Re: 4.2 KVM agent can't communicate with 4.1 management server

[Wido den Hollander]

On 07/22/2013 11:29 PM, Toshiaki Hatano wrote:

I got a question from this issue.
Should cloudstack allow version mismatch between management and agent?



I already discussed this on IRC yesterday and I wanted to start a new 
[DISCUSS] thread around this, didn't get to do that yet.


Wido


I think it's very good to have for ops.
But it looks very difficult for dev, since every little change could
break the interoperability. (We have to test it?)

--
Toshiaki

-Original Message-
From: Wido den Hollander [mailto:w...@widodh.nl]
Sent: Monday, July 22, 2013 07:53
To: dev@cloudstack.apache.org
Subject: 4.2 KVM agent can't communicate with 4.1 management server

Hi,

While reviewing 12775 I upgraded my Agents from 4.1 to 4.2, but kept my
management server at 4.1

When the Agent starts it sends a StartupRoutingCommand to the management
server, but this has changed it seems:

In 4.1 the Agent sends this JSON:

Sending Startup: Seq 4-0:  { Cmd , MgmtId: -1, via: 4, Ver: v1, Flags:
1, [{"StartupRoutingCommand":{

In 4.2 however the JSON data starts with:

Sending Startup: Seq 1-6:  { Cmd , MgmtId: -1, via: 1, Ver: v1, Flags:
1, [{"com.cloud.agent.api.StartupRoutingCommand":{

So the Agent sends the full name of the class and this confuses the
Management server, it throws an Exception:

Caused by: com.cloud.utils.exception.CloudRuntimeException: can't find
com.cloud.agent.api.com.cloud.agent.api.StartupRoutingCommand
at
com.cloud.agent.transport.ArrayTypeAdaptor.deserialize(ArrayTypeAdaptor.
java:79)
at
com.cloud.agent.transport.ArrayTypeAdaptor.deserialize(ArrayTypeAdaptor.
java:37)
at
com.google.gson.JsonDeserializerExceptionWrapper.deserialize(JsonDeseria
lizerExceptionWrapper.java:51)
... 15 more

So it's searching for
"com.cloud.agent.api.com.cloud.agent.api.StartupRoutingCommand" which
obviously fails.

I'm not sure how to fix this, since StartupRoutingCommand simply calls
"RouterPrivateIpStrategy.class.getCanonicalName()"

I created this issue for it:
https://issues.apache.org/jira/browse/CLOUDSTACK-3714

Any suggestions on how to fix this?

Wido



This email message is intended for the use of the person to whom it has been 
sent, and may contain information that is confidential or legally protected. If 
you are not the intended recipient or have received this message in error, you 
are not authorized to copy, distribute, or otherwise use this message or its 
attachments. Please notify the sender immediately by return e-mail and 
permanently delete this message and any attachments. Verio Inc. makes no 
warranty that this email is error or virus free.  Thank you.

Re: Review Request 12623: CLOUDSTACK-2328: Linux native VXLAN support on KVM hypervisor

[Chiradeep Vittal]

> On July 19, 2013, 4:56 p.m., Chiradeep Vittal wrote:
> > scripts/vm/network/vnet/modifyvxlan.sh, line 28
> > 
> >
> > I think there is a need to prevent the guest vm from spoofing the 
> > multicast?
> 
> Toshiaki Hatano wrote:
> We don't need to.
> 
> VM can only see Inner Frame of VXLAN packet.
> VXLAN uses multicast for Outer Packet that VM cannot see or manipulate.
> 
> Below is bridging diagram within KVM host.
> 
> InnerFrame: VM <-> eth*|vnet* <-> brethX-Y <-> vxlanY
>  || (*1)  
> 
> OuterPacket:   cloudbr* <-> eth* ==> 
> Outside of the Host   
> 
> 
> All frame that VM sent are encapsulated at (*1).
> Almost frames are mapped to unicast here, since vxlanY interface learns 
> mapping between other VMs' MAC address and Hosts IP address.
> Only when vxlanY interface haven't learned mapping yet or inner frame is 
> multicast or unicast frame, vxlanY interface uses multicast group statically 
> assigned at line 33 of modifyvxlan.sh.
> 
> Multicast group is assigned statically to vxlan interface of host, so VM 
> cannot spoof multicast group.

Yes, I agree.
What about throttling multicasts to prevent broadcast/multicast storms (using 
ebtables for example)?


On July 19, 2013, 4:56 p.m., Toshiaki Hatano wrote:
> > I just wanted to make sure that you have tested your patch with regular 
> > VLANs as well. 
> > And, what the behavior will be when VxLAN is enabled in the zone, but only 
> > Xen / VMW hypervisors are there
> > Also, some documentation on how the cloud operator can get this feature 
> > (KVM version/Which version of Centos/Ubuntu/etc), configuration of 
> > switches, bridges, etc would be useful.
> 
> Toshiaki Hatano wrote:
> Yes, I did same test with regular VLANs and it works fine.
> 
> I don't have Xen nor VMWare to test with, but as far as I read the code...
> Xen-agent would rise CloudRuntimeException("Unable to support this type 
> of network broadcast domain: " + nic.getBroadcastUri()) in 
> com.cloud.hypervisor.xen.resource.CitrixResourceBase.getNetwork(Connection, 
> NicTO) before they actually submit VIF.create to hypervisor.
> VMWare-agent would warn("Unrecognized broadcast type in VmwareResource, 
> type: " + nicTo.getBroadcastType().toString() + ". Use vlan info from 
> labeling: " + defaultVlan) in 
> com.cloud.hypervisor.vmware.resource.VmwareResource.getVlanInfo(NicTO, 
> String) and assign vlan interface with defultVlan instead of VXLAN.
> 
> In short, Xen-agent handle unknown isolation type as error and don't 
> start VM.
> VMWare-agent just ignore unknown isolation type and start VM with default 
> VLAN.
> 
> 
> Yes, I will write documentation.
> Is that requirement for this patch to be committed?

Generally speaking we need documentation. We have so many network plugins 
committed without any documentation, that *IMHO* renders them useless. QA is 
not able to include these test cases because they do not understand the 
feature. Developers will break compatibility because they do not know how to 
replicate / test the feature. And docs will ignore it because they have never 
heard of it.

Re: compatibility with other hypervisors, can you bring up this issue on the 
dev mailing list and we can discuss a better solution there?

Otherwise the patch is good.


- Chiradeep


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12623/#review23523
---


On July 23, 2013, 6:45 a.m., Toshiaki Hatano wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12623/
> ---
> 
> (Updated July 23, 2013, 6:45 a.m.)
> 
> 
> Review request for cloudstack, Alena Prokharchyk, Chiradeep Vittal, Murali 
> Reddy, Hugo Trippaers, and Sheng Yang.
> 
> 
> Bugs: https://issues.apache.org/jira/browse/CLOUDSTACK-2328
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-2328: Linux native VXLAN support on KVM hypervisor
> 
> Initial patch for VXLAN support.
> Fully functional, hopefully, for GuestNetwork - AdvancedZone.
> 
> Patch Note:
>  in cloudstack-server
> - Add isolation method VXLAN
> - Add VxlanGuestNetworkGuru as plugin for VXLAN isolation
> - Modify NetworkServiceImpl to handle extended vNet range for VXLAN isolation
> - Add VXLAN isolation option in zoneWizard UI
> 
>  in cloudstack-agent (kvm)
> - Add modifyvxlan.sh script that handle bridge/vxlan interface manipulation 
> script
> -- Usage is exactly same to modifyvlan.sh
> - BridgeVifDriver wil

Re: [DESIGN] Why is enum a class...

[Daan Hoogland]

+1 I am having a war on the enums in Networks.java that justifies your
solution, Frank


On Tue, Jul 23, 2013 at 2:57 AM, Alex Huang  wrote:

> +1 to do this for any enum that was designed to be added to by the plugins.
>
> I actually wrote a class for this before called Constant that did exactly
> this.  The problem I had was that there was no way to guarantee compile
> time enumeration of all of the values like Enum did so it's possible for
> errors to be introduced but certainly we can follow a convention in this
> regard.  Another way would be to use Spring to gather them and inject them
> into a holding class.
>
> --Alex
>
> > -Original Message-
> > From: Frank Zhang [mailto:frank.zh...@citrix.com]
> > Sent: Monday, July 22, 2013 5:45 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [DESIGN] Why is enum a class...
> >
> > Frankly speaking, if we are going to change enum, I would suggest not
> using
> > enmu anymore, instead, defining our own class like:
> >
> > public class VmType {
> >   private static Map types =
> > Collections.synchronizedMap(new HashMap());
> >   private final String typeName;
> >
> >   public VmType (String typeName) {
> >   this.typeName = typeName;
> >   types.put(typeName, this);
> >   }
> >
> >   public static VmType valueOf(String typeName) {
> >   VmType type = types.get(typeName);
> >   if (type == null) {
> >   throw new IllegalArgumentException("VmType type:
> > " + typeName + " was not registered by any VmType");
> >   }
> >   return type;
> >   }
> >
> >   @Override
> >   public String toString() {
> >   return typeName;
> >   }
> >
> >   @Override
> >   public boolean equals(Object t) {
> >   if (t == null || !(t instanceof VmType)) {
> >   return false;
> >   }
> >
> >   VmType type = (VmType)t;
> >   return type.toString().equals(typeName);
> >   }
> >
> >   @Override
> >   public int hashCode() {
> >   return typeName.hashCode();
> >   }
> >
> >   public static Set getAllTypeNames() {
> >   return types.keySet();
> >   }
> > }
> >
> > The only benefit of enum I see is you can use "==" instead of "equals()"
> > when comparing variables. However, it makes your code tight, every time
> > adding a new type you have to modify the enum.
> >
> > By above method, when a new plugin wants to extend a new VmType, it
> > simply does:
> >
> > class MagicVmManagerImpl {
> > public static final VmType type = new VmType("MagicVm"); }
> >
> >
> > > -Original Message-
> > > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > > Sent: Monday, July 22, 2013 5:23 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [DESIGN] Why is enum a class...
> > >
> > > BTW, this code already shows a bug that stems from the static method
> > usage.
> > > It says ElasticIpVm is not a system vm (which I don't believe is
> > > true).  Probably because whoever added elastic ip vm didn't see the
> > > static method and so didn't add the vm into the method.
> > >
> > > --Alex
> > >
> > > > -Original Message-
> > > > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > > > Sent: Monday, July 22, 2013 5:14 PM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: [DESIGN] Why is enum a class...
> > > >
> > > > I just went over this code and thought it was related and might be
> > > > interested to other developers.
> > > >
> > > > What's the difference between declaring a enum like this
> > > >
> > > > public enum Type {
> > > > User,
> > > > DomainRouter,
> > > > ConsoleProxy,
> > > > SecondaryStorageVm,
> > > > ElasticIpVm,
> > > > ElasticLoadBalancerVm,
> > > > InternalLoadBalancerVm,
> > > >
> > > > /*
> > > >  * UserBareMetal is only used for selecting
> > > > VirtualMachineGuru, there is no
> > > >  * VM with this type. UserBareMetal should treat exactly as
> User.
> > > >  */
> > > > UserBareMetal;
> > > >
> > > > public static boolean isSystemVM(VirtualMachine.Type vmtype)
> {
> > > > if (DomainRouter.equals(vmtype)
> > > > || ConsoleProxy.equals(vmtype)
> > > > || SecondaryStorageVm.equals(vmtype) ||
> > > > InternalLoadBalancerVm.equals(vmtype)) {
> > > > return true;
> > > > }
> > > > return false;
> > > > }
> > > > }
> > > >
> > > > Vs
> > > >
> > > > public enum Type {
> > > > User(false),
> > > > DomainRouter(true),
> > > > ConsoleProxy(true),
> > > > SecondaryStorageVm(true),
> > > > ElasticIpVm(true),
> > > > ElasticLoadBalancerVm(true),
> > > > InternalLoadBalancerVm(true),
> > > >
> > > > /*
> > > >  * UserBar

Re: Review Request 12720: CLOUDSTACK: 3382 Unable to Migrate VM's If the hosts are implicitly or explicitly dedicated.

[Devdeep Singh]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12720/#review23682
---

Ship it!


Committed to 4.2 and master in commits e2f2bc5f0 and 89b94bb75 respectively.

- Devdeep Singh


On July 22, 2013, 11:28 a.m., Saksham Srivastava wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12720/
> ---
> 
> (Updated July 22, 2013, 11:28 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: 3382
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Allow root admin to migrate VMs across dedicated hosts.
> Generate alerts for each migration performed between dedicated hosts.
> Alerts are generated for VM migration between:
> 1) Source host is dedicated and destination host is not.
> 2) Source host is not dedicated and destination host is dedicated.
> 3) Both hosts are dedicated.
> Alerts are now generated for both explicit and implicit dedication.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/deploy/dao/PlannerHostReservationDao.java 69118f1 
>   server/src/com/cloud/deploy/dao/PlannerHostReservationDaoImpl.java 41e0964 
>   server/src/com/cloud/vm/UserVmManagerImpl.java 9968690 
> 
> Diff: https://reviews.apache.org/r/12720/diff/
> 
> 
> Testing
> ---
> 
> Migrated VMs across explicitly dedicated hosts and alerts are generated.
> Migrated VMs across implicitly dedicated hosts and alerts are generated.
>  
> 
> 
> Thanks,
> 
> Saksham Srivastava
> 
>

Re: Tracing registerTemplate API, is it possible to connect Eclipse debugger to System VMs?

[Ryan Lei]

Great! I have successfully connected Eclipse debugger to the System VM
agent, but the steps are slightly different. See below:

* Step 4: I put that long string of arguments right after "java" instead of
at the very end.

If put at the very end, the command
"java -Djavax.net.ssl.trustStore=./certs/realhostip.keystore -mx${maxmem}m
-cp $CP com.cloud.agent.AgentShell $keyvalues $@ -Xdebug
-Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=8787" would result
in invalid parameters:

Invalid Parameter: -Xdebug
Invalid Parameter:
-Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=8787

* Step 6: I used "iptables -A INPUT -i eth1 -p tcp -m state --state NEW -m
tcp --dport 8787 -j ACCEPT" instead, as suggested by the Internet.
"iptables -I INPUT -I eth1 -p tcp -m state --state NEW -m tcp --dport 8787
-j ACCEPT" gave me error "iptables v1.4.8: Cannot use -I with -I"

The first -I may be OK, but at least the second -I should be the lowercase
-i, which means the network interface.

I'm not very familiar with such commands, so I'm not 100% sure if I'm
right. If necessary, please modify the wiki page for me.

---
Yu-Heng (Ryan) Lei, Associate Reasearcher
Chunghwa Telecom Laboratories / Cloud Computing Laboratory
ryan...@cht.com.tw
or
ryanlei750...@gmail.com


On Tue, Jul 23, 2013 at 4:53 AM, Edison Su  wrote:

> Yah, you need Eclipse connects to the java agent running on SSVM.
> You can use the following steps:
> 1. login into ssvm, either console proxy or ssh(port 3922)
> 2. kill all the processes named as(run.sh/_run.sh, and java)
> 3. cd /usr/local/cloud/systemvm
> 4. add the " -Xdebug
> -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address={8787}" at end of
> "java " in _run.sh
> 5. ./run.sh, the java agent will start, with debug port 8787 is listened
> on.
> 6. allow port 8787 in ssvm, "iptables -I INPUT -I eth1 -p tcp -m state
> --state NEW -m tcp --dport 8787 -j ACCEPT", either eth1 or eth2 is ok.
> Then you can connect to agent in ssvm, based on the ip address based on on
> eth1 or eth2, and port 8787
>
> Or above steps sound too complicated to you, you can debug resource code
> locally.
> For example, There is an unit test code, called,
> LocalNfsSecondaryStorageResourceTest, on 4.2 branch, you can mock the
> commands send to resource, then test it locally, but the requirement is
> that your local machine needs to be able to mount nfs(Both Linux/Mac works,
> Windows can't).
> I'd prefer unit test, as it's easier to test/debug/fix.
>
> > -Original Message-
> > From: ryanlei750...@gmail.com [mailto:ryanlei750...@gmail.com] On
> > Behalf Of Ryan Lei
> > Sent: Monday, July 22, 2013 3:10 AM
> > To: dev@cloudstack.apache.org
> > Subject: Tracing registerTemplate API, is it possible to connect Eclipse
> > debugger to System VMs?
> >
> > Hi, I've been trying to study how System VMs work in CloudStack code.
> >
> > Tracing the creation of SSVM and CPVM was not a problem. It has to do
> with
> > updateZone API (&allocationstate=Enabled) and SystemVmLoadScanner.java
> > for both secondary storage and console proxy.
> >
> > But the next usage case, registering a template, was not as
> straightforward.
> > Tracing the registerTemplate API lead me to a dead end in the send()
> > method in cloud-utils -> com.cloud.utils.nio.Link.java, whose previous
> step is
> > _link.send(req.toBytes()) in cloud-server ->
> > com.cloud.agent.manager.ConnectedAgentAttache.java.
> >
> > From the MS log, it seems after the DownloadCommand is sent in
> > agent.transport.Request, the registerTemplate API ends.
> > http://apaste.info/pM1E
> >
> > So how do I trace the next Java steps on the SSVM part? Is it possible to
> > connect Eclipse debugger to System VMs? Or what is the best way to trace
> > the interaction between MS and System VMs that involves an agent
> > command?
> >
> > Thanks
> >
> >
> ---
> > Yu-Heng (Ryan) Lei, Associate Reasearcher Chunghwa Telecom Laboratories /
> > Cloud Computing Laboratory
> > ryan...@cht.com.tw > wE1FEC3G0SWYpVkiWo8SsDdf3ZqO9AIuAPTzRnFYCUi-
> > z4YljtI_hyVKkNHfn9F1Bn-vUWJnQ4.&URL=mailto%3aryanlei%40cht.com.tw>
> > or
> > ryanlei750...@gmail.com
>

Re: Basic Zone network configuration

[Thomas Schneider]

Do you have an example of /etc/network/interface config for the host in
basic zone mode ?

