[CLOUDSTACK DAYS SEATTLE] Call for Volunteers
Hello CloudStack committers, If you would like to volunteer to help review proposals submitted for CloudStack Day Seattle and are a CloudStack Committer, please reach out to me or Karen (cc'd). We can start reviewing the submissions post the CFP deadline which is June 12th . Thanks, Likitha
Difference between the main and the shapeblue cloudstack?
Hi all, Is there a list of the patches (and their rationale) Shapeblue includes in “their” release of Cloudstack? Frank Louwers Openminds bvba Tel: +32 9 225 82 91
回复: Re: ISO文件突然间无法使用
删除ISO 重新上传，一定没有问题。
但是我想知道原因是什么。
会不会是跟宿主机的mount 目录有关？
div...@hotmail.com
发件人： 贝为标
发送时间： 2015-05-20 18:07
收件人： users-cn
主题： Re: ISO文件突然间无法使用
div8cn,您好！
一般是存储有问题，你删了iso重新传一次试试。
=== 2015-05-20 00:14:28 您在来信中写道：===
大家好，我现在遇到一个这样的问题，一个CentOS 6.5 dvd1.ISO之前一直是正常使用着的，突然发现无法挂载，也无法通过ISO创建实例。详细情况如下
环境： CENTOS 6.5 + CLOUDSTACK 4.4 +共享存储
CLOUDSTACK 下有二个集群：KVM集群，VMWARE集群
之前这个2003的ISO 在KVM集群下使用正常，在VMWARE集群下使用也正常。
现在的故障是：通过KVM集群下的任何主机去附加这个ISO，或通过这个ISO创建KVM集群的实例，全部报错
2015-05-20 00:09:03,478 DEBUG [c.c.a.t.Request] (AgentManager-Handler-10:null)
Seq 1-3492260036049719935: Processing: { Ans: , MgmtId: 279278805450823, via:
1, Ver: v1, Flags: 10,
[{com.cloud.agent.api.Answer:{result:false,details:com.cloud.utils.exception.CloudRuntimeException:
Can't find volume:206-2-51a56f1e-72dc-3c79-8f83-0c0a1a3967f1.iso\n\tat
com.cloud.hypervisor.kvm.storage.LibvirtStoragePool.getPhysicalDisk(LibvirtStoragePool.java:150)\n\tat
com.cloud.hypervisor.kvm.resource.LibvirtComputingResource.createVbd(LibvirtComputingResource.java:3918)\n\tat
com.cloud.hypervisor.kvm.resource.LibvirtComputingResource.execute(LibvirtComputingResource.java:3789)\n\tat
com.cloud.hypervisor.kvm.resource.LibvirtComputingResource.executeRequest(LibvirtComputingResource.java:1333)\n\tat
com.cloud.agent.Agent.processRequest(Agent.java:501)\n\tat
com.cloud.agent.Agent$AgentRequestHandler.doTask(Agent.java:808)\n\tat
com.cloud.utils.nio.Task.run(Task.java:84)\n\tat
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)\n\tat
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)\n\tat
java.lang.Thread.run(Thread.java:745)\n,wait:0}}] }
但是通过这个ISO 创建VMWARE 集群下的实例则一切正常，附加到VMWARE集群下的实例ISO也能正常使用。
这个问题困惑了我好久，忘遇到并解决了的朋友给予指导，万谢！
div...@hotmail.com
= = = = = = = = = = = = = = = = = = = =
致
礼！
贝为标
beiweib...@163.com
2015-05-20
Re: SSVM agent not starting
On Wed, May 20, 2015 at 3:05 PM, Martin Emrich martin.emr...@empolis.com wrote: Hi! I built (obviously OSS) packages for el6 using the packages/centos63/package.sh script and installed them on my test management server instead of the NONOSS packages. Now it works. Is it possible to replace the NONOSS packages with OSS packages on a production server? Or are dangerous changes to the database necessary? Switching between NONOSS and OSS shouldn't tamper with the database AFAIK. Or the other way around: How can I build NONOSS packages for Ubuntu? https://cwiki.apache.org/confluence/display/CLOUDSTACK/How+to+build+CloudStack#HowtobuildCloudStack-BuildingDEBpackages -- Erik
AW: SSVM agent not starting
Hi! I built (obviously OSS) packages for el6 using the packages/centos63/package.sh script and installed them on my test management server instead of the NONOSS packages. Now it works. Is it possible to replace the NONOSS packages with OSS packages on a production server? Or are dangerous changes to the database necessary? Or the other way around: How can I build NONOSS packages for Ubuntu? Thanks, Martin -Ursprüngliche Nachricht- Von: Martin Emrich [mailto:martin.emr...