl.java:913) ... 20 more Two system vms are destroyed and created new. 2013/9/5 Nguyen Anh Tu ng.t...@gmail.com Hi guys, I'm using an unofficial version ACS 4.1 (sdnextensions branch). I build my own cloud-agent package and install successfully into a kvm host and add this host success with two NICs: cloudbr0 (Management, Storage) and cloudbr1 (Guest, Public). I use openvswitch, follow Hugo's guide here: https://cwiki.apache.org/confluence/display/CLOUDSTACK/KVM+with+Ope nVS witch SSVM and CPVM start normally but I can not ping or ssh into via local-link (169.254.0.0/16) although both of them are already allocated IP addresses. On MS, local-link nic is cloud0 (169.254.0.1). No error log happen. In DB, they are consistently in starting state, not running. Anyone can help? -- N.g.U.y.e.N.A.n.H.t.U -- N.g.U.y.e.N.A.n.H.t.U -- N.g.U.y.e.N.A.n.H.t.U
Re: [DISCUSS] Remove async APIs from storage framework
Yeah, I understand the timing aspect. If there's already a lot in progress, then I'd rather not step in front of that train. So as long as we agree going forward we look to guava or something, then we can figure out how to get there. Regarding the storage framework, can Edison or Min chime in on how difficult/impactful it would be to move to synchronous? I could also help with it too. The longer we wait the more the impact if more storage drivers get contributed. Darren On Sep 5, 2013, at 4:49 PM, Kelven Yang kelven.y...@citrix.com wrote: To tell the truth, I don't like the use of cglib in this case either, it is per-request of making callback input available to IDE's intelli-sense editing and be friendly to automatic code-refactoring through IDE (Eclipse) which we do all the time. I'm fine with us to go with other standard approaches, my concern is the timing. Kelven On 9/5/13 3:25 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: Kelvin, I really have some concerns about the entire async framework that was put into place as part of the ipc framework. I don't know what VM sync is so I can't really comment too much on that. Basically my concern with the async framework is that it's yet another custom ACS thing. The use of cglib in the callbacks (or futures?) is to obscure and is not clear to a new person coming into ACS (for example, me). If we wish to do async stuff it should be based off of java 5 futures or guava's ListenableFutures. Those APIs and libraries are far more natural to java devs in the wild. Or we can get inspiration from scala and akka on how to do async well in java. In short, if we have a real use case for async, then I'd prefer we use an off the shelf library. In general I will lean towards most code being synchronous. To answer Chiradeeps comment on how do we do insane scale. That just requires true horizontal scalability of the management stack (which I'm not too sure if that's been fully proven beyond 3 servers and 30k hypervisors, but that's something we can address) and a really big database and somebody who enjoys really large failure domains. Darren On Sep 5, 2013, at 2:51 PM, Kelven Yang kelven.y...@citrix.com wrote: New VM sync and job flow handling are moving towards event-driven model which is async natural. Not sure about the details on how storage framework is using async mechanism, but argument on callback to assume memory state does not stand as strong as it might be, since arguments in any regular call pass information in memory the same way. Excessive callbacks (async method) usage pattern does encourage people to program in a wrong direction of async programming, I would prefer to substitute callback with event signaling and maintain/access flow-context through separated facility with well defined API, this can prevent developers to pass too much information directly and create a tight-coupling pattern. However, I'm a little concerning about yet another storage refactoring right after its previous refactoring work is barely done. Kelven On 9/5/13 12:57 PM, Chiradeep Vittal chiradeep.vit...@citrix.com wrote: +1 for removing complexity especially if the sync pattern is being used on top of the async pattern. I see this behavior in the AgentManager.send as well -- even thought the AM is capable of async, practically nobody uses it as such. But I guess the question will arise : what if I do want more than 10^n long-running storage jobs (cos my cloud is as successful as AWS :)) On 9/4/13 5:03 PM, Darren Shepherd darren.s.sheph...