Le 23/07/2013 04:54, Ahmad Emneina a écrit :
> Hey Thomas,
>
> Basic zones dont explicitly work with vlans. If you have vlan's behind the
> scenes, say native vlans, it shouldnt be an issue as long as your relevant
> networks can route to each other.
>
> Say for example I have my management server sitting on the
> 192.168.10.0/24segment with hosts, system vm's and guest vm's on the
> 192.168.11.0/24 segment, there shouldnt be an issue. All assuming they can
> communicate back and forth between those networks. HTH
>
>
> On Mon, Jul 22, 2013 at 9:36 AM, Thomas Schneider <
> thomas.schnei...@euskill.com> wrote:
>
>> Hi
>>
>> Do you have some examples of Basic Zone network configuration ?
>> Is it possible to configure it without using different vlan ?
>>
>> At this stage I configured manually the network without using vlans and
>> the consequence is that the System VM is stuck in starting mode.
>>
>> Best Regards,
>>
>> --
>> *Thomas Schneider*
>>
>>


-- 
*Thomas Schneider*
Directeur des Opérations
Euskill SARL
Web: www.euskill.com
Mobile: +33 (0)6 19 26 47 76
Mail: thomas.schnei...@euskill.com
5 rue de Phalsbourg
F-67000 Strasbourg

Review Request 12849: added backwards compatibility code to Networks enums

[daan Hoogland]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12849/
---

Review request for cloudstack and Koushik Das.


Repository: cloudstack-git


Description
---

Both BroadcastDomainType and IsolationType needed some extra code for backwards 
compatibility


Diffs
-

  api/src/com/cloud/network/Networks.java c76c3d4 
  
server/src/com/cloud/network/router/VpcVirtualNetworkApplianceManagerImpl.java 
6fafa3e 

Diff: https://reviews.apache.org/r/12849/diff/


Testing
---


Thanks,

daan Hoogland

Re: System VMs not coming up due to https://reviews.apache.org/r/12685/

[Daan Hoogland]

Koushik,

please review https://reviews.apache.org/r/12849/ It should solve the
problem.

regards,
Daan


On Mon, Jul 22, 2013 at 8:40 PM, Daan Hoogland wrote:

> Ignore my remark about master. I have been looking
> At an old log.
> I looked some more into this and it seems the even simpler
> if (scheme == Vlan.scheme) {
>
> return new URI(scheme + "://" + value);
> }
> would do the trick. I have not been able to reproduce the problem reliably
> yet.
>
> Daan,
>
> After manually updating the database entries to 'vlan://', the system VMs
> started working. So I don't think using a database before the commit is a
> problem. One piece of code that seems broken is
> CitrixResourceBase.getNetwork().
>
> Refer to the line of code
>
> long vlan = Long.parseLong(broadcastUri.getHost());
>
> This fails with a NumberFormatException for vlan:.
>
> -Koushik
>
> On 22-Jul-2013, at 9:38 PM, Daan Hoogland  wrote:
>
> > Koushik,
> >
> > thanks for reporting this. You are probably using a database form before
> > this commit. This should work but I suspect it doesn't.
> >public  URI toUri(T value) {
> >try {
> >// do we need to check that value does not contain a
> scheme
> >// part?
> >try {
> >Long.parseLong(value.toString());
> >return new URI(Vlan.scheme + "://" + value);
> >} catch (NumberFormatException e) {
> >if (value.toString().contains(":"))
> >return new URI(value.toString());
> >else
> >return new URI(scheme, value.toString(), null);
> >}
> >} catch (URISyntaxException e) {
> >throw new CloudRuntimeException(
> >"Unable to convert to broadcast URI: " + value);
> >}
> >}
> >
> > would be the correct code for the function that is your culprit, but
> master
> > doesn't work due to database mismatches between the code and the databse
> > schema. I will test this as soon as possible.
> >
> > regards,
> > Daan
> >
> >
> > On Mon, Jul 22, 2013 at 5:18 PM, Koushik Das 
> wrote:
> >
> >> Standard system VMs with systemvm.iso from latest master. But that
> >> shouldn't matter as the issue is in the deploy VM code.
> >>
> >> 
> >> From: Donal Lafferty
> >> Sent: 22/07/2013 6:37 PM
> >> To: 'dev@cloudstack.apache.org'; cloudstack-...@incubator.apache.org
> >> Subject: RE: System VMs not coming up due to
> >> https://reviews.apache.org/r/12685/
> >>
> >> Where are you getting your system VM from?
> >>
> >> DL
> >>
> >>> -Original Message-
> >>> From: Koushik Das [mailto:koushik@citrix.com]
> >>> Sent: 22 July 2013 11:09 AM
> >>> To: cloudstack-...@incubator.apache.org
> >>> Subject: System VMs not coming up due to
> >>> https://reviews.apache.org/r/12685/
> >>>
> >>> Commit id: 2d4464d2badc9aff842fd180bafc4c384a83a91d
> >>>
> >>> -return new URI(scheme + "://" + value);
> >>> +// do we need to check that value does not contain a
> >> scheme
> >>> +// part?
> >>> +if (value.toString().contains(":"))
> >>> +return new URI(value.toString());
> >>> +else
> >>> +return new URI(scheme, value.toString(), null);
> >>>
> >>>
> >>> As part of this commit system VMs are not working in latest master. I
> >> see as
> >>> part of this change the broadcast/isolation URI format has changed from
> >>> vlan:// to vlan:. The code in many places still depends
> >> on
> >>> the URI.getHosts() to extract the vlan id which is broken due to the
> new
> >>> format.
> >>> Can this be reverted?
> >>>
> >>>
> >>>
> >>
> >>
>
>

Re: Review Request 12849: added backwards compatibility code to Networks enums

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12849/#review23683
---


Review 12849 PASSED the build test
The url of build cloudstack-master-with-patch #20 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/20/

- Jenkins Cloudstack.org


On July 23, 2013, 10:02 a.m., daan Hoogland wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12849/
> ---
> 
> (Updated July 23, 2013, 10:02 a.m.)
> 
> 
> Review request for cloudstack and Koushik Das.
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Both BroadcastDomainType and IsolationType needed some extra code for 
> backwards compatibility
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java c76c3d4 
>   
> server/src/com/cloud/network/router/VpcVirtualNetworkApplianceManagerImpl.java
>  6fafa3e 
> 
> Diff: https://reviews.apache.org/r/12849/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> daan Hoogland
> 
>

Review Request 12853: CLOUDSTACK-3688: Test cases test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the domain to delete and failed.

[Sanjay Tripathi]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12853/
---

Review request for cloudstack, Devdeep Singh and Prasanna Santhanam.


Bugs: CLOUDSTACK-3688


Repository: cloudstack-git


Description
---


CLOUDSTACK-3688: Test cases 
test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the 
domain to delete and failed.

This issue is in CS product and not in the test_script.
The problem is coming because in case of deleteDomain with cleanup = true,
CS is not allowing deletion of domain if there are any account under that
domain needs clean up; though these accounts are removed and admin can't
see them in the listaccounts.
So CS should not restrict the deleteDomain in case of accounts needs 
cleanup.


Diffs
-

  server/src/com/cloud/user/DomainManagerImpl.java 1117ff0 
  test/integration/component/test_accounts.py 3c284bd 

Diff: https://reviews.apache.org/r/12853/diff/


Testing
---

Verified marvin test on my local cloudstack setup.


Thanks,

Sanjay Tripathi

Review Request 12852: CLOUDSTACK-2508: Template usage should report both physical and virtual sizes

[Saksham Srivastava]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12852/
---

Review request for cloudstack and Kishan Kavala.


Bugs: 2508


Repository: cloudstack-git


Description
---

Usage entries for template and iso should include virtual_size parameter.
Added new column to usage_event table which contains the virtual size.
Modified the usage event creation when a template is created.


Diffs
-

  api/src/org/apache/cloudstack/api/response/UsageRecordResponse.java 4b355cb 
  api/src/org/apache/cloudstack/usage/Usage.java c74d3b7 
  engine/schema/src/com/cloud/event/UsageEventVO.java c92972b 
  engine/schema/src/com/cloud/event/dao/UsageEventDaoImpl.java 004ab7c 
  engine/schema/src/com/cloud/usage/UsageStorageVO.java 1337edc 
  engine/schema/src/com/cloud/usage/UsageVO.java 18a3a6b 
  engine/schema/src/com/cloud/usage/dao/UsageDaoImpl.java 2237d56 
  engine/schema/src/com/cloud/usage/dao/UsageStorageDaoImpl.java 297c8f4 
  server/src/com/cloud/api/ApiResponseHelper.java e058105 
  server/src/com/cloud/event/UsageEventUtils.java 2d92fae 
  server/src/com/cloud/template/TemplateManagerImpl.java 5005ee0 
  usage/src/com/cloud/usage/UsageManagerImpl.java 4901dd9 
  usage/src/com/cloud/usage/parser/StorageUsageParser.java 337e8e2 

Diff: https://reviews.apache.org/r/12852/diff/


Testing
---

Usage event generated for creating template with virtual size.


Thanks,

Saksham Srivastava

Re: Review Request 12808: Fix for CLOUDSTACK-2536: updating parameters (cpu/memory)overcommit ratio and (cpu/memory).overprosioning.factor are redundant(cluster level)

[Harikrishna Patnala]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12808/
---

(Updated July 23, 2013, 11:41 a.m.)


Review request for cloudstack, Koushik Das and Nitin Mehta.


Changes
---

Removing parameters from both AddCluster and UpdateCluster APIs.
While creating new cluster by default (cpu/memory).overprosioning.factor values 
are taken from global configuration parameters


Bugs: CLOUDSTACK-1528 and CLOUDSTACK-2536


Repository: cloudstack-git


Description
---

CLOUDSTACK-2536:  updating parameters (cpu/memory)overcommit ratio and 
(cpu/memory).overprosioning.factor are redundant(cluster level) 
There are two APIs updateCluster and UpdateConfiguration where we can update 
these parameters. Ideally this should get updated using updateConfiguration 
API, so removing from updateCluster API


Diffs (updated)
-

  api/src/com/cloud/resource/ResourceService.java 25298cc 
  api/src/org/apache/cloudstack/api/ApiConstants.java 0550a39 
  api/src/org/apache/cloudstack/api/command/admin/cluster/AddClusterCmd.java 
c6ca9bc6 
  api/src/org/apache/cloudstack/api/command/admin/cluster/UpdateClusterCmd.java 
a14f905 
  server/src/com/cloud/configuration/ConfigurationManagerImpl.java 92178c9 
  server/src/com/cloud/resource/ResourceManagerImpl.java d7652f4 
  server/test/com/cloud/resource/MockResourceManagerImpl.java 2c66134 

Diff: https://reviews.apache.org/r/12808/diff/


Testing
---


Thanks,

Harikrishna Patnala

Re: Review Request 12852: CLOUDSTACK-2508: Template usage should report both physical and virtual sizes

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12852/#review23684
---


Review 12852 PASSED the build test
The url of build cloudstack-master-with-patch #21 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/21/

- Jenkins Cloudstack.org


On July 23, 2013, 11:18 a.m., Saksham Srivastava wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12852/
> ---
> 
> (Updated July 23, 2013, 11:18 a.m.)
> 
> 
> Review request for cloudstack and Kishan Kavala.
> 
> 
> Bugs: 2508
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Usage entries for template and iso should include virtual_size parameter.
> Added new column to usage_event table which contains the virtual size.
> Modified the usage event creation when a template is created.
> 
> 
> Diffs
> -
> 
>   api/src/org/apache/cloudstack/api/response/UsageRecordResponse.java 4b355cb 
>   api/src/org/apache/cloudstack/usage/Usage.java c74d3b7 
>   engine/schema/src/com/cloud/event/UsageEventVO.java c92972b 
>   engine/schema/src/com/cloud/event/dao/UsageEventDaoImpl.java 004ab7c 
>   engine/schema/src/com/cloud/usage/UsageStorageVO.java 1337edc 
>   engine/schema/src/com/cloud/usage/UsageVO.java 18a3a6b 
>   engine/schema/src/com/cloud/usage/dao/UsageDaoImpl.java 2237d56 
>   engine/schema/src/com/cloud/usage/dao/UsageStorageDaoImpl.java 297c8f4 
>   server/src/com/cloud/api/ApiResponseHelper.java e058105 
>   server/src/com/cloud/event/UsageEventUtils.java 2d92fae 
>   server/src/com/cloud/template/TemplateManagerImpl.java 5005ee0 
>   usage/src/com/cloud/usage/UsageManagerImpl.java 4901dd9 
>   usage/src/com/cloud/usage/parser/StorageUsageParser.java 337e8e2 
> 
> Diff: https://reviews.apache.org/r/12852/diff/
> 
> 
> Testing
> ---
> 
> Usage event generated for creating template with virtual size.
> 
> 
> Thanks,
> 
> Saksham Srivastava
> 
>

Re: [Discuss] Apply rules on Virtual Router

[Nguyen Anh Tu]

Anyone?


2013/7/22 Nguyen Anh Tu 

> Hi guys,
>
> While working with L3 network services, I found a problem in the process
> of applying iptables rules. It currently works not good in my opinion. When
> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router backups
> old rules and re-apply all of non-revoked rules related to source IP on the
> new rule, including this one. It causes a slow, especially when you have a
> lot of running rules. When you delete a rule, the process happens in the
> same. The deleting rule is marked as "revoked", so it doesn't appear in the
> list. I think we should have a better approach.
>
> Any idea?
>
> --
>
> N.g.U.y.e.N.A.n.H.t.U
>



-- 

N.g.U.y.e.N.A.n.H.t.U

Re: [Discuss] Apply rules on Virtual Router

[Prasanna Santhanam]

On Mon, Jul 22, 2013 at 08:52:34PM +0700, Nguyen Anh Tu wrote:
> 
> While working with L3 network services, I found a problem in the process of
> applying iptables rules. It currently works not good in my opinion. When
> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router backups
> old rules and re-apply all of non-revoked rules related to source IP on the
> new rule, including this one. It causes a slow, especially when you have a
> lot of running rules. When you delete a rule, the process happens in the
> same. The deleting rule is marked as "revoked", so it doesn't appear in the
> list. I think we should have a better approach.
> 

I'd like this: but know not how java can do this or what is involved:

Do you propose a diff based solution where we incrementally apply the
new set of rules? Also, do you think this can be done for any
configuration within the VR not just iptable rules?


-- 
Prasanna.,


Powered by BigRock.com

Re: [Discuss] Apply rules on Virtual Router

[Chiradeep Vittal]

It is quite hard to do a delta update correctly, so a complete rewrite of
the ruleset is the safest way to do it. Not sure why it is "slow", but I'd
compare it to the time taken to start a VM.
One way to make it slightly faster is to write the ruleset to a file and
use iptables-restore from the file.

On 7/23/13 5:22 PM, "Nguyen Anh Tu"  wrote:

>Anyone?
>
>
>2013/7/22 Nguyen Anh Tu 
>
>> Hi guys,
>>
>> While working with L3 network services, I found a problem in the process
>> of applying iptables rules. It currently works not good in my opinion.
>>When
>> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router
>>backups
>> old rules and re-apply all of non-revoked rules related to source IP on
>>the
>> new rule, including this one. It causes a slow, especially when you
>>have a
>> lot of running rules. When you delete a rule, the process happens in the
>> same. The deleting rule is marked as "revoked", so it doesn't appear in
>>the
>> list. I think we should have a better approach.
>>
>> Any idea?
>>
>> --
>>
>> N.g.U.y.e.N.A.n.H.t.U
>>
>
>
>
>-- 
>
>N.g.U.y.e.N.A.n.H.t.U

Re: Review Request 12853: CLOUDSTACK-3688: Test cases test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the domain to delete and failed.

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12853/#review23685
---


Review 12853 PASSED the build test
The url of build cloudstack-master-with-patch #22 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/22/

- Jenkins Cloudstack.org


On July 23, 2013, 11:06 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12853/
> ---
> 
> (Updated July 23, 2013, 11:06 a.m.)
> 
> 
> Review request for cloudstack, Devdeep Singh and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3688
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> 
> CLOUDSTACK-3688: Test cases 
> test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the 
> domain to delete and failed.
> 
> This issue is in CS product and not in the test_script.
> The problem is coming because in case of deleteDomain with cleanup = true,
> CS is not allowing deletion of domain if there are any account under that
> domain needs clean up; though these accounts are removed and admin can't
> see them in the listaccounts.
> So CS should not restrict the deleteDomain in case of accounts needs 
> cleanup.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/user/DomainManagerImpl.java 1117ff0 
>   test/integration/component/test_accounts.py 3c284bd 
> 
> Diff: https://reviews.apache.org/r/12853/diff/
> 
> 
> Testing
> ---
> 
> Verified marvin test on my local cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12849: added backwards compatibility code to Networks enums

[Koushik Das]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12849/#review23686
---



api/src/com/cloud/network/Networks.java


Why only Vlan in case of IsolationType and Vlan, Storage and Mido in the 
case of BroadcastDomainType? Earlier both used similar logic.