@empolis.com] Gesendet: Mittwoch, 20. Mai 2015 01:00 An: users@cloudstack.apache.org Betreff: AW: SSVM agent not starting I use these packages: Management Server (CentOS 6): http://cloudstack.apt-get.eu/centos/4.4/6/ (4.4.3-NONOSS_11) KVM Host (Ubuntu 14.04.2): http://cloudstack.apt-get.eu/ubuntu/dists/trusty/4.4/ (4.4.3) System-VM-Template: http://cloudstack.apt-get.eu/systemvm/4.4/systemvm64template-4.4.1-kvm.qcow2.bz2 (I also tried http://cloudstack.apt-get.eu/systemvm/4.5/systemvm64template-4.5-kvm.qcow2.bz2 out of curiosity, same result) Are there other packages? As our production management server runs CentOS 6 (also 4.4.3-NONOSS), but I need Ubuntu for KVM hosts (as qemu/libvirt versions are too old on CentOS 6), I have no choice but to mix distributions here... Ciao Martin -Ursprüngliche Nachricht- Von: Erik Weber [mailto:terbol...@gmail.com] Gesendet: Dienstag, 19. Mai 2015 23:19 An: users@cloudstack.apache.org Betreff: Re: SSVM agent not starting Your error message is typical when you try to run non-oss mgmt server with oss system vm template (or was it the other way around - i never remember). Which system vm template did you use (for kvm)? -- Erik On Tue, May 19, 2015 at 10:29 PM, Martin Emrich martin.emr...@empolis.com wrote: Could you elaborate? I use the rpms (Management) and .debs (KVM Host) from cloudstack.apt-get.eu. The rpms say something NONOSS in the version, the debs do not. Thanks, Martin Von meinem iPhone gesendet Am 19.05.2015 um 18:23 schrieb Erik Weber terbol...@gmail.com: Did you mix non-oss with oss? Erik Den tirsdag 19. mai 2015 skrev Martin Emrich martin.emr...@empolis.com følgende: Hi! After using ACS with XenServer for some time now, I am now trying ACS (4.4.3) with KVM in my lab environment. But the SSVM does not start the agent. After stumbling upon https://cwiki.apache.org/confluence/display/CLOUDSTACK/SSVM%2C+templat es%2C+Secondary+storage+troubleshooting , I saw that I have the same error message: Unable to start agent: Resource class not found: com.cloud.storage.resource.PremiumSecondaryStorageResource due to: java.lang.ClassNotFoundException: com.cloud.storage.resource.PremiumSecondaryStorageResource Ssvm-check.sh gives me this error: ... NFS server is eth2 Ping: unknown host ... ERROR: Java process not running. Try restarting the SSVM. I can mount my secondary storage manually from inside the SSVM, so it's not a network or NFS issue, but it seems that the configuration is somehow damaged. I tried it with the SystemVM-template 4.4.1 as well as 4.5. The page mentioned did not help me find the fault... there is a systemvm.iso on both the management host and the KVM host. Any Ideas what could be wrong? Thanks Martin
Re: Compatibility of CS 4.5.1 with XenServer 6.5 SP1
I guess it wouldn't be a problem, but I'm also interested if someone has tested and verified this yet. 2015-05-19 14:57 GMT+02:00 Ian Service iserv...@ts2.ca: The release notes for CS 4.5.1 don't specifically list XenServer 6.5 SP1 compatibility, likely because they overlapped in release timelines. Are there any compatibility concerns between the two latest versions of each? Thanks, - Ian
KVM memory overprovisioning behaviour
Hi! I just tried memory overprovisioning with ACS 4.4.3 and KVM on Ubuntu 14.04. With a ratio of 2.0, a VM configured for 1024MB RAM only sees 512MB RAM (the VM is also running Ubuntu 14.04 with CONFIG_VIRT_BALLOON=y) even if the host is not yet overloaded. This is obviously useless, as I could have configured an offering with 512MB in the first place ;) I can use virsh setmem manually to claim the full 1024MB. I would expect to have the 1024MB unless the host is overloaded, in which case the VMs are reduced proportionally (as XenServer does it) or based on actual usage. There are lots of threads and wiki pages on memory overcommit or memory overprovisioning, but most seem outdated or incomplete. Is such a behaviour possible with current CloudStack, libvirt and KVM? Thanks Martin
Re: Incorrect Network setup on agent
Found the problem by googling renamed network interfaces rom pXpY
to ethZ
https://issues.apache.org/jira/browse/CLOUDSTACK-3959
On 20.05.2015 18:30, Vladislav Nazarenko wrote:
Hi Vadim,
thanx for your response ...