@gmail.com wrote: I've been reading over the storage code and have come to the conclusion that the async aspects of the storage framework should be removed. Whenever one introduces an async pattern you have to give a lot of consideration to its use, benefits, and impact. Within the context of ACS and given the current state of its code, I do not think it will be possible to realize any benefits of the current callback approach. Since nothing else in ACS uses callbacks, all of the async methods are essentially wrapped in synchronous calls. So nothing as it stands is actually async. Besides the current implementation, you need to conciser how you would expect an implementation of the storage framework to use the callback. The problem with callbacks is that they assume some in memory state. This means if the process/server crashes that state is lost. Many will say just serialize the callback to the DB, but that is very impractical. Since ACS doesn't actually stand in the data path, an async pattern won't really even allow it to have better performance. ACS is just waiting for some storage operation to happen. ACS can easily spawn 1000 threads and have them all wait. If you were to get to this point, you'd find that downstream you'll most likely have issues as you have 1000 create
Re: KVM Assistance
So, I'm trying to follow the directions here on updating my interfaces file on Ubuntu 12.04.1: http://cloudstack.apache.org/docs/en-US/Apache_CloudStack/4.1.1/html/Installation_Guide/hypervisor-kvm-install-flow.html I am running Ubuntu in a VMware Fusion VM with using what Fusion calls a custom network (I believe this is a host-only network). I'm not using VLANs. Any thoughts on how I should modify my interfaces file for my situation (below is the interfaces file from the CS docs): Thanks! auto lo iface lo inet loopback # The primary network interface auto eth0.100 iface eth0.100 inet static address 192.168.42.11 netmask 255.255.255.240 gateway 192.168.42.1 dns-nameservers 8.8.8.8 8.8.4.4 dns-domain lab.example.org # Public network auto cloudbr0 iface cloudbr0 inet manual bridge_ports eth0.200 bridge_fd 5 bridge_stp off bridge_maxwait 1 # Private network auto cloudbr1 iface cloudbr1 inet manual bridge_ports eth0.300 bridge_fd 5 bridge_stp off bridge_maxwait 1 On Fri, Aug 30, 2013 at 8:06 PM, Marcus Sorensen shadow...@gmail.comwrote: I'm not sure about the history of the agent. The libvirt api is remotely accessible, but it has proven to be a moving target, and at least early on it was insufficient. There are other things the agent does as well, which are not directly related to libvirt, like logging into virtual routers and programming them. On the Xen side I believe the APIs are sufficient for that, but on Linux /kvm the alternative would be ssh commands wrapped in ssh commands. I believe with the new architecture your plugin can still configure the SAN via mgmt server, but connecting the disk on the host needs to be done by the agent. On Aug 30, 2013 7:46 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Hey Marcus, I have a general design question here. For the XenServer and VMware plug-ins, the Java code runs on the CSMS, but for KVM it runs on the host. Can you give me an idea why there is this difference? I'm just getting up to speed on libvert, so maybe it doesn't have a daemon that runs that accepts API calls (or its API is insufficient)? Can you clarify this for me? Thanks! On Fri, Aug 30, 2013 at 7:28 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Thanks, Marcus I believe I will only need to modify LibvirtComputingResource for KVM support (based on my experience in 4.2 with CitrixResourceBase and VmwareResource). It will need to be updated to work with CloudStack's managed storage (first introduced in the new storage framework in 4.2 and supported in XenServer and VMware). Edison made large modifications to the storage component in 4.2 and I wrote a plug-in for SolidFire storage. To fully leverage what the plug-in does, I needed to modify logic in the storage framework and hypervisor plug-ins (so they didn't assume that storage was necessarily preallocated - it might be created on demand now). On Fri, Aug 30, 2013 at 5:51 PM, Marcus Sorensen shadow...