If all system VMs (SSVM, CPVM and router VMs) are getting started after 
this then should be fine.

Also ensure that upgraded setups work fine as well.


- Koushik Das


On July 23, 2013, 10:02 a.m., daan Hoogland wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12849/
> ---
> 
> (Updated July 23, 2013, 10:02 a.m.)
> 
> 
> Review request for cloudstack and Koushik Das.
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Both BroadcastDomainType and IsolationType needed some extra code for 
> backwards compatibility
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java c76c3d4 
>   
> server/src/com/cloud/network/router/VpcVirtualNetworkApplianceManagerImpl.java
>  6fafa3e 
> 
> Diff: https://reviews.apache.org/r/12849/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> daan Hoogland
> 
>

Re: Review Request 12817: CLOUDSTACK-3689: Test case test_accounts.TesttemplateHierarchy.test_01_template_hierarchy failed during ListTemplate call

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12817/#review23687
---


Commit 3e6e66fee1bf73844e0ab7158e18f038aac7071d in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=3e6e66f ]

CLOUDSTACK-3689: Test case 
test_accounts.TesttemplateHierarchy.test_01_template_hierarchy failed during 
ListTemplate call.

Signed-off-by: Prasanna Santhanam 
(cherry picked from commit 6b2c9bf3ae1653693381431ed2554e7b7e4dfae4)


- ASF Subversion and Git Services


On July 22, 2013, 3:25 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12817/
> ---
> 
> (Updated July 22, 2013, 3:25 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3689
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3689: Test case 
> test_accounts.TesttemplateHierarchy.test_01_template_hierarchy  failed during 
> ListTemplate call.
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_accounts.py 7d0aeb1 
> 
> Diff: https://reviews.apache.org/r/12817/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on my cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12817: CLOUDSTACK-3689: Test case test_accounts.TesttemplateHierarchy.test_01_template_hierarchy failed during ListTemplate call

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12817/#review23688
---


Commit 6b2c9bf3ae1653693381431ed2554e7b7e4dfae4 in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=6b2c9bf ]

CLOUDSTACK-3689: Test case 
test_accounts.TesttemplateHierarchy.test_01_template_hierarchy failed during 
ListTemplate call.

Signed-off-by: Prasanna Santhanam 


- ASF Subversion and Git Services


On July 22, 2013, 3:25 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12817/
> ---
> 
> (Updated July 22, 2013, 3:25 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3689
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3689: Test case 
> test_accounts.TesttemplateHierarchy.test_01_template_hierarchy  failed during 
> ListTemplate call.
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_accounts.py 7d0aeb1 
> 
> Diff: https://reviews.apache.org/r/12817/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on my cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12817: CLOUDSTACK-3689: Test case test_accounts.TesttemplateHierarchy.test_01_template_hierarchy failed during ListTemplate call

[Prasanna Santhanam]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12817/#review23689
---

Ship it!


Ship It!

- Prasanna Santhanam


On July 22, 2013, 3:25 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12817/
> ---
> 
> (Updated July 22, 2013, 3:25 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3689
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3689: Test case 
> test_accounts.TesttemplateHierarchy.test_01_template_hierarchy  failed during 
> ListTemplate call.
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_accounts.py 7d0aeb1 
> 
> Diff: https://reviews.apache.org/r/12817/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on my cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12853: CLOUDSTACK-3688: Test cases test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the domain to delete and failed.

[Prasanna Santhanam]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12853/
---

(Updated July 23, 2013, 12:34 p.m.)


Review request for cloudstack, Alena Prokharchyk, Devdeep Singh, and Prasanna 
Santhanam.


Changes
---

Change to test is fine. Not sure about the DomainManager change. Adding Alena 
as reviewer.


Bugs: CLOUDSTACK-3688


Repository: cloudstack-git


Description
---


CLOUDSTACK-3688: Test cases 
test_accounts.TestDomainForceRemove.test_forceDeleteDomain, not found the 
domain to delete and failed.

This issue is in CS product and not in the test_script.
The problem is coming because in case of deleteDomain with cleanup = true,
CS is not allowing deletion of domain if there are any account under that
domain needs clean up; though these accounts are removed and admin can't
see them in the listaccounts.
So CS should not restrict the deleteDomain in case of accounts needs 
cleanup.


Diffs
-

  server/src/com/cloud/user/DomainManagerImpl.java 1117ff0 
  test/integration/component/test_accounts.py 3c284bd 

Diff: https://reviews.apache.org/r/12853/diff/


Testing
---

Verified marvin test on my local cloudstack setup.


Thanks,

Sanjay Tripathi

Re: Review Request 12810: DnsMasqConfigurator need to be rewrite in bash script

[bharat kumar]

> On July 22, 2013, 10:23 p.m., Sheng Yang wrote:
> > patches/systemvm/debian/config/root/dnsmasq.sh, line 54
> > 
> >
> > You don't need so complicate way to find dns is enabled or not. Just 
> > check /var/cache/cloud/cmdline for entry " dns1", " dns2"(remember add 
> > leading space to avoid ip6dns)
> > 
> > And if cmdline contained "useextdns=true", then dns1/dns2 should be 
> > used directly in such case. You don't need to check eth0 ip is dns or not.
> > 
> > In fact I don't think you need to add dhcp-option=tag:xxx,6,xxx for dns 
> > server of every range you added. You should able to use the original 
> > dhcp-option=6,xxx to cover all the cases. It should by default cover 
> > everything.

Hi sheng,

in case if useextndns is not true we have to add the routerip corresponding to 
each subnet so we need to add the option like this dhcp-option=tag:xxx,6,xxx. 


- bharat


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12810/#review23649
---


On July 22, 2013, 12:54 p.m., bharat kumar wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12810/
> ---
> 
> (Updated July 22, 2013, 12:54 p.m.)
> 
> 
> Review request for cloudstack, Alena Prokharchyk and Sheng Yang.
> 
> 
> Bugs: CLOUDSTACK-3694
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> https://issues.apache.org/jira/browse/CLOUDSTACK-3694
> 
> wrote the dnsmasq config in bash instead of creating the config file in java 
> and overwriting. 
> 
> 
> Diffs
> -
> 
>   core/src/com/cloud/agent/api/routing/DnsMasqConfigCommand.java 521ad70 
>   
> core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
> 0b26220 
>   core/src/com/cloud/network/DnsMasqConfigurator.java 3fc61df 
>   patches/systemvm/debian/config/root/createIpAlias.sh 5498195 
>   patches/systemvm/debian/config/root/dnsmasq.sh b70e2d3 
>   
> plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/resource/VmwareResource.java
>  c7f487e 
>   
> plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
>  f80d4b6 
>   scripts/vm/hypervisor/xenserver/vmops f8c0253 
>   server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java 
> 6c0f7a1 
> 
> Diff: https://reviews.apache.org/r/12810/diff/
> 
> 
> Testing
> ---
> 
> Tested on old master using xenserver.
>  Could not test on the latest one as it is broken.
> 
> 
> Thanks,
> 
> bharat kumar
> 
>

Re: Review Request 12810: DnsMasqConfigurator need to be rewrite in bash script

[bharat kumar]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12810/
---

(Updated July 23, 2013, 12:42 p.m.)


Review request for cloudstack, Alena Prokharchyk and Sheng Yang.


Changes
---

Incorporated the review comments.


Bugs: CLOUDSTACK-3694


Repository: cloudstack-git


Description
---

https://issues.apache.org/jira/browse/CLOUDSTACK-3694

wrote the dnsmasq config in bash instead of creating the config file in java 
and overwriting. 


Diffs (updated)
-

  build/replace.properties 265f335 
  core/src/com/cloud/agent/api/routing/DnsMasqConfigCommand.java 521ad70 
  core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
0b26220 
  core/src/com/cloud/network/DnsMasqConfigurator.java 3fc61df 
  patches/systemvm/debian/config/root/createIpAlias.sh 5498195 
  patches/systemvm/debian/config/root/dnsmasq.sh b70e2d3 
  
plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/resource/VmwareResource.java
 c7f487e 
  
plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
 f80d4b6 
  scripts/vm/hypervisor/xenserver/vmops f8c0253 
  server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java 
6c0f7a1 

Diff: https://reviews.apache.org/r/12810/diff/


Testing
---

Tested on old master using xenserver.
 Could not test on the latest one as it is broken.


Thanks,

bharat kumar

Review Request 12854: CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on restart of managment server.

[Sanjay Tripathi]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12854/
---

Review request for cloudstack and Devdeep Singh.


Bugs: CLOUDSTACK-3727


Repository: cloudstack-git


Description
---

CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
restart of managment server.

The method is listing all the accounts/domains including the removed ones which 
causes unnecessary insertion of the entries in the resource_count table.


Diffs
-

  server/src/com/cloud/server/ConfigurationServerImpl.java 9a9b145 

Diff: https://reviews.apache.org/r/12854/diff/


Testing
---

Verified that during the startup/restart of MS, resourcetype insertion for 
removed accounts is not happening.


Thanks,

Sanjay Tripathi

Re: Review Request 12810: DnsMasqConfigurator need to be rewrite in bash script

[bharat kumar]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12810/
---

(Updated July 23, 2013, 1:07 p.m.)


Review request for cloudstack, Alena Prokharchyk and Sheng Yang.


Changes
---

added the remove multiple_ranges.conf statement in cloud-early-config.


Bugs: CLOUDSTACK-3694


Repository: cloudstack-git


Description
---

https://issues.apache.org/jira/browse/CLOUDSTACK-3694

wrote the dnsmasq config in bash instead of creating the config file in java 
and overwriting. 


Diffs (updated)
-

  core/src/com/cloud/agent/api/routing/DnsMasqConfigCommand.java 521ad70 
  core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
0b26220 
  core/src/com/cloud/network/DnsMasqConfigurator.java 3fc61df 
  patches/systemvm/debian/config/etc/init.d/cloud-early-config c04ff90 
  patches/systemvm/debian/config/root/dnsmasq.sh b70e2d3 
  
plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/resource/VmwareResource.java
 c7f487e 
  
plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
 f80d4b6 
  scripts/vm/hypervisor/xenserver/vmops f8c0253 
  server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java 
6c0f7a1 

Diff: https://reviews.apache.org/r/12810/diff/


Testing
---

Tested on old master using xenserver.
 Could not test on the latest one as it is broken.


Thanks,

bharat kumar

Review Request 12856: [GSoC] Add midterm review

[Ian Duffy]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12856/
---

Review request for cloudstack, Abhinandan Prateek and Sebastien Goasguen.


Repository: cloudstack-git


Description
---

Add documentation for the midterm review.
This is an overview of work completed to date.


Diffs
-

  docs/en-US/gsoc-midsummer-ian.xml 348418f 
  docs/en-US/gsoc-midsummer.xml ffb031a 
  docs/en-US/images/add-ldap-configuration-failure.png PRE-CREATION 
  docs/en-US/images/add-ldap-configuration.png PRE-CREATION 
  docs/en-US/images/delete-ldap-configuration-failure.png PRE-CREATION 
  docs/en-US/images/delete-ldap.png PRE-CREATION 
  docs/en-US/images/jenkins-pipeline.png PRE-CREATION 
  docs/en-US/images/ldap-account-addition.png PRE-CREATION 
  docs/en-US/images/ldap-configuration.png PRE-CREATION 
  docs/en-US/images/ldap-global-settings.png PRE-CREATION 
  docs/en-US/images/ldap-list-users.png PRE-CREATION 
  docs/en-US/images/list-ldap-configuration.png PRE-CREATION 

Diff: https://reviews.apache.org/r/12856/diff/


Testing
---

Complied with publican. 


Thanks,

Ian Duffy

RE: [Discuss] Apply rules on Virtual Router

[Alex Huang]

The file approach will definitely make it faster.  

Just thinking out loud, If we can write all of the rules on a file, why not do 
an iptables-save, perform a diff and apply the difference?  

--Alex

> -Original Message-
> From: Chiradeep Vittal [mailto:chiradeep.vit...@citrix.com]
> Sent: Tuesday, July 23, 2013 5:08 AM
> To: dev@cloudstack.apache.org
> Cc: Nguyen Anh Tu
> Subject: Re: [Discuss] Apply rules on Virtual Router
> 
> It is quite hard to do a delta update correctly, so a complete rewrite of the
> ruleset is the safest way to do it. Not sure why it is "slow", but I'd 
> compare it
> to the time taken to start a VM.
> One way to make it slightly faster is to write the ruleset to a file and use
> iptables-restore from the file.
> 
> On 7/23/13 5:22 PM, "Nguyen Anh Tu"  wrote:
> 
> >Anyone?
> >
> >
> >2013/7/22 Nguyen Anh Tu 
> >
> >> Hi guys,
> >>
> >> While working with L3 network services, I found a problem in the
> >>process  of applying iptables rules. It currently works not good in my
> opinion.
> >>When
> >> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router
> >>backups  old rules and re-apply all of non-revoked rules related to
> >>source IP on the  new rule, including this one. It causes a slow,
> >>especially when you have a  lot of running rules. When you delete a
> >>rule, the process happens in the  same. The deleting rule is marked as
> >>"revoked", so it doesn't appear in the  list. I think we should have a
> >>better approach.
> >>
> >> Any idea?
> >>
> >> --
> >>
> >> N.g.U.y.e.N.A.n.H.t.U
> >>
> >
> >
> >
> >--
> >
> >N.g.U.y.e.N.A.n.H.t.U

Re: [DESIGN] Why is enum a class...

[John Burwell]

All,

+1 to Alex's design suggestion.  Another little know feature of enumerations is 
the ability to define abstract methods.  Therefore, Alex's example can be 
expanded as follows:

public enum Type {
   User(false)
{
@Override
public void doWork() {
}
},
   DomainRouter(true)
{
@Override
public void doWork() {
}
},
   ConsoleProxy(true)
{
@Override
public void doWork(){
{
},
   SecondaryStorageVm(true)
{
@Override
public void doWork(){
{
},
   ElasticIpVm(true)
{
@Override
public void doWork(){
{
},
   ElasticLoadBalancerVm(true)
{
@Override
public void doWork(){
{
},
   InternalLoadBalancerVm(true)
{
@Override
public void doWork(){
{
}

   /*
* UserBareMetal is only used for selecting VirtualMachineGuru, there is 
no
* VM with this type. UserBareMetal should treat exactly as User.
*/
   UserBareMetal(false);

  private boolean _isSystemVm;

  private Type(Boolean isSystemVm) {
  _isSystemVm = isSystemVm;
  }

   public boolean isSystemVM() {
  return _isSystemVm;
   }

  public abstract doWork();

   }

Personally, I see using switch statements with Java enumerations as a bad code 
smell.  Java enumerations were specifically designed to allow the 
implementation of behavior on values.  In addition to simplifying code, it 
greatly increases cohesion by defining all of the uses of the value in one 
place and using the compiler to validate coverage.  Finally, it makes unit 
testing of this conditional functionality much more straight forward.

To Frank's point, I don't see the need for another class.  The Java compiler 
properly generates the boilerplate methods and ensures that all enumeration 
values are singletons within a class loader.  Regardless of the state added to 
an enumeration value, User==User will always be true.

Thanks,
-John

On Jul 23, 2013, at 3:48 AM, Daan Hoogland  wrote:

> +1 I am having a war on the enums in Networks.java that justifies your
> solution, Frank
> 
> 
> On Tue, Jul 23, 2013 at 2:57 AM, Alex Huang  wrote:
> 
>> +1 to do this for any enum that was designed to be added to by the plugins.
>> 
>> I actually wrote a class for this before called Constant that did exactly
>> this.  The problem I had was that there was no way to guarantee compile
>> time enumeration of all of the values like Enum did so it's possible for
>> errors to be introduced but certainly we can follow a convention in this
>> regard.  Another way would be to use Spring to gather them and inject them
>> into a holding class.
>> 
>> --Alex
>> 
>>> -Original Message-
>>> From: Frank Zhang [mailto:frank.zh...@citrix.com]
>>> Sent: Monday, July 22, 2013 5:45 PM
>>> To: dev@cloudstack.apache.org
>>> Subject: RE: [DESIGN] Why is enum a class...
>>> 
>>> Frankly speaking, if we are going to change enum, I would suggest not
>> using
>>> enmu anymore, instead, defining our own class like:
>>> 
>>> public class VmType {
>>>  private static Map types =
>>> Collections.synchronizedMap(new HashMap());
>>>  private final String typeName;
>>> 
>>>  public VmType (String typeName) {
>>>  this.typeName = typeName;
>>>  types.put(typeName, this);
>>>  }
>>> 
>>>  public static VmType valueOf(String typeName) {
>>>  VmType type = types.get(typeName);
>>>  if (type == null) {
>>>  throw new IllegalArgumentException("VmType type:
>>> " + typeName + " was not registered by any VmType");
>>>  }
>>>  return type;
>>>  }
>>> 
>>>  @Override
>>>  public String toString() {
>>>  return typeName;
>>>  }
>>> 
>>>  @Override
>>>  public boolean equals(Object t) {
>>>  if (t == null || !(t instanceof VmType)) {
>>>  return false;
>>>  }
>>> 
>>>  VmType type = (VmType)t;
>>>  return type.toString().equals(typeName);
>>>  }
>>> 
>>>  @Override
>>>  public int hashCode() {
>>>  return typeName.hashCode();
>>>  }
>>> 
>>>  public static Set getAllTypeNames() {
>>>  return types.keySet();
>>>  }
>>> }
>>> 
>>> The only benefit of enum I see is you can use "==" instead of "equals()"
>>> when comparing variables. However, it makes your code tight, every time
>>> adding a new type you have to modify the enum.
>>> 
>>> By above method, when a new plugin wants to extend a new VmType, it
>>> simply does:
>>> 
>>> class MagicVmManagerImpl {
>>>  

Re: Review Request 12854: CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on restart of managment server.