my setup:
1. KVM
2. Network
eth0
cloudbr0 (mgmt) (X.X.X.176/28)
public (X.X.X.0/24)
eth1
cloudbr1
guest 10.0.30.0/21 VLAN 20-39
storage 10.0.40.0/24 VLAN 40
3. Cloudstack 4.4.3 on centos 7
Advanced zone
4. Agent Conf
#Storage
#Wed May 20 18:27:31 CEST 2015
guest.network.device=cloudbr1
workers=5
private.network.device=cloudbr0
port=8250
resource=com.cloud.hypervisor.kvm.resource.LibvirtComputingResource
pod=1
zone=1
hypervisor.type=kvm
guid=f6ce77af-ebe3-31b7-b8b6-fcc58dad925e
public.network.device=cloudbr0
cluster=1
local.storage.uuid=c37e767f-911b-4503-816a-7b965a962661
domr.scripts.dir=scripts/network/domr/kvm
LibvirtComputingResource.id=0
host=X.X.X.178
Logs on the compute node:
2015-05-20 18:28:44,327{GMT} INFO [utils.nio.NioClient]
(Agent-Selector:) SSL: Handshake done
2015-05-20 18:28:44,328{GMT} INFO [utils.nio.NioClient]
(Agent-Selector:) Connected to X.X.X.178:8250
2015-05-20 18:28:44,452{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Proccess agent startup answer, agent id = 0
2015-05-20 18:28:44,452{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Set agent id 0
2015-05-20 18:28:44,453{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Startup Response Received: agent id = 0
2015-05-20 18:28:44,638{GMT} INFO [agent.transport.Request]
(UgentTask-5:) not building log message for '[{}]', _cmds.length == 1
2015-05-20 18:28:44,639{GMT} WARN [cloud.agent.Agent] (UgentTask-5:)
Unable to send request: null
2015-05-20 18:28:44,726{GMT} INFO [agent.transport.Request]
(Agent-Handler-4:) not building log message for '[{}]', _cmds.length == 1
2015-05-20 18:28:44,726{GMT} WARN [cloud.agent.Agent]
(Agent-Handler-4:) Unable to send response: null
2015-05-20 18:28:49,329{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-1:) Connected to the server
2015-05-20 18:28:49,600{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-1:) Lost connection to the server. Dealing with the
remaining commands...
Alerts in the dashboard
Incorrect Network setup on agentReinitialize agent after network names
are setupdetails : Can not find network: cloudbr0
20 May 2015 18:11:52
On 19.05.2015 21:39, Vadim Kimlaychuk wrote:
Vlad,
You should provide a little bit more than that. At least
the following:
1. Hypervisor type and version.
2. Network setup (physical and logical). In brief. I guess you use
KVM, so do you have openvswitch or do you use linux bridge?
3. Cloudstack version and type of zone you are trying to configure
(basic, advanced) ?