@gmail.com wrote: Er. I mean normal users don't have to deal with jar files. I always install the agent code via rpm when I'm developing, rather than messing with jars, since you rarely just touch one thing in the agent, it's usually a script, or LibvirtComputingResource + some other class files, and it can be easy to miss something. It's easier to just built the packages during the compile and install the whole thing. On Fri, Aug 30, 2013 at 5:49 PM, Marcus Sorensen shadow...@gmail.com wrote: Package refers to a .deb or a .rpm, the packages used to install. Normal users never have to deal with them. Since you're not developing, you can get them from the website and install as a normal user would, but to create them look in the 'packaging' directory, there's a script for ubuntu(debs) and one for centos (rpms). On Fri, Aug 30, 2013 at 2:23 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Looking at some CS docs on the web, it appears we are using the term package here to refer to a JAR file. On Fri, Aug 30, 2013 at 2:20 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: I should be a litter clearer. :) I was referring to what you said here, Marcus: If you are developing, I'd generate the .debs, copy agent and common packages over, install those What I'm not sure how to do is generate the .deb files for CS 4.2. Also, when you say copy agent and common packages over, do you just mean the JAR files? If so, do you know where I put them? Is there an install script of some sort on the KVM side I should be running then? Thanks! On Fri, Aug 30, 2013 at 2:16 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Thanks for that info, Marcus I did have to assign a
Re: KVM Assistance
Well, you sort of have to know what you want to do, that is decide where management, guest, public traffic is going to go. You don't really need working public traffic just to run vms, and guest traffic will stay local to the vm, so that leaves management traffic, which also doesn't really need to access anything outside of the VM unless you've got an external secondary storage or are running the mgmt server on your mac or something. I think if you've solved your fusion networking problem (meaning you can create a vm and the host can talk to it), you've already got a head start with devcloud-kvm. You may have to set it to use dhcp again, ignoring the NAT setup that it does (which I mainly just use to register templates and debug VPC networking) and ignore one of the interfaces, but it's preconfigured with secondary storage, primary storage, and dependencies. In other words, I hope you didn't give up on it and start from scratch just because of the NAT issues, as it would be much easier just to tweak it to do what you're doing now. In either case, assuming you do care about networking, but host-only is fine and you want that to be management, I'd set up a bridge on eth0, assign it an IP that the host can reach (dhcp?), and then when you fill out the pod network info I'd leave the vlan field blank. Then filling out the network icons, I'd set the 'traffic label' to the name of the bridge you created for eth0. Then for public traffic I'd make up an ip range and vlan, and for guest traffic I'd also make up a vlan range. Those bridges will be created automatically, and will work just fine within the scope of the single host. You can probably find ubuntu-specific info on how to make a bridge on eth0, just a shot in the dark, might be something like: # change eth0 to have no config iface eth0 inet manual # mgt network auto cloudmgt iface cloudmgt inet dhcp bridge_ports eth0 bridge_fd 5 bridge_stp off bridge_maxwait 1 On Thu, Sep 5, 2013 at 9:05 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: So, I'm trying to follow the directions here on updating my interfaces file on Ubuntu 12.04.1: http://cloudstack.apache.org/docs/en-US/Apache_CloudStack/4.1.1/html/Installation_Guide/hypervisor-kvm-install-flow.html I am running Ubuntu in a VMware Fusion VM with using what Fusion calls a custom network (I believe this is a host-only network). I'm not using VLANs. Any thoughts on how I should modify my interfaces file for my situation (below is the interfaces file from the CS docs): Thanks! auto lo iface lo inet loopback # The primary network interface auto eth0.100 iface eth0.100 inet static address 192.168.42.11 netmask 255.255.255.240 gateway 192.168.42.1 dns-nameservers 8.8.8.8 8.8.4.4 dns-domain lab.example.org # Public network auto cloudbr0 iface cloudbr0 inet manual bridge_ports eth0.200 bridge_fd 5 bridge_stp off bridge_maxwait 1 # Private network auto cloudbr1 iface cloudbr1 inet manual bridge_ports eth0.300 bridge_fd 5 bridge_stp off bridge_maxwait 1 On Fri, Aug 30, 2013 at 8:06 PM, Marcus Sorensen shadow...