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12854/#review23691
---


Review 12854 PASSED the build test
The url of build cloudstack-master-with-patch #23 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/23/

- Jenkins Cloudstack.org


On July 23, 2013, 12:54 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12854/
> ---
> 
> (Updated July 23, 2013, 12:54 p.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3727
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
> restart of managment server.
> 
> The method is listing all the accounts/domains including the removed ones 
> which causes unnecessary insertion of the entries in the resource_count table.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/server/ConfigurationServerImpl.java 9a9b145 
> 
> Diff: https://reviews.apache.org/r/12854/diff/
> 
> 
> Testing
> ---
> 
> Verified that during the startup/restart of MS, resourcetype insertion for 
> removed accounts is not happening.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Reformatting UI code

[Sebastien Goasguen]

On Jul 22, 2013, at 6:54 PM, Jessica Wang  wrote:

> Sebastien,
> 
> Brian and I are fix 4.2 bugs.
> Any fix for 4.2 bugs should go to both 4.2 branch and master branch.
> That's why Brian and I still need to merge our check-in between master branch 
> and 4.2 branch.

Ok got it .

Ian next time you send a patch related to the UI maybe you can set Jessica and 
Brian as reviewers.

thanks,

> 
> Jessica
> 
> -Original Message-
> From: Sebastien Goasguen [mailto:run...@gmail.com] 
> Sent: Monday, July 22, 2013 3:53 PM
> To: Brian Federle
> Cc: Jessica Wang; Pranav Saxena; Ian Duffy (i...@ianduffy.ie); 
> dev@cloudstack.apache.org
> Subject: Re: Reformatting UI code
> 
> 
> 
> On 23 Jul 2013, at 00:34, Brian Federle  wrote:
> 
>> In this case it was fine, I could resolve the conflicts by git's 
>> -Xignore-space-change and reformatting accordingly.
>> 
>> The main issue was that I didn't see any indication that this was being 
>> committed from reading the discussion thread. Since we're doing a lot of 
>> bugfixing right now for 4.2, a large sweeping commit like this should at 
>> least be posted as part of thread, so that there is heads-up so that we can 
>> prepare any pending changes we have to avoid being blocked.
>> 
> 
> Ok but this is not for 4.2. It was committed to master. How is it impacting 
> your 4.2 fixes ?
> 
> 
>> -Brian
>> 
>> -Original Message-
>> From: Sebastien Goasguen [mailto:run...@gmail.com] 
>> Sent: Monday, July 22, 2013 3:26 PM
>> To: Jessica Wang
>> Cc: Pranav Saxena; Ian Duffy (i...@ianduffy.ie); dev@cloudstack.apache.org; 
>> Brian Federle
>> Subject: Re: Reformatting UI code
>> 
>> Hi jessica,
>> 
>> Sorry this is causing you trouble. Ian us developing his ldap plugin in a 
>> feature branch .
>> This was a UI cosmetic code change, i saw that pranav shipped it in RB but 
>> did not apply the patch. Since the three if us work in more or less the same 
>> time zone i applied the patch quickly to master. I did think about other UI 
>> work but since 4.2 is in feature freeze i did not think committing to master 
>> would be a problem.
>> 
>> Moreover i did not know u were working on UI feature branches. Where can we 
>> look at the feature descriptions ? 
>> 
>> I suppose u could revert the patch . But for features for 4.3 i think it 
>> should be rebase of your feature branches . I dont think this change should 
>> be impacting 4.2 not sure why u say thats the case.
>> 
>> Thoughts ?
>> 
>> -Sebastien
>> 
>> On 23 Jul 2013, at 00:07, Jessica Wang  wrote:
>> 
>>> Pranav, Ian, Sebastien,
>>> 
>>> The problem is the duration between the time Ian brought it up on the 
>>> mailing list and the time Ian/Sebastien checked in the change to master 
>>> branch is too short (less than 4 hours).
>>> 
>>> Ian brought it up on the mailing list at Thu 7/18/2013 5:44 AM (email 
>>> subject is "Auto format javascript"). 
>>> Sebastien checked in Ian's change to master branch at 7/18/2013 9:34 AM 
>>> (Commit hash: ad69bc8da3244b783dd003ddf3184fca2762c514).
>>> 
>>> This is a big change of UI code.
>>> In GIT's view, every line in JS files has been changed (If you look at code 
>>> difference in GIT's history).
>>> GIT sees it as "delete all lines and add new different lines".
>>> I was unable to merge my check-in from master branch to 4.2 branch (or any 
>>> other branch) since GIT sees JS files in master branch and other branches 
>>> are totally different.
>>> 
>>> Shouldn't this kind of big change be checked in to a different branch (not 
>>> master branch) first? Then, submit a merge request to community, wait for 
>>> 72 hours, then merge to master branch eventually?
>>> 
>>> Jessica
>>> 
>>> 
>>> -Original Message-
>>> From: Pranav Saxena [mailto:psb...@gmail.com]
>>> Sent: Monday, July 22, 2013 12:18 PM
>>> To: dev@cloudstack.apache.org; Brian Federle
>>> Cc: Ian Duffy (i...@ianduffy.ie)
>>> Subject: Re: Reformatting UI code
>>> 
>>> Hey Brian,
>>> 
>>> Sorry to hear that it caused merge conflicts for you . But Ian did 
>>> bring it up on the mailing list and I suggested him to use the js 
>>> beautifier tool for reformatting the js code which I guess you missed 
>>> probably because of the "heavy" traffic on the dev list and thereafter 
>>> Sebastien merged the code when the discussion and the reviews ended. 
>>> Anyways, I'll also try to ping you personally on such occasions in the 
>>> future to let you know if any major changes are being committed. Now , 
>>> probably you would need to do a lot of rebasing , sorry for that !!
>>> 
>>> Thanks,
>>> Pranav
>>> 
>>> 
>>> On Tue, Jul 23, 2013 at 12:29 AM, Jessica Wang 
>>> wrote:
>>> 
 +1
 
 -Original Message-
 From: Brian Federle [mailto:brian.fede...@citrix.com]
 Sent: Monday, July 22, 2013 11:33 AM
 To: dev@cloudstack.apache.org
 Cc: Ian Duffy (i...@ianduffy.ie)
 Subject: Reformatting UI code
 
 Hello,
 
 Recently I discovered that all JS and UI code have been

Review Request 12858: CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network offering.

[Sanjay Tripathi]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12858/
---

Review request for cloudstack and Prasanna Santhanam.


Bugs: CLOUDSTACK-3724


Repository: cloudstack-git


Description
---

CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network 
offering.

The test_script was broken because "nw_off" got appended twice for cleanup 
instead of "nw_off_no_lb".


Diffs
-

  test/integration/component/test_vpc_vms_deployment.py c599d96 

Diff: https://reviews.apache.org/r/12858/diff/


Testing
---

Verified the marvin test locally on cloudstack setup.


Thanks,

Sanjay Tripathi

Re: Review Request 12854: CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on restart of managment server.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12854/#review23694
---


Commit e2e3b9a2884d4f1029aa5e0bb5137160fd794b24 in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=e2e3b9a ]

CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
restart of managment server.

Signed-off-by: Prasanna Santhanam 
(cherry picked from commit a7084b2dac594a5851e32be692863fb9f7b23d9d)


- ASF Subversion and Git Services


On July 23, 2013, 12:54 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12854/
> ---
> 
> (Updated July 23, 2013, 12:54 p.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3727
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
> restart of managment server.
> 
> The method is listing all the accounts/domains including the removed ones 
> which causes unnecessary insertion of the entries in the resource_count table.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/server/ConfigurationServerImpl.java 9a9b145 
> 
> Diff: https://reviews.apache.org/r/12854/diff/
> 
> 
> Testing
> ---
> 
> Verified that during the startup/restart of MS, resourcetype insertion for 
> removed accounts is not happening.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12854: CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on restart of managment server.

[Prasanna Santhanam]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12854/#review23695
---

Ship it!


Ship It!

- Prasanna Santhanam


On July 23, 2013, 12:54 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12854/
> ---
> 
> (Updated July 23, 2013, 12:54 p.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3727
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
> restart of managment server.
> 
> The method is listing all the accounts/domains including the removed ones 
> which causes unnecessary insertion of the entries in the resource_count table.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/server/ConfigurationServerImpl.java 9a9b145 
> 
> Diff: https://reviews.apache.org/r/12854/diff/
> 
> 
> Testing
> ---
> 
> Verified that during the startup/restart of MS, resourcetype insertion for 
> removed accounts is not happening.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12854: CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on restart of managment server.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12854/#review23696
---


Commit a7084b2dac594a5851e32be692863fb9f7b23d9d in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=a7084b2 ]

CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
restart of managment server.

Signed-off-by: Prasanna Santhanam 


- ASF Subversion and Git Services


On July 23, 2013, 12:54 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12854/
> ---
> 
> (Updated July 23, 2013, 12:54 p.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3727
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3727: Resource Count calculation happens for deleted accounts on 
> restart of managment server.
> 
> The method is listing all the accounts/domains including the removed ones 
> which causes unnecessary insertion of the entries in the resource_count table.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/server/ConfigurationServerImpl.java 9a9b145 
> 
> Diff: https://reviews.apache.org/r/12854/diff/
> 
> 
> Testing
> ---
> 
> Verified that during the startup/restart of MS, resourcetype insertion for 
> removed accounts is not happening.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Fwd: [GSOC] Update to document

[Sebastien Goasguen]

Hi Guys (Ian, Dharmesh, Shiva, Meng, Nguyen),

Google needs to get our evaluation of your progress so far to decide if you 
keep on going or not.

I sent the email below on July 5th. You now have only few days left to get it 
done ( I need it before Friday).

I am looking for 3/4 pages describing what you have done in the last month and 
half, describe your progress, your challenges, what you learn about OSS 
development and how you see your deliverables going forward.

-sebastien

Begin forwarded message:

> From: Sebastien Goasguen 
> Subject: [GSOC] Update to document
> Date: July 5, 2013 9:18:07 AM EDT
> To: "dev@cloudstack.apache.org" 
> 
> Hi guys,
> 
> I modified the format of our "report", I moved the proposals to a chapter and 
> started a new chapter called "updates".
> Each of you should use a section to write a small report/updates on your 
> progress so far.
> 
> You can see the doc at:
> http://jenkins.cloudstack.org/job/docs-4.3-gsoc-guide/lastSuccessfulBuild/artifact/Apache_CloudStack-4.2.0-CloudStack_GSoC_Guide-en-US.pdf
> 
> Your respective sections should be self-explanatory when you checkout the 
> code.
> 
> We will use this for the mid-summer evaluation that Google wants/needs.
> 
> A good example of progress report is from Ian at:
> 
> http://imduffy15.blogspot.co.uk
> 
> You can include specific things about your project but also things related to 
> CloudStack itself, what you are learning (jira, git etc..) and what you 
> contributed so far outside your project (e.g Meng gave us the quick install 
> guide for 4.1 and Dharmesh is looking at a significant update to the code 
> with CLLU)
> 
> Cheers,
> 
> -Sebastien

Review Request 12859: fixed account objects CLOUDSTACK-3726

[Sowmya Krishnan]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12859/
---

Review request for cloudstack and Prasanna Santhanam.


Bugs: CLOUDSTACK-3726


Repository: cloudstack-git


Description
---

Fixed account objects for all tests


Diffs
-

  test/integration/component/test_netscaler_nw_off.py c328f48 

Diff: https://reviews.apache.org/r/12859/diff/


Testing
---

Locally, script passes


Thanks,

Sowmya Krishnan

Re: Review Request 12859: fixed account objects CLOUDSTACK-3726

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12859/#review23697
---


Commit d2ce17ea73248ed92cda58df85d27c771dbfdfb3 in branch refs/heads/master 
from Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=d2ce17e ]

Fixed account objects CLOUDSTACK-3726

Signed-off-by: Prasanna Santhanam 
(cherry picked from commit 1cd435e25a36e8737d37edce1bb5da70d9b670c9)


- ASF Subversion and Git Services


On July 23, 2013, 3:11 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12859/
> ---
> 
> (Updated July 23, 2013, 3:11 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3726
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed account objects for all tests
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_netscaler_nw_off.py c328f48 
> 
> Diff: https://reviews.apache.org/r/12859/diff/
> 
> 
> Testing
> ---
> 
> Locally, script passes
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12859: fixed account objects CLOUDSTACK-3726

[Prasanna Santhanam]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12859/#review23700
---

Ship it!


Ship It!

- Prasanna Santhanam


On July 23, 2013, 3:11 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12859/
> ---
> 
> (Updated July 23, 2013, 3:11 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3726
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed account objects for all tests
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_netscaler_nw_off.py c328f48 
> 
> Diff: https://reviews.apache.org/r/12859/diff/
> 
> 
> Testing
> ---
> 
> Locally, script passes
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12858: CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network offering.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12858/#review23699
---


Commit a06b3f40eba6ebc3af5d7c366ec757703a085f67 in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=a06b3f4 ]

CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network 
offering.

Signed-off-by: Prasanna Santhanam 


- ASF Subversion and Git Services


On July 23, 2013, 2:53 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12858/
> ---
> 
> (Updated July 23, 2013, 2:53 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3724
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find 
> network offering.
> 
> The test_script was broken because "nw_off" got appended twice for cleanup 
> instead of "nw_off_no_lb".
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_vpc_vms_deployment.py c599d96 
> 
> Diff: https://reviews.apache.org/r/12858/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12858: CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network offering.

[Prasanna Santhanam]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12858/#review23701
---

Ship it!


Ship It!

- Prasanna Santhanam


On July 23, 2013, 2:53 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12858/
> ---
> 
> (Updated July 23, 2013, 2:53 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3724
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find 
> network offering.
> 
> The test_script was broken because "nw_off" got appended twice for cleanup 
> instead of "nw_off_no_lb".
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_vpc_vms_deployment.py c599d96 
> 
> Diff: https://reviews.apache.org/r/12858/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12859: fixed account objects CLOUDSTACK-3726

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12859/#review23698
---


Commit 1cd435e25a36e8737d37edce1bb5da70d9b670c9 in branch refs/heads/4.2 from 
Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=1cd435e ]

Fixed account objects CLOUDSTACK-3726

Signed-off-by: Prasanna Santhanam 


- ASF Subversion and Git Services


On July 23, 2013, 3:11 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12859/
> ---
> 
> (Updated July 23, 2013, 3:11 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3726
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed account objects for all tests
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_netscaler_nw_off.py c328f48 
> 
> Diff: https://reviews.apache.org/r/12859/diff/
> 
> 
> Testing
> ---
> 
> Locally, script passes
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12856: [GSoC] Add midterm review

[Sebastien Goasguen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12856/#review23702
---

Ship it!


applied to master 06484061f39f27dd3249572016e8a9adbf95851a
mark as submitted

thanks

- Sebastien Goasguen


On July 23, 2013, 1:18 p.m., Ian Duffy wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12856/
> ---
> 
> (Updated July 23, 2013, 1:18 p.m.)
> 
> 
> Review request for cloudstack, Abhinandan Prateek and Sebastien Goasguen.
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Add documentation for the midterm review.
> This is an overview of work completed to date.
> 
> 
> Diffs
> -
> 
>   docs/en-US/gsoc-midsummer-ian.xml 348418f 
>   docs/en-US/gsoc-midsummer.xml ffb031a 
>   docs/en-US/images/add-ldap-configuration-failure.png PRE-CREATION 
>   docs/en-US/images/add-ldap-configuration.png PRE-CREATION 
>   docs/en-US/images/delete-ldap-configuration-failure.png PRE-CREATION 
>   docs/en-US/images/delete-ldap.png PRE-CREATION 
>   docs/en-US/images/jenkins-pipeline.png PRE-CREATION 
>   docs/en-US/images/ldap-account-addition.png PRE-CREATION 
>   docs/en-US/images/ldap-configuration.png PRE-CREATION 
>   docs/en-US/images/ldap-global-settings.png PRE-CREATION 
>   docs/en-US/images/ldap-list-users.png PRE-CREATION 
>   docs/en-US/images/list-ldap-configuration.png PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/12856/diff/
> 
> 
> Testing
> ---
> 
> Complied with publican. 
> 
> 
> Thanks,
> 
> Ian Duffy
> 
>

Re: Reformatting UI code

[Mike Tutkowski]

One other thought here is that we might want to wait until, say, the day
after code freeze - if allowable - and then check in these kinds of changes
(basically give them special permission to go in after code freeze).

I think they're valuable since they do make the code much more readable,
but it might be better if they go in once development is "frozen."