4. Agent conf file would be also great
Vadim
From: Vladislav Nazarenko vladislav.nazare...@gmail.com
Sent: Tuesday, May 19, 2015 19:09
To: users@cloudstack.apache.org
Subject: Incorrect Network setup on agent
Hi Everybody,
getting the following error trying to install a cloudstack agent. Can
anybody help?
Incorrect Network setup on agent, Reinitialize agent after network names
are setup, details : Can not find network: mgmtbr
Please advice
Vlad
Re: Incorrect Network setup on agent
Hi Vadim,
thanx for your response ...
my setup:
1. KVM
2. Network
eth0
cloudbr0 (mgmt) (X.X.X.176/28)
public (X.X.X.0/24)
eth1
cloudbr1
guest 10.0.30.0/21 VLAN 20-39
storage 10.0.40.0/24 VLAN 40
3. Cloudstack 4.4.3 on centos 7
Advanced zone
4. Agent Conf
#Storage
#Wed May 20 18:27:31 CEST 2015
guest.network.device=cloudbr1
workers=5
private.network.device=cloudbr0
port=8250
resource=com.cloud.hypervisor.kvm.resource.LibvirtComputingResource
pod=1
zone=1
hypervisor.type=kvm
guid=f6ce77af-ebe3-31b7-b8b6-fcc58dad925e
public.network.device=cloudbr0
cluster=1
local.storage.uuid=c37e767f-911b-4503-816a-7b965a962661
domr.scripts.dir=scripts/network/domr/kvm
LibvirtComputingResource.id=0
host=X.X.X.178
Logs on the compute node:
2015-05-20 18:28:44,327{GMT} INFO [utils.nio.NioClient]
(Agent-Selector:) SSL: Handshake done
2015-05-20 18:28:44,328{GMT} INFO [utils.nio.NioClient]
(Agent-Selector:) Connected to X.X.X.178:8250
2015-05-20 18:28:44,452{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Proccess agent startup answer, agent id = 0
2015-05-20 18:28:44,452{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Set agent id 0
2015-05-20 18:28:44,453{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-2:) Startup Response Received: agent id = 0
2015-05-20 18:28:44,638{GMT} INFO [agent.transport.Request]
(UgentTask-5:) not building log message for '[{}]', _cmds.length == 1
2015-05-20 18:28:44,639{GMT} WARN [cloud.agent.Agent] (UgentTask-5:)
Unable to send request: null
2015-05-20 18:28:44,726{GMT} INFO [agent.transport.Request]
(Agent-Handler-4:) not building log message for '[{}]', _cmds.length == 1
2015-05-20 18:28:44,726{GMT} WARN [cloud.agent.Agent]
(Agent-Handler-4:) Unable to send response: null
2015-05-20 18:28:49,329{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-1:) Connected to the server
2015-05-20 18:28:49,600{GMT} INFO [cloud.agent.Agent]
(Agent-Handler-1:) Lost connection to the server. Dealing with the
remaining commands...
Alerts in the dashboard
Incorrect Network setup on agentReinitialize agent after network names
are setupdetails : Can not find network: cloudbr0
20 May 2015 18:11:52
On 19.05.2015 21:39, Vadim Kimlaychuk wrote:
Vlad,
You should provide a little bit more than that. At least the
following:
1. Hypervisor type and version.
2. Network setup (physical and logical). In brief. I guess you use KVM, so do
you have openvswitch or do you use linux bridge?
3. Cloudstack version and type of zone you are trying to configure (basic,
advanced) ?
4. Agent conf file would be also great
Vadim
From: Vladislav Nazarenko vladislav.nazare...@gmail.com
Sent: Tuesday, May 19, 2015 19:09
To: users@cloudstack.apache.org
Subject: Incorrect Network setup on agent
Hi Everybody,
getting the following error trying to install a cloudstack agent. Can
anybody help?