@gmail.comwrote: I'm not sure about the history of the agent. The libvirt api is remotely accessible, but it has proven to be a moving target, and at least early on it was insufficient. There are other things the agent does as well, which are not directly related to libvirt, like logging into virtual routers and programming them. On the Xen side I believe the APIs are sufficient for that, but on Linux /kvm the alternative would be ssh commands wrapped in ssh commands. I believe with the new architecture your plugin can still configure the SAN via mgmt server, but connecting the disk on the host needs to be done by the agent. On Aug 30, 2013 7:46 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Hey Marcus, I have a general design question here. For the XenServer and VMware plug-ins, the Java code runs on the CSMS, but for KVM it runs on the host. Can you give me an idea why there is this difference? I'm just getting up to speed on libvert, so maybe it doesn't have a daemon that runs that accepts API calls (or its API is insufficient)? Can you clarify this for me? Thanks! On Fri, Aug 30, 2013 at 7:28 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: Thanks, Marcus I believe I will only need to modify LibvirtComputingResource for KVM support (based on my experience in 4.2 with CitrixResourceBase and VmwareResource). It will need to be updated to work with CloudStack's managed storage (first introduced in the new storage framework in 4.2 and supported in XenServer and VMware). Edison made large modifications to the storage component in 4.2 and I wrote a plug-in for SolidFire storage. To fully leverage what the plug-in does, I needed to modify logic in the storage framework and
Re: KVM Assistance
Hey Marcus, Thanks for all the detail. Yeah, I'm running the CSMS on my host OS, which is Mac OS X 10.8.3. I have VMware Fusion running Ubuntu 12.04.1 with KVM and am able to run VMs from Virtual Machine Manager. I have not yet tried adding this KVM host to CS because I was afraid my network settings were not configured properly. Currently - without making any networking changes like those described in that help doc - I've assigned an IP address statically to Ubuntu (from VMware host-only network) and can ping it from Mac OS X. I guess I wasn't sure how far I needed to take the networking to get this to the point where I could add this KVM host to CS and be able to run VMs on it. Thanks! On Thu, Sep 5, 2013 at 10:25 PM, Marcus Sorensen shadow...@gmail.comwrote: make sure all of the traffic labels are set to the name of the bridge... this is the 'edit' button under the round network icons during setup. On Thu, Sep 5, 2013 at 10:23 PM, Marcus Sorensen shadow...@gmail.com wrote: Well, you sort of have to know what you want to do, that is decide where management, guest, public traffic is going to go. You don't really need working public traffic just to run vms, and guest traffic will stay local to the vm, so that leaves management traffic, which also doesn't really need to access anything outside of the VM unless you've got an external secondary storage or are running the mgmt server on your mac or something. I think if you've solved your fusion networking problem (meaning you can create a vm and the host can talk to it), you've already got a head start with devcloud-kvm. You may have to set it to use dhcp again, ignoring the NAT setup that it does (which I mainly just use to register templates and debug VPC networking) and ignore one of the interfaces, but it's preconfigured with secondary storage, primary storage, and dependencies. In other words, I hope you didn't give up on it and start from scratch just because of the NAT issues, as it would be much easier just to tweak it to do what you're doing now. In either case, assuming you do care about networking, but host-only is fine and you want that to be management, I'd set up a bridge