On Tue, Jul 23, 2013 at 8:42 AM, Sebastien Goasguen wrote:

>
> On Jul 22, 2013, at 6:54 PM, Jessica Wang  wrote:
>
> > Sebastien,
> >
> > Brian and I are fix 4.2 bugs.
> > Any fix for 4.2 bugs should go to both 4.2 branch and master branch.
> > That's why Brian and I still need to merge our check-in between master
> branch and 4.2 branch.
>
> Ok got it .
>
> Ian next time you send a patch related to the UI maybe you can set Jessica
> and Brian as reviewers.
>
> thanks,
>
> >
> > Jessica
> >
> > -Original Message-
> > From: Sebastien Goasguen [mailto:run...@gmail.com]
> > Sent: Monday, July 22, 2013 3:53 PM
> > To: Brian Federle
> > Cc: Jessica Wang; Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> dev@cloudstack.apache.org
> > Subject: Re: Reformatting UI code
> >
> >
> >
> > On 23 Jul 2013, at 00:34, Brian Federle 
> wrote:
> >
> >> In this case it was fine, I could resolve the conflicts by git's
> -Xignore-space-change and reformatting accordingly.
> >>
> >> The main issue was that I didn't see any indication that this was being
> committed from reading the discussion thread. Since we're doing a lot of
> bugfixing right now for 4.2, a large sweeping commit like this should at
> least be posted as part of thread, so that there is heads-up so that we can
> prepare any pending changes we have to avoid being blocked.
> >>
> >
> > Ok but this is not for 4.2. It was committed to master. How is it
> impacting your 4.2 fixes ?
> >
> >
> >> -Brian
> >>
> >> -Original Message-
> >> From: Sebastien Goasguen [mailto:run...@gmail.com]
> >> Sent: Monday, July 22, 2013 3:26 PM
> >> To: Jessica Wang
> >> Cc: Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> dev@cloudstack.apache.org; Brian Federle
> >> Subject: Re: Reformatting UI code
> >>
> >> Hi jessica,
> >>
> >> Sorry this is causing you trouble. Ian us developing his ldap plugin in
> a feature branch .
> >> This was a UI cosmetic code change, i saw that pranav shipped it in RB
> but did not apply the patch. Since the three if us work in more or less the
> same time zone i applied the patch quickly to master. I did think about
> other UI work but since 4.2 is in feature freeze i did not think committing
> to master would be a problem.
> >>
> >> Moreover i did not know u were working on UI feature branches. Where
> can we look at the feature descriptions ?
> >>
> >> I suppose u could revert the patch . But for features for 4.3 i think
> it should be rebase of your feature branches . I dont think this change
> should be impacting 4.2 not sure why u say thats the case.
> >>
> >> Thoughts ?
> >>
> >> -Sebastien
> >>
> >> On 23 Jul 2013, at 00:07, Jessica Wang  wrote:
> >>
> >>> Pranav, Ian, Sebastien,
> >>>
> >>> The problem is the duration between the time Ian brought it up on the
> mailing list and the time Ian/Sebastien checked in the change to master
> branch is too short (less than 4 hours).
> >>>
> >>> Ian brought it up on the mailing list at Thu 7/18/2013 5:44 AM (email
> subject is "Auto format javascript").
> >>> Sebastien checked in Ian's change to master branch at 7/18/2013 9:34
> AM (Commit hash: ad69bc8da3244b783dd003ddf3184fca2762c514).
> >>>
> >>> This is a big change of UI code.
> >>> In GIT's view, every line in JS files has been changed (If you look at
> code difference in GIT's history).
> >>> GIT sees it as "delete all lines and add new different lines".
> >>> I was unable to merge my check-in from master branch to 4.2 branch (or
> any other branch) since GIT sees JS files in master branch and other
> branches are totally different.
> >>>
> >>> Shouldn't this kind of big change be checked in to a different branch
> (not master branch) first? Then, submit a merge request to community, wait
> for 72 hours, then merge to master branch eventually?
> >>>
> >>> Jessica
> >>>
> >>>
> >>> -Original Message-
> >>> From: Pranav Saxena [mailto:psb...@gmail.com]
> >>> Sent: Monday, July 22, 2013 12:18 PM
> >>> To: dev@cloudstack.apache.org; Brian Federle
> >>> Cc: Ian Duffy (i...@ianduffy.ie)
> >>> Subject: Re: Reformatting UI code
> >>>
> >>> Hey Brian,
> >>>
> >>> Sorry to hear that it caused merge conflicts for you . But Ian did
> >>> bring it up on the mailing list and I suggested him to use the js
> >>> beautifier tool for reformatting the js code which I guess you missed
> >>> probably because of the "heavy" traffic on the dev list and thereafter
> >>> Sebastien merged the code when the discussion and the reviews ended.
> >>> Anyways, I'll also try to ping you personally on such occasions in the
> >>> future to let you know if any major changes are being committed. Now ,
> >>> prob

Re: [DESIGN] Why is enum a class...

[Mike Tutkowski]

The first thing I noticed about Frank's comment is that he started it with
"Frankly speaking." (Frank/Frankly) Probably not intended as a pun, but
funny nonetheless. :)


On Mon, Jul 22, 2013 at 6:45 PM, Frank Zhang  wrote:

> Frankly speaking, if we are going to change enum, I would suggest not
> using enmu anymore, instead, defining our own class like:
>
> public class VmType {
> private static Map types =
> Collections.synchronizedMap(new HashMap());
> private final String typeName;
>
> public VmType (String typeName) {
> this.typeName = typeName;
> types.put(typeName, this);
> }
>
> public static VmType valueOf(String typeName) {
> VmType type = types.get(typeName);
> if (type == null) {
> throw new IllegalArgumentException("VmType type: "
> + typeName + " was not registered by any VmType");
> }
> return type;
> }
>
> @Override
> public String toString() {
> return typeName;
> }
>
> @Override
> public boolean equals(Object t) {
> if (t == null || !(t instanceof VmType)) {
> return false;
> }
>
> VmType type = (VmType)t;
> return type.toString().equals(typeName);
> }
>
> @Override
> public int hashCode() {
> return typeName.hashCode();
> }
>
> public static Set getAllTypeNames() {
> return types.keySet();
> }
> }
>
> The only benefit of enum I see is you can use "==" instead of "equals()"
> when comparing variables. However, it makes your code tight, every time
> adding a new type you have to modify the enum.
>
> By above method, when a new plugin wants to extend a new VmType, it simply
> does:
>
> class MagicVmManagerImpl {
> public static final VmType type = new VmType("MagicVm");
> }
>
>
> > -Original Message-
> > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > Sent: Monday, July 22, 2013 5:23 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [DESIGN] Why is enum a class...
> >
> > BTW, this code already shows a bug that stems from the static method
> usage.
> > It says ElasticIpVm is not a system vm (which I don't believe is true).
>  Probably
> > because whoever added elastic ip vm didn't see the static method and so
> didn't
> > add the vm into the method.
> >
> > --Alex
> >
> > > -Original Message-
> > > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > > Sent: Monday, July 22, 2013 5:14 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: [DESIGN] Why is enum a class...
> > >
> > > I just went over this code and thought it was related and might be
> > > interested to other developers.
> > >
> > > What's the difference between declaring a enum like this
> > >
> > > public enum Type {
> > > User,
> > > DomainRouter,
> > > ConsoleProxy,
> > > SecondaryStorageVm,
> > > ElasticIpVm,
> > > ElasticLoadBalancerVm,
> > > InternalLoadBalancerVm,
> > >
> > > /*
> > >  * UserBareMetal is only used for selecting
> > > VirtualMachineGuru, there is no
> > >  * VM with this type. UserBareMetal should treat exactly as
> User.
> > >  */
> > > UserBareMetal;
> > >
> > > public static boolean isSystemVM(VirtualMachine.Type vmtype) {
> > > if (DomainRouter.equals(vmtype)
> > > || ConsoleProxy.equals(vmtype)
> > > || SecondaryStorageVm.equals(vmtype) ||
> > > InternalLoadBalancerVm.equals(vmtype)) {
> > > return true;
> > > }
> > > return false;
> > > }
> > > }
> > >
> > > Vs
> > >
> > > public enum Type {
> > > User(false),
> > > DomainRouter(true),
> > > ConsoleProxy(true),
> > > SecondaryStorageVm(true),
> > > ElasticIpVm(true),
> > > ElasticLoadBalancerVm(true),
> > > InternalLoadBalancerVm(true),
> > >
> > > /*
> > >  * UserBareMetal is only used for selecting
> > > VirtualMachineGuru, there is no
> > >  * VM with this type. UserBareMetal should treat exactly as
> User.
> > >  */
> > > UserBareMetal(false);
> > >
> > >private boolean _isSystemVm;
> > >
> > >private Type(Boolean isSystemVm) {
> > >_isSystemVm = isSystemVm;
> > >}
> > >
> > > public boolean isSystemVM() {
> > >return _isSystemVm;
> > > }
> > > }
> > >
> > > The second declaration is more self-descriptive:
> > >
> > > - It tells developer when they add more to this enum, they have to
> > > specify whether it is a system vm.  They may have missed the static
> > > method in the first declaration and causes failures later.
> > > - It tells developers using the enu

[DISCUSS] API changes / compatibility between 4.1 and 4.2

[Wido den Hollander]

Hi,

Yesterday I encountered this bug: 
https://issues.apache.org/jira/browse/CLOUDSTACK-3714


A short summary: A 4.2 agent can't communicate with a 4.2 management 
server and vise versa.


What this means is that during an upgrade you can't upgrade the Agents 
one by one, since a upgraded Agent won't be able to participate in the 
cluster.


This is a big breaker for me, since going from 4.1 to 4.2 and having to 
shut down your management server and all your Agents at once seems like 
a major upgrade to me.


I also encountered the problems with the new SystemVMs today while 
upgrading my management server to 4.2 and it seems to me that this 
upgrade won't be an easy one for a lot of users.


So I'd like to start a discussing about this, since I'm almost thinking 
that 4.2 should become 5.0 because of the major changes.


These major changes are good and they bring a lot of functionality to 
CloudStack, but sometimes it seems like stuff is coming in so fast that 
it's almost impossible to keep up with.


How do other people think about this API change between 4.1 and 4.2?

Do you also think that 4.2 should actually be called 5.0?

Wido

Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[Sowmya Krishnan]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/
---

Review request for cloudstack and Prasanna Santhanam.


Bugs: CLOUDSTACK-3697


Repository: cloudstack-git


Description
---

LB is already deleted when project is cleaned up. Removed it from cleanup queue


Diffs
-

  test/integration/component/test_project_resources.py 378238e 

Diff: https://reviews.apache.org/r/12864/diff/


Testing
---

Tested script with fix locally and it passes now.


Thanks,

Sowmya Krishnan

Re: Review Request 12810: DnsMasqConfigurator need to be rewrite in bash script

[Sheng Yang]

> On July 22, 2013, 10:23 p.m., Sheng Yang wrote:
> > patches/systemvm/debian/config/root/dnsmasq.sh, line 54
> > 
> >
> > You don't need so complicate way to find dns is enabled or not. Just 
> > check /var/cache/cloud/cmdline for entry " dns1", " dns2"(remember add 
> > leading space to avoid ip6dns)
> > 
> > And if cmdline contained "useextdns=true", then dns1/dns2 should be 
> > used directly in such case. You don't need to check eth0 ip is dns or not.
> > 
> > In fact I don't think you need to add dhcp-option=tag:xxx,6,xxx for dns 
> > server of every range you added. You should able to use the original 
> > dhcp-option=6,xxx to cover all the cases. It should by default cover 
> > everything.
> 
> bharat kumar wrote:
> Hi sheng,
> 
> in case if useextndns is not true we have to add the routerip 
> corresponding to each subnet so we need to add the option like this 
> dhcp-option=tag:xxx,6,xxx. 
> 
>

Yes, you're right. dhcp-option=6,xxx only apply to useextdns.


- Sheng


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12810/#review23649
---


On July 23, 2013, 1:07 p.m., bharat kumar wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12810/
> ---
> 
> (Updated July 23, 2013, 1:07 p.m.)
> 
> 
> Review request for cloudstack, Alena Prokharchyk and Sheng Yang.
> 
> 
> Bugs: CLOUDSTACK-3694
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> https://issues.apache.org/jira/browse/CLOUDSTACK-3694
> 
> wrote the dnsmasq config in bash instead of creating the config file in java 
> and overwriting. 
> 
> 
> Diffs
> -
> 
>   core/src/com/cloud/agent/api/routing/DnsMasqConfigCommand.java 521ad70 
>   
> core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
> 0b26220 
>   core/src/com/cloud/network/DnsMasqConfigurator.java 3fc61df 
>   patches/systemvm/debian/config/etc/init.d/cloud-early-config c04ff90 
>   patches/systemvm/debian/config/root/dnsmasq.sh b70e2d3 
>   
> plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/resource/VmwareResource.java
>  c7f487e 
>   
> plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
>  f80d4b6 
>   scripts/vm/hypervisor/xenserver/vmops f8c0253 
>   server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java 
> 6c0f7a1 
> 
> Diff: https://reviews.apache.org/r/12810/diff/
> 
> 
> Testing
> ---
> 
> Tested on old master using xenserver.
>  Could not test on the latest one as it is broken.
> 
> 
> Thanks,
> 
> bharat kumar
> 
>

Re: Review Request 12858: CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network offering.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12858/#review23704
---


Commit 5c45645932f7537aa93f4d060e114a40dc7cbcb6 in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=5c45645 ]

CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find network 
offering.

Signed-off-by: Prasanna Santhanam 
(cherry picked from commit a06b3f40eba6ebc3af5d7c366ec757703a085f67)


- ASF Subversion and Git Services


On July 23, 2013, 2:53 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12858/
> ---
> 
> (Updated July 23, 2013, 2:53 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3724
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3724: Test case "test_vpc_vms_deployment.py" failed to find 
> network offering.
> 
> The test_script was broken because "nw_off" got appended twice for cleanup 
> instead of "nw_off_no_lb".
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_vpc_vms_deployment.py c599d96 
> 
> Diff: https://reviews.apache.org/r/12858/diff/
> 
> 
> Testing
> ---
> 
> Verified the marvin test locally on cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: [DISCUSS] API changes / compatibility between 4.1 and 4.2

[Marcus Sorensen]

I think maybe there's a typo there, you mean to say a 4.1 agent can't
communicate to a 4.2 mgmt server, and vice versa?

The systemvm problem, do you mean that the system vms need to be
upgraded, or are you talking about a specific bug?

On Tue, Jul 23, 2013 at 10:00 AM, Wido den Hollander  wrote:
> Hi,
>
> Yesterday I encountered this bug:
> https://issues.apache.org/jira/browse/CLOUDSTACK-3714
>
> A short summary: A 4.2 agent can't communicate with a 4.2 management server
> and vise versa.
>
> What this means is that during an upgrade you can't upgrade the Agents one
> by one, since a upgraded Agent won't be able to participate in the cluster.
>
> This is a big breaker for me, since going from 4.1 to 4.2 and having to shut
> down your management server and all your Agents at once seems like a major
> upgrade to me.
>
> I also encountered the problems with the new SystemVMs today while upgrading
> my management server to 4.2 and it seems to me that this upgrade won't be an
> easy one for a lot of users.
>
> So I'd like to start a discussing about this, since I'm almost thinking that
> 4.2 should become 5.0 because of the major changes.
>
> These major changes are good and they bring a lot of functionality to
> CloudStack, but sometimes it seems like stuff is coming in so fast that it's
> almost impossible to keep up with.
>
> How do other people think about this API change between 4.1 and 4.2?
>
> Do you also think that 4.2 should actually be called 5.0?
>
> Wido

Re: [rant] stupid test cases

[Prasanna Santhanam]

here's another one:

def test_07_delete_network_with_rules(self):
""" Test delete network that has PF/staticNat/LB rules/Network Acl

# Validate the following
# 1. Create a VPC with cidr - 10.1.1.1/16
# 2. Add network1(10.1.1.1/24) and network2(10.1.2.1/24) to this VPC.
# 3. Deploy vm1 and vm2 in network1 and vm3 and vm4 in network2.
# 4. Create a PF /Static Nat/LB rule for vms in network1.
# 5. Create a PF /Static Nat/LB rule for vms in network2.
# 6. Create ingress network ACL for allowing all the above rules from
#public ip range on network1 and network2.
# 7. Create egress network ACL for network1 and network2 to access
#google.com.
# 8. Create a private gateway for this VPC and add a static route to
#this gateway
# 9. Create a VPN gateway for this VPC and add a static route to this
#gateway.
# 10. Make sure that all the PF,LB, Static NAT rules work as expected
# 11. Make sure that we are able to access google from all user Vms
# 12. Make sure that the newly added private gateway's and VPN
#gateway's static routes work as expected.
# Steps:
# 1. Delete the 1st network.
# 2. Delete account
# Validations:
# 1. As part of network deletion all the resources attached with
#network should get deleted. All other VMs and rules shall work as
#expected
# 2. All the resources associated with account should be deleted

This is such a complicated test. I can see breaking it down into at least 5
tests. The point I'm trying to make here is simply this -

When we don't have simple tests that make sure if ACLs are working correctly we
shouldn't overindulge in this kind of testing. Bear in mind that testing that
simple ACL ingress works correctly is also a "system integration" test. There's
no reason why only unittests shold be the most condensed in form. 

if we had one test for ACL, one for VPC offerings, one for the VPN gw, one for
the check whether ACLs are able to connect to an external service like
'google'. The tests can be much simpler to debug and much better indicators of
failure

This entire scenario not just convolutes analysis of a failure but also makes
maintaining this a problem. I'm completely fine with integration tests as being
tests that put together things as a whole. But Sanjay worked on this today and
Sheng worked on another scenario in related to VPC load balancing. And it takes
a few hours to get this entire thing debugged and fixed. And the entire suite
(atleast 8 to 10 tests) are such scenarios.