Incorrect Network setup on agent, Reinitialize agent after network names
are setup, details : Can not find network: mgmtbr
Please advice
Vlad
Re: Compatibility of CS 4.5.1 with XenServer 6.5 SP1
I am planning to do it within a couple of days. Will let you know Vadim. From: Jan-Arve Nygård jan.arve.nyg...@gmail.com Sent: Wednesday, May 20, 2015 15:38 To: users@cloudstack.apache.org Subject: Re: Compatibility of CS 4.5.1 with XenServer 6.5 SP1 I guess it wouldn't be a problem, but I'm also interested if someone has tested and verified this yet. 2015-05-19 14:57 GMT+02:00 Ian Service iserv...@ts2.ca: The release notes for CS 4.5.1 don't specifically list XenServer 6.5 SP1 compatibility, likely because they overlapped in release timelines. Are there any compatibility concerns between the two latest versions of each? Thanks, - Ian
Re: Compatibility of CS 4.5.1 with XenServer 6.5 SP1
We have tested upgrading to XS 6.5 SP1 with CloudStack 4.5.1 and we didn't run into any issues. We tested the following without any problems: Start of existing VM Deploy VM Destroy VM Attach ISO Detach ISO Snapshot of VM Live migration of VM to another host Storage migration to another SR Add NIC Remove NIC And probably some other stuff that i don't remember right now. -thinktwo 2015-05-20 21:17 GMT+02:00 Vadim Kimlaychuk vadim.kimlayc...@elion.ee: I am planning to do it within a couple of days. Will let you know Vadim. From: Jan-Arve Nygård jan.arve.nyg...@gmail.com Sent: Wednesday, May 20, 2015 15:38 To: users@cloudstack.apache.org Subject: Re: Compatibility of CS 4.5.1 with XenServer 6.5 SP1 I guess it wouldn't be a problem, but I'm also interested if someone has tested and verified this yet. 2015-05-19 14:57 GMT+02:00 Ian Service iserv...@ts2.ca: The release notes for CS 4.5.1 don't specifically list XenServer 6.5 SP1 compatibility, likely because they overlapped in release timelines. Are there any compatibility concerns between the two latest versions of each? Thanks, - Ian
Multi-point VPC DRS on XenServer utilizing CloudStack
Is it possible to do a multiple VPC IPsec tunnel? I have CloudStack 4.5.1 installed and VPC works find connecting to PfSense physical firewall on site at a location. Now I have a second location that would also like a IPsec tunnel to the hosted servers in CloudStack. I am using Advanced networking zones. Also I have a physical server outside of CloudStack I am going to take the VLAN CloudStack assigned the VM on the private network and assign that specific VLAN on the 5596 to the 10G port connected to the server. It doesn't look like it would cause a problem but does anyone have any ideas if it is a bad plan. Also I want to look into a CloudStack script that does similar works as DRS on Vmware. My Hypervisor if XenServer 6.5 SP1 and I'm using shared storage via iSCSI multipath. I don't want to have to worry about hypervisor over utilization so since CloudStack checks the hosts utilization and the VM utilization why can I not use both of those features along with live migration to offset the load on hosts. If this is already available and I just can't find it that would be awesome. Jeremy
Re: Multi-point VPC DRS on XenServer utilizing CloudStack
Hi, It is possible to have multiple IPsec tunnels from vpc with each tunnel pointing to different gateway. Your VLAN configuration works perfectly fine. Cloudstack can check the host and storage utilization and allocates the host and storage based on the availability of resources for vm deployment. What do you mean by VM utilization? -Sanjeev On Thu, May 21, 2015 at 6:23 AM, Jeremy Peterson jpeter...@acentek.net wrote: Is it possible to do a multiple VPC IPsec tunnel? I have CloudStack 4.5.1 installed and VPC works find connecting to PfSense physical firewall on site at a location. Now I have a second location that would also like a IPsec tunnel to the hosted servers in CloudStack. I am using Advanced networking zones. Also I have a physical server outside of CloudStack I am going to take the VLAN CloudStack assigned the VM on the private network and assign that specific VLAN on the 5596 to the 10G port connected to the server. It doesn't look like it would cause a problem but does anyone have any ideas if it is a bad plan. Also I want to look into a CloudStack script that does similar works as DRS on Vmware. My Hypervisor if XenServer 6.5 SP1 and I'm using shared storage via iSCSI multipath. I don't want to have to worry about hypervisor over utilization so since CloudStack checks the hosts utilization and the VM utilization why can I not use both of those features along with live migration to offset the load on hosts. If this is already available and I just can't find it that would be awesome. Jeremy