on eth0, assign it an IP that the host can reach (dhcp?), and then when you fill out the pod network info I'd leave the vlan field blank. Then filling out the network icons, I'd set the 'traffic label' to the name of the bridge you created for eth0. Then for public traffic I'd make up an ip range and vlan, and for guest traffic I'd also make up a vlan range. Those bridges will be created automatically, and will work just fine within the scope of the single host. You can probably find ubuntu-specific info on how to make a bridge on eth0, just a shot in the dark, might be something like: # change eth0 to have no config iface eth0 inet manual # mgt network auto cloudmgt iface cloudmgt inet dhcp bridge_ports eth0 bridge_fd 5 bridge_stp off bridge_maxwait 1 On Thu, Sep 5, 2013 at 9:05 PM, Mike Tutkowski mike.tutkow...@solidfire.com wrote: So, I'm trying to follow the directions here on updating my interfaces file on Ubuntu 12.04.1: http://cloudstack.apache.org/docs/en-US/Apache_CloudStack/4.1.1/html/Installation_Guide/hypervisor-kvm-install-flow.html I am running Ubuntu in a VMware Fusion VM with using what Fusion calls a custom network (I believe this is a host-only network). I'm not using VLANs. Any thoughts on how I should modify my interfaces file for my situation (below is the interfaces file from the CS docs): Thanks! auto lo iface lo inet loopback # The primary network interface auto eth0.100 iface eth0.100 inet static address 192.168.42.11 netmask 255.255.255.240 gateway 192.168.42.1 dns-nameservers 8.8.8.8 8.8.4.4 dns-domain lab.example.org # Public network auto cloudbr0 iface cloudbr0 inet manual bridge_ports eth0.200 bridge_fd 5 bridge_stp off bridge_maxwait 1 # Private network auto cloudbr1 iface cloudbr1 inet manual bridge_ports eth0.300 bridge_fd 5 bridge_stp off bridge_maxwait 1 On Fri, Aug 30, 2013 at 8:06 PM, Marcus Sorensen shadow...@gmail.com wrote: I'm not sure about the history of the agent. The libvirt api is remotely accessible, but it has proven to be a moving target, and at least early on it was insufficient. There are other things the agent does as well, which are not directly related to libvirt, like logging into virtual routers and programming them. On the Xen side I believe the APIs are sufficient for that, but on Linux /kvm the alternative would be ssh commands wrapped in ssh commands. I believe with the new architecture your plugin can still configure the SAN via mgmt server, but connecting the disk on the host needs to be done by the agent. On Aug 30, 2013
Re: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers (2)
On Thu, Sep 05, 2013 at 05:06:02PM -0600, Marcus Sorensen wrote:
Thanks. It must be the marvin deployment then that's causing me grief.
I haven't manually filled out the UI in awhile.
Marvin handles internallbvm deployment as well. Your providers should
now look like below.Then sync marvin and you should be ready to go.
providers: [
{
broadcastdomainrange: ZONE,
name: VirtualRouter
},
{
broadcastdomainrange: ZONE,
name: VpcVirtualRouter
},
{
broadcastdomainrange: ZONE,
name: InternalLbVm
}
],
see the setup/dev/advanced.cfg. I keep that updated usually.
--
Prasanna.,
Powered by BigRock.com
Re: Questions about 4.2 RC - internal-load-balancing-between-vpc-tiers (2)
So this look right to you?
diff --git a/tools/devcloud-kvm/devcloud-kvm-advanced.cfg
b/tools/devcloud-kvm/devcloud-kvm-advanced.cfg
index 45bdfce..8e77809 100644
--- a/tools/devcloud-kvm/devcloud-kvm-advanced.cfg
+++ b/tools/devcloud-kvm/devcloud-kvm-advanced.cfg
@@ -47,6 +47,10 @@
{
broadcastdomainrange: ZONE,
name: VpcVirtualRouter
+},
+{
+broadcastdomainrange: ZONE,
+name: InternalLbVm
}
]
},
On Thu, Sep 5, 2013 at 10:37 PM, Prasanna Santhanam t...@apache.org wrote:
On Thu, Sep 05, 2013 at 05:06:02PM -0600, Marcus Sorensen wrote:
Thanks. It must be the marvin deployment then that's causing me grief.
I haven't manually filled out the UI in awhile.
Marvin handles internallbvm deployment as well. Your providers should
now look like below.Then sync marvin and you should be ready to go.
providers: [
{
broadcastdomainrange: ZONE,
name: VirtualRouter
},
{
broadcastdomainrange: ZONE,
name: VpcVirtualRouter
},
{
broadcastdomainrange: ZONE,
name: InternalLbVm
}
],
see the setup/dev/advanced.cfg. I keep that updated usually.
--
Prasanna.,
Powered by BigRock.com