We need to start working on critically reviewing the tests that come through.

Of course I'd like to hear others thoughts on this.

On Fri, Jul 19, 2013 at 06:06:18PM +0530, Prasanna Santhanam wrote:
> My problem is that we have tests that already check cleanup of
> accounts. A test should do the most crystallized set of steps to
> achieve the scenario and not try to put everything and the kitchen
> sink in to it. If we see ourselves doing that - we need to break down
> our tests into smaller blocks. They'll still be
> system/integration/live tests only.
> 
> As to the simulator - yes you can run these tests on a simulator
> today.
> 
> On Fri, Jul 19, 2013 at 01:12:14AM +, Alex Huang wrote:
> > I disagree.  Error handling should be part of our testing.
> > 
> > We should incorporate the simulator into the BVT and regression
> > tests.  On testcases that really is to test the business logic
> > rather than the provisioning code, the test case should perform all
> > of the provisioning on the simulator instead.  Then simulator can be
> > programmed to simulate VM stopped failure etc and see how the
> > business responds to these problems.
> > 
> > --Alex
> > 
> > > -Original Message-
> > > From: Anthony Xu [mailto:xuefei...@citrix.com]
> > > Sent: Thursday, July 18, 2013 3:02 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [rant] stupid test cases
> > > 
> > > +1   VM can be in "Stopped" state
> > > 
> > > 
> > > Anthony
> > > 
> > > -Original Message-
> > > From: Marcus Sorensen [mailto:shadow...@gmail.com]
> > > Sent: Wednesday, July 17, 2013 10:47 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: Re: [rant] stupid test cases
> > > 
> > > I can understand that we may want to test that everything related to the
> > > domain gets cleaned up properly. We have run into all sorts of things when
> > > deleting accounts, for example where resources won't clean up because the
> > > account is gone and we throw null pointers because a bunch of code looks 
> > > up
> > > account when deleting. However, to your point, VMs can be created in a
> > > "stopped" state and that wouldn't incur the overhead of deployment.
> > > On Jul 17, 2013 11:33 PM, "Prasanna Santhanam"  wrote:
> > > 
> > > > I was just going through one of the automated test 

Re: [DISCUSS] API changes / compatibility between 4.1 and 4.2

[Wido den Hollander]

Op 23 jul. 2013 om 18:16 heeft "Marcus Sorensen"  het 
volgende geschreven:

> I think maybe there's a typo there, you mean to say a 4.1 agent can't
> communicate to a 4.2 mgmt server, and vice versa?
> 

Correct, that was a typo. I mean 4.1 and 4.2

> The systemvm problem, do you mean that the system vms need to be
> upgraded, or are you talking about a specific bug?
> 

The upgrade process of the System VMs indeed.

Wido

> On Tue, Jul 23, 2013 at 10:00 AM, Wido den Hollander  wrote:
>> Hi,
>> 
>> Yesterday I encountered this bug:
>> https://issues.apache.org/jira/browse/CLOUDSTACK-3714
>> 
>> A short summary: A 4.2 agent can't communicate with a 4.2 management server
>> and vise versa.
>> 
>> What this means is that during an upgrade you can't upgrade the Agents one
>> by one, since a upgraded Agent won't be able to participate in the cluster.
>> 
>> This is a big breaker for me, since going from 4.1 to 4.2 and having to shut
>> down your management server and all your Agents at once seems like a major
>> upgrade to me.
>> 
>> I also encountered the problems with the new SystemVMs today while upgrading
>> my management server to 4.2 and it seems to me that this upgrade won't be an
>> easy one for a lot of users.
>> 
>> So I'd like to start a discussing about this, since I'm almost thinking that
>> 4.2 should become 5.0 because of the major changes.
>> 
>> These major changes are good and they bring a lot of functionality to
>> CloudStack, but sometimes it seems like stuff is coming in so fast that it's
>> almost impossible to keep up with.
>> 
>> How do other people think about this API change between 4.1 and 4.2?
>> 
>> Do you also think that 4.2 should actually be called 5.0?
>> 
>> Wido

Re: [Discuss] Apply rules on Virtual Router

[Nguyen Anh Tu]

Just still thinking about the incremental applying solution...

+1 for writing rules to file.


2013/7/23 Alex Huang 

> The file approach will definitely make it faster.
>
> Just thinking out loud, If we can write all of the rules on a file, why
> not do an iptables-save, perform a diff and apply the difference?
>
> --Alex
>
> > -Original Message-
> > From: Chiradeep Vittal [mailto:chiradeep.vit...@citrix.com]
> > Sent: Tuesday, July 23, 2013 5:08 AM
> > To: dev@cloudstack.apache.org
> > Cc: Nguyen Anh Tu
> > Subject: Re: [Discuss] Apply rules on Virtual Router
> >
> > It is quite hard to do a delta update correctly, so a complete rewrite
> of the
> > ruleset is the safest way to do it. Not sure why it is "slow", but I'd
> compare it
> > to the time taken to start a VM.
> > One way to make it slightly faster is to write the ruleset to a file and
> use
> > iptables-restore from the file.
> >
> > On 7/23/13 5:22 PM, "Nguyen Anh Tu"  wrote:
> >
> > >Anyone?
> > >
> > >
> > >2013/7/22 Nguyen Anh Tu 
> > >
> > >> Hi guys,
> > >>
> > >> While working with L3 network services, I found a problem in the
> > >>process  of applying iptables rules. It currently works not good in my
> > opinion.
> > >>When
> > >> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router
> > >>backups  old rules and re-apply all of non-revoked rules related to
> > >>source IP on the  new rule, including this one. It causes a slow,
> > >>especially when you have a  lot of running rules. When you delete a
> > >>rule, the process happens in the  same. The deleting rule is marked as
> > >>"revoked", so it doesn't appear in the  list. I think we should have a
> > >>better approach.
> > >>
> > >> Any idea?
> > >>
> > >> --
> > >>
> > >> 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: Reformatting UI code

[Chip Childers]

On Tue, Jul 23, 2013 at 09:37:26AM -0600, Mike Tutkowski wrote:
> One other thought here is that we might want to wait until, say, the day
> after code freeze - if allowable - and then check in these kinds of changes
> (basically give them special permission to go in after code freeze).
> 
> I think they're valuable since they do make the code much more readable,
> but it might be better if they go in once development is "frozen."

The challenge with that, is that sweeping changes like this (or
architectural changes as well) are best done *early* in a release cycle.
The challenge we've run into here is that while 4.2 work is proceeding,
master is open for 4.3 changes (and there is a preference that if
something big is going to come in, nows the time to do it).

Perhaps that model is what's broken?

> 
> 
> On Tue, Jul 23, 2013 at 8:42 AM, Sebastien Goasguen wrote:
> 
> >
> > On Jul 22, 2013, at 6:54 PM, Jessica Wang  wrote:
> >
> > > Sebastien,
> > >
> > > Brian and I are fix 4.2 bugs.
> > > Any fix for 4.2 bugs should go to both 4.2 branch and master branch.
> > > That's why Brian and I still need to merge our check-in between master
> > branch and 4.2 branch.
> >
> > Ok got it .
> >
> > Ian next time you send a patch related to the UI maybe you can set Jessica
> > and Brian as reviewers.
> >
> > thanks,
> >
> > >
> > > Jessica
> > >
> > > -Original Message-
> > > From: Sebastien Goasguen [mailto:run...@gmail.com]
> > > Sent: Monday, July 22, 2013 3:53 PM
> > > To: Brian Federle
> > > Cc: Jessica Wang; Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > dev@cloudstack.apache.org
> > > Subject: Re: Reformatting UI code
> > >
> > >
> > >
> > > On 23 Jul 2013, at 00:34, Brian Federle 
> > wrote:
> > >
> > >> In this case it was fine, I could resolve the conflicts by git's
> > -Xignore-space-change and reformatting accordingly.
> > >>
> > >> The main issue was that I didn't see any indication that this was being
> > committed from reading the discussion thread. Since we're doing a lot of
> > bugfixing right now for 4.2, a large sweeping commit like this should at
> > least be posted as part of thread, so that there is heads-up so that we can
> > prepare any pending changes we have to avoid being blocked.
> > >>
> > >
> > > Ok but this is not for 4.2. It was committed to master. How is it
> > impacting your 4.2 fixes ?
> > >
> > >
> > >> -Brian
> > >>
> > >> -Original Message-
> > >> From: Sebastien Goasguen [mailto:run...@gmail.com]
> > >> Sent: Monday, July 22, 2013 3:26 PM
> > >> To: Jessica Wang
> > >> Cc: Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > dev@cloudstack.apache.org; Brian Federle
> > >> Subject: Re: Reformatting UI code
> > >>
> > >> Hi jessica,
> > >>
> > >> Sorry this is causing you trouble. Ian us developing his ldap plugin in
> > a feature branch .
> > >> This was a UI cosmetic code change, i saw that pranav shipped it in RB
> > but did not apply the patch. Since the three if us work in more or less the
> > same time zone i applied the patch quickly to master. I did think about
> > other UI work but since 4.2 is in feature freeze i did not think committing
> > to master would be a problem.
> > >>
> > >> Moreover i did not know u were working on UI feature branches. Where
> > can we look at the feature descriptions ?
> > >>
> > >> I suppose u could revert the patch . But for features for 4.3 i think
> > it should be rebase of your feature branches . I dont think this change
> > should be impacting 4.2 not sure why u say thats the case.
> > >>
> > >> Thoughts ?
> > >>
> > >> -Sebastien
> > >>
> > >> On 23 Jul 2013, at 00:07, Jessica Wang  wrote:
> > >>
> > >>> Pranav, Ian, Sebastien,
> > >>>
> > >>> The problem is the duration between the time Ian brought it up on the
> > mailing list and the time Ian/Sebastien checked in the change to master
> > branch is too short (less than 4 hours).
> > >>>
> > >>> Ian brought it up on the mailing list at Thu 7/18/2013 5:44 AM (email
> > subject is "Auto format javascript").
> > >>> Sebastien checked in Ian's change to master branch at 7/18/2013 9:34
> > AM (Commit hash: ad69bc8da3244b783dd003ddf3184fca2762c514).
> > >>>
> > >>> This is a big change of UI code.
> > >>> In GIT's view, every line in JS files has been changed (If you look at
> > code difference in GIT's history).
> > >>> GIT sees it as "delete all lines and add new different lines".
> > >>> I was unable to merge my check-in from master branch to 4.2 branch (or
> > any other branch) since GIT sees JS files in master branch and other
> > branches are totally different.
> > >>>
> > >>> Shouldn't this kind of big change be checked in to a different branch
> > (not master branch) first? Then, submit a merge request to community, wait
> > for 72 hours, then merge to master branch eventually?
> > >>>
> > >>> Jessica
> > >>>
> > >>>
> > >>> -Original Message-
> > >>> From: Pranav Saxena [mailto:psb...@gmail.com]
> > >>> Sent: Monday, July 22, 2013 12:18 PM
> > >>> To: dev@clo

Re: Reformatting UI code

[David Nalley]

On Tue, Jul 23, 2013 at 12:36 PM, Chip Childers
 wrote:
> On Tue, Jul 23, 2013 at 09:37:26AM -0600, Mike Tutkowski wrote:
>> One other thought here is that we might want to wait until, say, the day
>> after code freeze - if allowable - and then check in these kinds of changes
>> (basically give them special permission to go in after code freeze).
>>
>> I think they're valuable since they do make the code much more readable,
>> but it might be better if they go in once development is "frozen."
>
> The challenge with that, is that sweeping changes like this (or
> architectural changes as well) are best done *early* in a release cycle.
> The challenge we've run into here is that while 4.2 work is proceeding,
> master is open for 4.3 changes (and there is a preference that if
> something big is going to come in, nows the time to do it).
>
> Perhaps that model is what's broken?
>

I agree - and the package name change is another such change, that
when it lands will result in a ton of extra work for dealing with bugs
across branches.

If it's really just reformatting - why aren't we reformatting 4.2 as well.

Re: [Discuss] Apply rules on Virtual Router

[Wei ZHOU]

+1
It is also much better if CS supports adding multiple rules in one request.



2013/7/23 Nguyen Anh Tu 

> Just still thinking about the incremental applying solution...
>
> +1 for writing rules to file.
>
>
> 2013/7/23 Alex Huang 
>
> > The file approach will definitely make it faster.
> >
> > Just thinking out loud, If we can write all of the rules on a file, why
> > not do an iptables-save, perform a diff and apply the difference?
> >
> > --Alex
> >
> > > -Original Message-
> > > From: Chiradeep Vittal [mailto:chiradeep.vit...@citrix.com]
> > > Sent: Tuesday, July 23, 2013 5:08 AM
> > > To: dev@cloudstack.apache.org
> > > Cc: Nguyen Anh Tu
> > > Subject: Re: [Discuss] Apply rules on Virtual Router
> > >
> > > It is quite hard to do a delta update correctly, so a complete rewrite
> > of the
> > > ruleset is the safest way to do it. Not sure why it is "slow", but I'd
> > compare it
> > > to the time taken to start a VM.
> > > One way to make it slightly faster is to write the ruleset to a file
> and
> > use
> > > iptables-restore from the file.
> > >
> > > On 7/23/13 5:22 PM, "Nguyen Anh Tu"  wrote:
> > >
> > > >Anyone?
> > > >
> > > >
> > > >2013/7/22 Nguyen Anh Tu 
> > > >
> > > >> Hi guys,
> > > >>
> > > >> While working with L3 network services, I found a problem in the
> > > >>process  of applying iptables rules. It currently works not good in
> my
> > > opinion.
> > > >>When
> > > >> you apply a new rule (eg. StaticNat or Egress rule), Virtual Router
> > > >>backups  old rules and re-apply all of non-revoked rules related to
> > > >>source IP on the  new rule, including this one. It causes a slow,
> > > >>especially when you have a  lot of running rules. When you delete a
> > > >>rule, the process happens in the  same. The deleting rule is marked
> as
> > > >>"revoked", so it doesn't appear in the  list. I think we should have
> a
> > > >>better approach.
> > > >>
> > > >> Any idea?
> > > >>
> > > >> --
> > > >>
> > > >> 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: Regression failures

[Prasanna Santhanam]

On Tue, Jul 23, 2013 at 11:56:02AM +0530, Prasanna Santhanam wrote:
> Thanks Sowmya, I see more issues in the list now when the component is
> Automation.

Filtered this further down to :

"""
project = CloudStack AND (status = Open OR status = Reopened OR status = "Ready 
To Review") AND (summary ~ Automation OR component = Automation) and type = Bug 
ORDER BY key DESC
"""

The previous filter showed subtasks that are planned efforts to automate entire
features which are removed now to show only failures in the tests that we are
fixing.

-- 
Prasanna.,


Powered by BigRock.com

Review Request 12865: CLOUDSTACK-3693 - Fix more account objects

[Sowmya Krishnan]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12865/
---

Review request for cloudstack and Prasanna Santhanam.


Bugs: CLOUDSTACK-3693


Repository: cloudstack-git


Description
---

Fixed more account objects. Fixes CLOUDSTACK-3693


Diffs
-

  test/integration/component/test_egress_rules.py c8b3eee 

Diff: https://reviews.apache.org/r/12865/diff/


Testing
---


Thanks,

Sowmya Krishnan

New Components on JIRA

[Prasanna Santhanam]

The following new components have been added to JIRA to narrow down
issues into the right buckets. 

(name, description)
---
* Infra - Infrastructure managed by the project - CloudStack - (jenkins, 
builds, repos, mirrors)
* SystemVM - SystemVM appliances, images, scripts, ssvm, cpvm
* Virtual Router - Anything related to the Virtual Router appliance
* XCP - Xen Cloud Platform
---

Anyone see any problems?

-- 
Prasanna.,


Powered by BigRock.com

Review Request 12874: Fix reboot failures with wait time

[Sowmya Krishnan]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12874/
---

Review request for cloudstack and Prasanna Santhanam.


Repository: cloudstack-git


Description
---

Fix reboot failures with sleep time since Netscaler at times takes almost 60 - 
90 seconds for identifying the service to be up after rebooting a VM.


Diffs
-


Diff: https://reviews.apache.org/r/12874/diff/


Testing
---

Tested locally and the script now passes


Thanks,

Sowmya Krishnan

[Proposal] Routing between guest networks in VLAN isolation method.

[Nguyen Anh Tu]

Hi guys,

I write a proposal about implementing routing method for guest networks
using VLAN isolation. At the moment, they can reach each other due to
interVLAN routing in VPC model, but can not in Guest network model. So the
key point is make some static routes between them, including iptables rules
for filtering ports and protocols. Please take a look on my proposal, link
below.

https://cwiki.apache.org/confluence/display/CLOUDSTACK/Routing+between+Guest+networks


-- 

N.g.U.y.e.N.A.n.H.t.U

Re: New Components on JIRA

[Ahmad Emneina]

+1, good stuff Prasanna. I'm a fan of more concise components to be filing
against.