RE: KVM memory overprovisioning behaviour
Hi, Just check the agent.properties on the KVM host /etc/cloudstack/agent/agent.properties vm.memballoon.disable=true set the mem.overprovisioning.factor in the Global Settings to 2 Try provision again Thanks Glenn Wagner Senior Consultant, South Africa Phone: +27 21 527 0091 | Mobile: +27 73 917 4111 glenn.wag...@shapeblue.com | www.shapeblue.com | Twitter:@shapeBlue ShapeBlue SA (Pty) Ltd, 2nd Floor, Oudehuis Centre, 122 Main Rd, Somerset West, Cape Town 7130 -Original Message- From: Martin Emrich [mailto:martin.emr...@empolis.com] Sent: 20 May 2015 03:57 PM To: users@cloudstack.apache.org Subject: KVM memory overprovisioning behaviour Hi! I just tried memory overprovisioning with ACS 4.4.3 and KVM on Ubuntu 14.04. With a ratio of 2.0, a VM configured for 1024MB RAM only sees 512MB RAM (the VM is also running Ubuntu 14.04 with CONFIG_VIRT_BALLOON=y) even if the host is not yet overloaded. This is obviously useless, as I could have configured an offering with 512MB in the first place ;) I can use virsh setmem manually to claim the full 1024MB. I would expect to have the 1024MB unless the host is overloaded, in which case the VMs are reduced proportionally (as XenServer does it) or based on actual usage. There are lots of threads and wiki pages on memory overcommit or memory overprovisioning, but most seem outdated or incomplete. Is such a behaviour possible with current CloudStack, libvirt and KVM? Thanks Martin Find out more about ShapeBlue and our range of CloudStack related services IaaS Cloud Design Buildhttp://shapeblue.com/iaas-cloud-design-and-build// CSForge – rapid IaaS deployment frameworkhttp://shapeblue.com/csforge/ CloudStack Consultinghttp://shapeblue.com/cloudstack-consultancy/ CloudStack Software Engineeringhttp://shapeblue.com/cloudstack-software-engineering/ CloudStack Infrastructure Supporthttp://shapeblue.com/cloudstack-infrastructure-support/ CloudStack Bootcamp Training Courseshttp://shapeblue.com/cloudstack-training/ 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 a company incorporated in India and is operated under license from Shape Blue Ltd. Shape Blue Brasil Consultoria Ltda is a company incorporated in Brasil and is operated under license from Shape Blue Ltd. ShapeBlue SA Pty Ltd is a company registered by The Republic of South Africa and is traded under license from Shape Blue Ltd. ShapeBlue is a registered trademark.
Re: KVM memory overprovisioning behaviour
With a ratio of 2.0, a VM configured for 1024MB RAM only sees 512MB RAM This is by desing - so memory overprovisioning is useless in this particular sense, and not to be used AFAIK. CPU - maybe, storage - maybe/yes... On 20 May 2015 at 15:57, Martin Emrich martin.emr...@empolis.com wrote: Hi! I just tried memory overprovisioning with ACS 4.4.3 and KVM on Ubuntu 14.04. With a ratio of 2.0, a VM configured for 1024MB RAM only sees 512MB RAM (the VM is also running Ubuntu 14.04 with CONFIG_VIRT_BALLOON=y) even if the host is not yet overloaded. This is obviously useless, as I could have configured an offering with 512MB in the first place ;) I can use virsh setmem manually to claim the full 1024MB. I would expect to have the 1024MB unless the host is overloaded, in which case the VMs are reduced proportionally (as XenServer does it) or based on actual usage. There are lots of threads and wiki pages on memory overcommit or memory overprovisioning, but most seem outdated or incomplete. Is such a behaviour possible with current CloudStack, libvirt and KVM? Thanks Martin -- Andrija Panić