On Tue, Jul 23, 2013 at 10:04 AM, Prasanna Santhanam  wrote:

> The following new components have been added to JIRA to narrow down
> issues into the right buckets.
>
> (name, description)
> ---
> * Infra - Infrastructure managed by the project - CloudStack - (jenkins,
> builds, repos, mirrors)
> * SystemVM - SystemVM appliances, images, scripts, ssvm, cpvm
> * Virtual Router - Anything related to the Virtual Router appliance
> * XCP - Xen Cloud Platform
> ---
>
> Anyone see any problems?
>
> --
> Prasanna.,
>
> 
> Powered by BigRock.com
>
>

RE: New Components on JIRA

[Ram Ganesh]

> -Original Message-
> From: Prasanna Santhanam [mailto:t...@apache.org]
> Sent: 23 July 2013 22:36
> To: CloudStack Dev
> Subject: New Components on JIRA
> 
> The following new components have been added to JIRA to narrow down
> issues into the right buckets.
> 
> (name, description)
> ---
> * Infra - Infrastructure managed by the project - CloudStack - (jenkins, 
> builds,
> repos, mirrors)
> * SystemVM - SystemVM appliances, images, scripts, ssvm, cpvm
> * Virtual Router - Anything related to the Virtual Router appliance
> * XCP - Xen Cloud Platform
> ---
> 
> Anyone see any problems?

Prasanna,

How about  - automation-product and automation-script components? 
Automation-product for all product bugs discovered by the automation engine and 
automation-script for all automation script issues?

Thanks,
RamG
 

> 
> --
> Prasanna.,
> 
> 
> Powered by BigRock.com

Re: [rant] stupid test cases

[Ahmad Emneina]

In terms of integration tests, it could be broken up into smaller pieces,
but as part of an overall 'networking functional' suite. Maybe we need to
further divide the tests between functional tests and unit or integration
tests. We really need to improve the testing around error handling and the
lifecycle of a feature.

For example network.gc does its job in cleaning up rules after a network
has shut down. But when restarting a vm in said network, none of the
previous rules get reinstated. That to me is brittle and needs to mature...
sorry for the anti-rant :p


On Tue, Jul 23, 2013 at 9:22 AM, Prasanna Santhanam  wrote:

> here's another one:
>
> def test_07_delete_network_with_rules(self):
> """ Test delete network that has PF/staticNat/LB rules/Network Acl
>
> # Validate the following
> # 1. Create a VPC with cidr - 10.1.1.1/16
> # 2. Add network1(10.1.1.1/24) and network2(10.1.2.1/24) to this
> VPC.
> # 3. Deploy vm1 and vm2 in network1 and vm3 and vm4 in network2.
> # 4. Create a PF /Static Nat/LB rule for vms in network1.
> # 5. Create a PF /Static Nat/LB rule for vms in network2.
> # 6. Create ingress network ACL for allowing all the above rules
> from
> #public ip range on network1 and network2.
> # 7. Create egress network ACL for network1 and network2 to access
> #google.com.
> # 8. Create a private gateway for this VPC and add a static route
> to
> #this gateway
> # 9. Create a VPN gateway for this VPC and add a static route to
> this
> #gateway.
> # 10. Make sure that all the PF,LB, Static NAT rules work as
> expected
> # 11. Make sure that we are able to access google from all user Vms
> # 12. Make sure that the newly added private gateway's and VPN
> #gateway's static routes work as expected.
> # Steps:
> # 1. Delete the 1st network.
> # 2. Delete account
> # Validations:
> # 1. As part of network deletion all the resources attached with
> #network should get deleted. All other VMs and rules shall
> work as
> #expected
> # 2. All the resources associated with account should be deleted
>
> This is such a complicated test. I can see breaking it down into at least 5
> tests. The point I'm trying to make here is simply this -
>
> When we don't have simple tests that make sure if ACLs are working
> correctly we
> shouldn't overindulge in this kind of testing. Bear in mind that testing
> that
> simple ACL ingress works correctly is also a "system integration" test.
> There's
> no reason why only unittests shold be the most condensed in form.
>
> if we had one test for ACL, one for VPC offerings, one for the VPN gw, one
> for
> the check whether ACLs are able to connect to an external service like
> 'google'. The tests can be much simpler to debug and much better
> indicators of
> failure
>
> This entire scenario not just convolutes analysis of a failure but also
> makes
> maintaining this a problem. I'm completely fine with integration tests as
> being
> tests that put together things as a whole. But Sanjay worked on this today
> and
> Sheng worked on another scenario in related to VPC load balancing. And it
> takes
> a few hours to get this entire thing debugged and fixed. And the entire
> suite
> (atleast 8 to 10 tests) are such scenarios.
>
> We need to start working on critically reviewing the tests that come
> through.
>
> Of course I'd like to hear others thoughts on this.
>
> On Fri, Jul 19, 2013 at 06:06:18PM +0530, Prasanna Santhanam wrote:
> > My problem is that we have tests that already check cleanup of
> > accounts. A test should do the most crystallized set of steps to
> > achieve the scenario and not try to put everything and the kitchen
> > sink in to it. If we see ourselves doing that - we need to break down
> > our tests into smaller blocks. They'll still be
> > system/integration/live tests only.
> >
> > As to the simulator - yes you can run these tests on a simulator
> > today.
> >
> > On Fri, Jul 19, 2013 at 01:12:14AM +, Alex Huang wrote:
> > > I disagree.  Error handling should be part of our testing.
> > >
> > > We should incorporate the simulator into the BVT and regression
> > > tests.  On testcases that really is to test the business logic
> > > rather than the provisioning code, the test case should perform all
> > > of the provisioning on the simulator instead.  Then simulator can be
> > > programmed to simulate VM stopped failure etc and see how the
> > > business responds to these problems.
> > >
> > > --Alex
> > >
> > > > -Original Message-
> > > > From: Anthony Xu [mailto:xuefei...@citrix.com]
> > > > Sent: Thursday, July 18, 2013 3:02 PM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: RE: [rant] stupid test cases
> > > >
> > > > +1   VM can be in "Stopped" state
> > > >
> > > >
> > > > Anthony
> > > >
> > >

Re: [Proposal] Routing between guest networks in VLAN isolation method.

[David Nalley]

On Tue, Jul 23, 2013 at 1:21 PM, Nguyen Anh Tu  wrote:
> Hi guys,
>
> I write a proposal about implementing routing method for guest networks
> using VLAN isolation. At the moment, they can reach each other due to
> interVLAN routing in VPC model, but can not in Guest network model. So the
> key point is make some static routes between them, including iptables rules
> for filtering ports and protocols. Please take a look on my proposal, link
> below.
>
> https://cwiki.apache.org/confluence/display/CLOUDSTACK/Routing+between+Guest+networks
>

Isn't this exactly the case that VPC is designed to solve?
What's the benefit of doing this? If we did this, would we continue using VPC?

--David

Re: [Proposal] Routing between guest networks in VLAN isolation method.

[Ahmad Emneina]

it looks like this is meant to enable routing between two different guest
isolated networks. Also suggests the ability to allow natted traffic
through another accounts network. I'd like to know more as to why someone
would want this, I dont see how the natting would work since the CIDR for
both networks are identical (forgive the ignorance here).


On Tue, Jul 23, 2013 at 10:26 AM, David Nalley  wrote:

> On Tue, Jul 23, 2013 at 1:21 PM, Nguyen Anh Tu  wrote:
> > Hi guys,
> >
> > I write a proposal about implementing routing method for guest networks
> > using VLAN isolation. At the moment, they can reach each other due to
> > interVLAN routing in VPC model, but can not in Guest network model. So
> the
> > key point is make some static routes between them, including iptables
> rules
> > for filtering ports and protocols. Please take a look on my proposal,
> link
> > below.
> >
> >
> https://cwiki.apache.org/confluence/display/CLOUDSTACK/Routing+between+Guest+networks
> >
>
> Isn't this exactly the case that VPC is designed to solve?
> What's the benefit of doing this? If we did this, would we continue using
> VPC?
>
> --David
>

Re: [Proposal] Routing between guest networks in VLAN isolation method.

[Chip Childers]

On Tue, Jul 23, 2013 at 01:26:08PM -0400, David Nalley wrote:
> On Tue, Jul 23, 2013 at 1:21 PM, Nguyen Anh Tu  wrote:
> > Hi guys,
> >
> > I write a proposal about implementing routing method for guest networks
> > using VLAN isolation. At the moment, they can reach each other due to
> > interVLAN routing in VPC model, but can not in Guest network model. So the
> > key point is make some static routes between them, including iptables rules
> > for filtering ports and protocols. Please take a look on my proposal, link
> > below.
> >
> > https://cwiki.apache.org/confluence/display/CLOUDSTACK/Routing+between+Guest+networks
> >
> 
> Isn't this exactly the case that VPC is designed to solve?
> What's the benefit of doing this? If we did this, would we continue using VPC?
> 
> --David
> 

Well right now, the main issue is that VPC follows the AWS VPC concepts
of allocating a single block for the VPC.  This isn't actually flexible
enough for some environments, and Nguyen's proposal is something that I've
been looking into myself.

Nguyen, when you state "All configurations are done by admin only.",
which admin?  Root?  If root only, why?

Localization pages on Jira

[Milamber]

Hello,

I've add a new section "Localization L10N" in Development section.
https://cwiki.apache.org/confluence/display/CLOUDSTACK/Localization+%28L10N%29


Currently there have 2 pages:

  * Debug Web UI with another language than English


(to find the issue of the white page after login)

  * Update L10N files from Transifex to git repo


(a good thing to do before a new release)


Milamber

[GSOC]cannot add host with the lastest master

[Han,Meng]

Hi all,

When I deploy CloudStack using the latest master, I can start jetty but 
it gave me the following error:


com.google.gson.JsonParseException: The JsonDeserializer 
com.cloud.agent.transport.ArrayTypeAdaptor@3bfd6f73 failed to 
deserialize json object 
[{"StartupRoutingCommand":{"cpus":2,"speed":3000,"memory":3844370432,"dom0MinMemory":384437043,"poolSync":false,"vms":{},"caps":"hvm,snapshot","pool":"/root","hypervisorType":"KVM","hostDetails":{"com.cloud.network.Networks.RouterPrivateIpStrategy":"HostLocal","Host.OS":"CentOS","Host.OS.Kernel.Version":"2.6.32-358.el6.x86_64","Host.OS.Version":"6.4"},"type":"Routing","dataCenter":"1","pod":"1","cluster":"1","guid":"a7320748-6346-3c9a-975e-90ac4ae4a986-LibvirtComputingResource","name":"meng.acis.ufl.edu","id":1,"version":"4.1.0","publicIpAddress":"10.244.18.55","publicNetmask":"255.255.255.0","publicMacAddress":"00:23:ae:94:f7:22","privateIpAddress":"10.244.18.55","privateMacAddress":"00:23:ae:94:f7:22","privateNetmask":"255.255.255.0","storageIpAddress":"10.244.18.55","storageNetmask":"255.255.255.0","storageMacAddress":"00:23:ae:94:f7:22","resourceName":"LibvirtComputingResource","gatewayIpAddress":"10.244.18.1","contextMap":{},"wait":0}},{"StartupStorageCommand":{"totalSi!

ze":0,"po
olInfo":{"uuid":"2b96eacb-4af5-48f3-bd37-31123a14bf70","host":"10.244.18.55","localPath":"/var/lib/libvirt/images/","hostPath":"/var/lib/libvirt/images/","poolType":"Filesystem","capacityBytes":52844687360,"availableBytes":23876186112},"resourceType":"STORAGE_POOL","hostDetails":{},"type":"Storage","dataCenter":"1","pod":"1","guid":"a7320748-6346-3c9a-975e-90ac4ae4a986-LibvirtComputingResource","name":"meng.acis.ufl.edu","id":1,"version":"4.1.0","resourceName":"LibvirtComputingResource","contextMap":{},"wait":0}}] 
given the type class [Lcom.cloud.agent.api.Command;
	at 
com.google.gson.JsonDeserializerExceptionWrapper.deserialize(JsonDeserializerExceptionWrapper.java:64)
	at 
com.google.gson.JsonDeserializationVisitor.invokeCustomDeserializer(JsonDeserializationVisitor.java:92)
	at 
com.google.gson.JsonDeserializationVisitor.visitUsingCustomHandler(JsonDeserializationVisitor.java:80)

at com.google.gson.ObjectNavigator.accept(ObjectNavigator.java:101)
	at 
com.google.gson.JsonDeserializationContextDefault.fromJsonArray(JsonDeserializationContextDefault.java:67)
	at 
com.google.gson.JsonDeserializationContextDefault.deserialize(JsonDeserializationContextDefault.java:52)

at com.google.gson.Gson.fromJson(Gson.java:551)
at com.google.gson.Gson.fromJson(Gson.java:498)
at com.cloud.agent.transport.Request.getCommands(Request.java:235)
	at 
com.cloud.agent.manager.AgentManagerImpl$AgentHandler.processRequest(AgentManagerImpl.java:1201)
	at 
com.cloud.agent.manager.AgentManagerImpl$AgentHandler.doTask(AgentManagerImpl.java:1354)
	at 
com.cloud.agent.manager.ClusteredAgentManagerImpl$ClusteredAgentHandler.doTask(ClusteredAgentManagerImpl.java:666)

at com.cloud.utils.nio.Task.run(Task.java:83)
	at 
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110)
	at 
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603)

at java.lang.Thread.run(Thread.java:722)
Caused by: com.cloud.utils.exception.CloudRuntimeException: can't find 
StartupRoutingCommand
	at 
com.cloud.agent.transport.ArrayTypeAdaptor.deserialize(ArrayTypeAdaptor.java:75)
	at 
com.cloud.agent.transport.ArrayTypeAdaptor.deserialize(ArrayTypeAdaptor.java:36)
	at 
com.google.gson.JsonDeserializerExceptionWrapper.deserialize(JsonDeserializerExceptionWrapper.java:51)

... 15 more

Due to this error, I am not able to add any KVM host. Could anyone give 
me some pointers?


The following is the command I used to build CloudStack.

mvn -P developer clean install
mvn -P developer -pl developer -Ddeploydb
mvn -pl  :cloud-client-ui jetty:run

Thank you all.
Cheers,
Meng

Re: [Proposal] Routing between guest networks in VLAN isolation method.

[Nguyen Anh Tu]

@David: it just only add a feature for guest network mode. If a VPC has too
much tiers, maybe one VPC router is not enough.
@Ahmad: this proposal use a route instead of a nat. Vms can talk via
privateIP. Of course, VMs in two guest networks currently can reach each
other via hairpinNAT.
@Chip: For the beginning, I just wanna limit privilege to Root admin. If
users can easily config route, maybe it cause some conflict rules.


2013/7/24 Chip Childers 

> On Tue, Jul 23, 2013 at 01:26:08PM -0400, David Nalley wrote:
> > On Tue, Jul 23, 2013 at 1:21 PM, Nguyen Anh Tu 
> wrote:
> > > Hi guys,
> > >
> > > I write a proposal about implementing routing method for guest networks
> > > using VLAN isolation. At the moment, they can reach each other due to
> > > interVLAN routing in VPC model, but can not in Guest network model. So
> the
> > > key point is make some static routes between them, including iptables
> rules
> > > for filtering ports and protocols. Please take a look on my proposal,
> link
> > > below.
> > >
> > >
> https://cwiki.apache.org/confluence/display/CLOUDSTACK/Routing+between+Guest+networks
> > >
> >
> > Isn't this exactly the case that VPC is designed to solve?
> > What's the benefit of doing this? If we did this, would we continue
> using VPC?
> >
> > --David
> >
>
> Well right now, the main issue is that VPC follows the AWS VPC concepts
> of allocating a single block for the VPC.  This isn't actually flexible
> enough for some environments, and Nguyen's proposal is something that I've
> been looking into myself.
>
> Nguyen, when you state "All configurations are done by admin only.",
> which admin?  Root?  If root only, why?
>



-- 

N.g.U.y.e.N.A.n.H.t.U

Re: Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/#review23706
---


Review 12864 PASSED the build test
The url of build cloudstack-master-with-patch #24 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/24/

- Jenkins Cloudstack.org


On July 23, 2013, 4:01 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12864/
> ---
> 
> (Updated July 23, 2013, 4:01 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3697
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> LB is already deleted when project is cleaned up. Removed it from cleanup 
> queue
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_resources.py 378238e 
> 
> Diff: https://reviews.apache.org/r/12864/diff/
> 
> 
> Testing
> ---
> 
> Tested script with fix locally and it passes now.
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

RE: [DESIGN] Why is enum a class...

[Frank Zhang]

Emm, it's not intended as a pun, but now I learned this pun :)

> -Original Message-
> From: Mike Tutkowski [mailto:mike.tutkow...@solidfire.com]
> Sent: Tuesday, July 23, 2013 8:43 AM
> To: dev@cloudstack.apache.org
> Subject: Re: [DESIGN] Why is enum a class...
> 
> The first thing I noticed about Frank's comment is that he started it with
> "Frankly speaking." (Frank/Frankly) Probably not intended as a pun, but funny
> nonetheless. :)
> 
> 
> On Mon, Jul 22, 2013 at 6:45 PM, Frank Zhang 
> wrote:
> 
> > Frankly speaking, if we are going to change enum, I would suggest not
> > using enmu anymore, instead, defining our own class like:
> >
> > public class VmType {
> > private static Map types =
> > Collections.synchronizedMap(new HashMap());
> > private final String typeName;
> >
> > public VmType (String typeName) {
> > this.typeName = typeName;
> > types.put(typeName, this);
> > }
> >
> > public static VmType valueOf(String typeName) {
> > VmType type = types.get(typeName);
> > if (type == null) {
> > throw new IllegalArgumentException("VmType type: "
> > + typeName + " was not registered by any VmType");
> > }
> > return type;
> > }
> >
> > @Override
> > public String toString() {
> > return typeName;
> > }
> >
> > @Override
> > public boolean equals(Object t) {
> > if (t == null || !(t instanceof VmType)) {
> > return false;
> > }
> >
> > VmType type = (VmType)t;
> > return type.toString().equals(typeName);
> > }
> >
> > @Override
> > public int hashCode() {
> > return typeName.hashCode();
> > }
> >
> > public static Set getAllTypeNames() {
> > return types.keySet();
> > }
> > }
> >
> > The only benefit of enum I see is you can use "==" instead of "equals()"
> > when comparing variables. However, it makes your code tight, every
> > time adding a new type you have to modify the enum.
> >
> > By above method, when a new plugin wants to extend a new VmType, it
> > simply
> > does:
> >
> > class MagicVmManagerImpl {
> > public static final VmType type = new VmType("MagicVm"); }
> >
> >
> > > -Original Message-
> > > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > > Sent: Monday, July 22, 2013 5:23 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [DESIGN] Why is enum a class...
> > >
> > > BTW, this code already shows a bug that stems from the static method
> > usage.
> > > It says ElasticIpVm is not a system vm (which I don't believe is true).
> >  Probably
> > > because whoever added elastic ip vm didn't see the static method and so
> > didn't
> > > add the vm into the method.
> > >
> > > --Alex
> > >
> > > > -Original Message-
> > > > From: Alex Huang [mailto:alex.hu...@citrix.com]
> > > > Sent: Monday, July 22, 2013 5:14 PM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: [DESIGN] Why is enum a class...
> > > >
> > > > I just went over this code and thought it was related and might be
> > > > interested to other developers.
> > > >
> > > > What's the difference between declaring a enum like this
> > > >
> > > > public enum Type {
> > > > User,
> > > > DomainRouter,
> > > > ConsoleProxy,
> > > > SecondaryStorageVm,
> > > > ElasticIpVm,
> > > > ElasticLoadBalancerVm,
> > > > InternalLoadBalancerVm,
> > > >
> > > > /*
> > > >  * UserBareMetal is only used for selecting
> > > > VirtualMachineGuru, there is no
> > > >  * VM with this type. UserBareMetal should treat exactly as
> > User.
> > > >  */
> > > > UserBareMetal;
> > > >
> > > > public static boolean isSystemVM(VirtualMachine.Type vmtype) {
> > > > if (DomainRouter.equals(vmtype)
> > > > || ConsoleProxy.equals(vmtype)
> > > > || SecondaryStorageVm.equals(vmtype) ||
> > > > InternalLoadBalancerVm.equals(vmtype)) {
> > > > return true;
> > > > }
> > > > return false;
> > > > }
> > > > }
> > > >
> > > > Vs
> > > >
> > > > public enum Type {
> > > > User(false),
> > > > DomainRouter(true),
> > > > ConsoleProxy(true),
> > > > SecondaryStorageVm(true),
> > > > ElasticIpVm(true),
> > > > ElasticLoadBalancerVm(true),
> > > > InternalLoadBalancerVm(true),
> > > >
> > > > /*
> > > >  * UserBareMetal is only used for selecting
> > > > VirtualMachineGuru, there is no
> > > >  * VM with this type. UserBareMetal should treat exactly as
> > User.
> > > >  */
> > > > UserBareMetal(false);
> > > >
> > > >private boolean _

Re: [GSOC]cannot add host with the lastest master

[Wei ZHOU]

I suggest updating the agent on host as well

Re: [rant] stupid test cases

[Mathias Mullins]

Prasanna, 

If we broke these tests down into smaller test specific test cases, could
we automate them into one specific end-to-end test together?

I.e. (And I'm over simplifying this for the sake of the e-mail)

1. Create Zone
2. Create Domain
3. Create Accounts
4. Create Users
5. Create Networks
6. Create VMs
7. Destroy VMs
8. Delete Users
9. Delete Accounts
10. Delete Zone

Each of these would have multiple steps and then would have a integration
test where you could pick and choose the unit test cases to be ran
together? I don't know the testing system well, but I know there are
automated systems out there that have done this, so I'm asking what our
systems can do. 

Matt 



On 7/23/13 1:26 PM, "Ahmad Emneina"  wrote:

>In terms of integration tests, it could be broken up into smaller pieces,
>but as part of an overall 'networking functional' suite. Maybe we need to
>further divide the tests between functional tests and unit or integration
>tests. We really need to improve the testing around error handling and the
>lifecycle of a feature.
>
>For example network.gc does its job in cleaning up rules after a network
>has shut down. But when restarting a vm in said network, none of the
>previous rules get reinstated. That to me is brittle and needs to
>mature...
>sorry for the anti-rant :p
>
>
>On Tue, Jul 23, 2013 at 9:22 AM, Prasanna Santhanam 
>wrote:
>
>> here's another one:
>>
>> def test_07_delete_network_with_rules(self):
>> """ Test delete network that has PF/staticNat/LB rules/Network
>>Acl
>>
>> # Validate the following
>> # 1. Create a VPC with cidr - 10.1.1.1/16
>> # 2. Add network1(10.1.1.1/24) and network2(10.1.2.1/24) to this
>> VPC.
>> # 3. Deploy vm1 and vm2 in network1 and vm3 and vm4 in network2.
>> # 4. Create a PF /Static Nat/LB rule for vms in network1.
>> # 5. Create a PF /Static Nat/LB rule for vms in network2.
>> # 6. Create ingress network ACL for allowing all the above rules
>> from
>> #public ip range on network1 and network2.
>> # 7. Create egress network ACL for network1 and network2 to
>>access
>> #google.com.
>> # 8. Create a private gateway for this VPC and add a static
>>route
>> to
>> #this gateway
>> # 9. Create a VPN gateway for this VPC and add a static route to
>> this
>> #gateway.
>> # 10. Make sure that all the PF,LB, Static NAT rules work as
>> expected
>> # 11. Make sure that we are able to access google from all user
>>Vms
>> # 12. Make sure that the newly added private gateway's and VPN
>> #gateway's static routes work as expected.
>> # Steps:
>> # 1. Delete the 1st network.
>> # 2. Delete account
>> # Validations:
>> # 1. As part of network deletion all the resources attached with
>> #network should get deleted. All other VMs and rules shall
>> work as
>> #expected
>> # 2. All the resources associated with account should be deleted
>>
>> This is such a complicated test. I can see breaking it down into at
>>least 5
>> tests. The point I'm trying to make here is simply this -
>>
>> When we don't have simple tests that make sure if ACLs are working
>> correctly we
>> shouldn't overindulge in this kind of testing. Bear in mind that testing
>> that
>> simple ACL ingress works correctly is also a "system integration" test.
>> There's
>> no reason why only unittests shold be the most condensed in form.
>>
>> if we had one test for ACL, one for VPC offerings, one for the VPN gw,
>>one
>> for
>> the check whether ACLs are able to connect to an external service like
>> 'google'. The tests can be much simpler to debug and much better
>> indicators of
>> failure
>>
>> This entire scenario not just convolutes analysis of a failure but also
>> makes
>> maintaining this a problem. I'm completely fine with integration tests
>>as
>> being
>> tests that put together things as a whole. But Sanjay worked on this
>>today
>> and
>> Sheng worked on another scenario in related to VPC load balancing. And
>>it
>> takes
>> a few hours to get this entire thing debugged and fixed. And the entire
>> suite
>> (atleast 8 to 10 tests) are such scenarios.
>>
>> We need to start working on critically reviewing the tests that come
>> through.
>>
>> Of course I'd like to hear others thoughts on this.
>>
>> On Fri, Jul 19, 2013 at 06:06:18PM +0530, Prasanna Santhanam wrote:
>> > My problem is that we have tests that already check cleanup of
>> > accounts. A test should do the most crystallized set of steps to
>> > achieve the scenario and not try to put everything and the kitchen
>> > sink in to it. If we see ourselves doing that - we need to break down
>> > our tests into smaller blocks. They'll still be
>> > system/integration/live tests only.
>> >
>> > As to the simulator - yes you can run these tests on a simulator
>> > today.
>> >
>> > On Fri, Jul 19, 2

RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Animesh Chaturvedi]

Ilya, Chip

Are your concerns addressed? If so I would like to proceed with adding this 
upgrade path.

Animesh

> -Original Message-
> From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> Sent: Friday, July 19, 2013 3:54 PM
> To: dev@cloudstack.apache.org
> Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> 
> 
> 
> > -Original Message-
> > From: Musayev, Ilya [mailto:imusa...@webmd.net]
> > Sent: Friday, July 19, 2013 2:13 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> >
> > Animesh,
> >
> > Would this have any effect on non-CCP installations?
> [Animesh>] No CCP -> ACS upgrade path files will not affects ACS->ACS
> upgrade path
> >
> > I see this move as a positive thing and it would be ideal to have in
> > 4.2, but we don't want to risk stability, hence I'm curious on the
> > safety and possible issues for non-CCP users.
> >
> > Thanks
> > ilya
> >
> >
> > > -Original Message-
> > > From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> > > Sent: Friday, July 19, 2013 4:43 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > >
> > >
> > >
> > > > -Original Message-
> > > > From: Chip Childers [mailto:chip.child...@sungard.com]
> > > > Sent: Friday, July 19, 2013 9:39 AM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > > >
> > > > On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi
> wrote:
> > > > > [Animesh>] Yes this seems late for 4.2, but it makes it easier
> > > > > to add
> > > > support now. If there is objection we can track it for 4.3.
> > > >
> > > > Are you sure it's easier to add support now?  That basically puts
> > > > us in a position where it's *possible* to attempt it.  That means
> > > > we have to test it, because if things go wrong then we are in a
> > > > mess
> > for 4.3.
> > > >
> > > > Am I getting it wrong?
> > > [Animesh>] Yes it is one upgrade path from 3.0.7 to 4.1 and follow
> > > through same upgrade path from 4.1 to 4.2 with couple of SQL files
> > > only. . Upgrade from 3.0,4-> 3.0.7 has already been tested in CCP
> > > releases and becomes the checkpoint. It was planned to be tested for
> > > CCP anyway. Sudha do you see it as a concern. Again if community
> > > thinks it is late, we can track for 4.3
> >

Re: New user - Introduction

[Mathias Mullins]

Welcome René! Glad to have you join.

Matt 


On 7/19/13 4:19 AM, "René Diepstraten"  wrote:

>Hi everyone.
>
>Since I have joined the mailing list, I'd like to say hello and introduce
>myself.
>My name is Rene Diepstraten, am a system administrator and the company I
>work for ( PCextreme ) uses Apache CloudStack.
>
>I'd like to contribute here and there with my skills in python and bash,
>and perhaps learn some java.
>A few small patches that I wrote have been accepted into Master, and I'd
>like to contribute more to this great project in the future.
>
>Best regards,
>
>Rene Diepstraten

Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Chip Childers]

My only concern was testing.  Looks like that's handled.  Let's roll
with it.

On Tue, Jul 23, 2013 at 06:15:33PM +, Animesh Chaturvedi wrote:
> Ilya, Chip
> 
> Are your concerns addressed? If so I would like to proceed with adding this 
> upgrade path.
> 
> Animesh
> 
> > -Original Message-
> > From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> > Sent: Friday, July 19, 2013 3:54 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > 
> > 
> > 
> > > -Original Message-
> > > From: Musayev, Ilya [mailto:imusa...@webmd.net]
> > > Sent: Friday, July 19, 2013 2:13 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > >
> > > Animesh,
> > >
> > > Would this have any effect on non-CCP installations?
> > [Animesh>] No CCP -> ACS upgrade path files will not affects ACS->ACS
> > upgrade path
> > >
> > > I see this move as a positive thing and it would be ideal to have in
> > > 4.2, but we don't want to risk stability, hence I'm curious on the
> > > safety and possible issues for non-CCP users.
> > >
> > > Thanks
> > > ilya
> > >
> > >
> > > > -Original Message-
> > > > From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> > > > Sent: Friday, July 19, 2013 4:43 PM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > > >
> > > >
> > > >
> > > > > -Original Message-
> > > > > From: Chip Childers [mailto:chip.child...@sungard.com]
> > > > > Sent: Friday, July 19, 2013 9:39 AM
> > > > > To: dev@cloudstack.apache.org
> > > > > Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > > > >
> > > > > On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi
> > wrote:
> > > > > > [Animesh>] Yes this seems late for 4.2, but it makes it easier
> > > > > > to add
> > > > > support now. If there is objection we can track it for 4.3.
> > > > >
> > > > > Are you sure it's easier to add support now?  That basically puts
> > > > > us in a position where it's *possible* to attempt it.  That means
> > > > > we have to test it, because if things go wrong then we are in a
> > > > > mess
> > > for 4.3.
> > > > >
> > > > > Am I getting it wrong?
> > > > [Animesh>] Yes it is one upgrade path from 3.0.7 to 4.1 and follow
> > > > through same upgrade path from 4.1 to 4.2 with couple of SQL files
> > > > only. . Upgrade from 3.0,4-> 3.0.7 has already been tested in CCP
> > > > releases and becomes the checkpoint. It was planned to be tested for
> > > > CCP anyway. Sudha do you see it as a concern. Again if community
> > > > thinks it is late, we can track for 4.3
> > >
> 
> 

RE: [DESIGN] Why is enum a class...

[Frank Zhang]

Speaking of method on enum, I used to think it's very handy feature but finally 
I failed on some scenarios and switched to own defined class.
The problem is enum is singleton that you can't save stateful data in it. For 
example:

public enum Type {
private Object userData;

private Object getUserData() {...}
 private void setUserData(Object data) {...}
 

Type SOME_TYPE;
}

This kind of thing is somewhat useless, userData here is actually global, every 
call to SOME_TYPE.setUserData() would change its value. 
So for usage perspective the methods you define on enmu are like static method 
on class which has its limitations.

As for methods in enmu body in your example, the only usage I can recall now is 
like what TimeUnit does. For CloudStack, we can do the similar
thing: SizeUnit that translates storage/memory size in different quantities.

public enum SizeUnit {
BYTE {
public long toKiloByte(long s) {
return (s / (k / b));
}

public long toMegaByte(long s) {
return (s / (m / b));
}

public long toGigaByte(long s) {
return (s / (g / b));
}

public long toTeraByte(long s) {
return (s / (t / b));
}
},
KILOBYTE {
public long toByte(long s) {
return (s * (k / b));
}

public long toMegaByte(long s) {
return (s / (m / k));
}

public long toGigaByte(long s) {
return (s / (g / k));
}

public long toTeraByte(long s) {
return (s / (t / k));
}
},
MEGABYTE {
public long toByte(long s) {
return (s * (m / b));
}

public long toKiloByte(long s) {
return (s * (m / k));
}

public long toGigaByte(long s) {
return (s / (g / m));
}

public long toTeraByte(long s) {
return (s / (t / m));
}
},
GIGABYTE {
public long toByte(long s) {
return (s * (g / b));
}

public long toKiloByte(long s) {
return (s * (g / k));
}

public long toMegaByte(long s) {
return (s * (g / m));
}

public long toTeraByte(long s) {
return (s / (t / g));
}
},
TERABYTE {
public long toByte(long s) {
return (s * (t / b));
}

public long toKiloByte(long s) {
return (s * (t / k));
}

public long toMegaByte(long s) {
return (s * (t / m));
}

public long toGigaByte(long s) {
return (s * (t / g));
}
};

private static final long b = 1;
private static final long k = b * 1024;
private static final long m = k * 1024;
private static final long g = m * 1024;
private static final long t = g * 1024;

public long toByte(long s) {
throw new AbstractMethodError();
}
public long toKiloByte(long s) {
throw new AbstractMethodError();
}   
public long toMegaByte(long s) {
throw new AbstractMethodError();
}
public long toGigaByte(long s) {
throw new AbstractMethodError();
}
public long toTeraByte(long s) {
throw new AbstractMethodError();
}
}

This may explain some of my views on java enum. It does have lots of advantages 
like compiler check, native == operator. Just like
Alex mentioned, for constants that unlikely to change in future, we can stick 
to enum. But for types that plugin may extend, we'd
better to use own defined classes. 

From: John Burwell [mailto:jburw...@basho.com] 
Sent: Tuesday, July 23, 2013 6:25 AM
To: dev@cloudstack.apache.org
Subject: Re: [DESIGN] Why is enum a class...

All,

+1 to Alex's design suggestion.  Another little know feature of enumerations is 
the ability to define abstract methods.  Therefore, Alex's example can be 
expanded as follows:

public enum Type {
   User(false)
{
@Override
public void doWork() {
}
},
   DomainRouter(true)
{
@Override
public void doWork() {
}
},
   ConsoleProxy(true)
{
@Override
public void doWork(){
{
},
   SecondaryStorageVm(true)
{
@Override
public void doWork(){
{
},
   ElasticIpVm(true)
{
@Override
public void doWork(){
{
},
   ElasticLoadBalancerVm(true)
{
@Override
public void doWork(){
{
},
   InternalLoadBalancerVm(true)
{
@Override
public void doWork(){
{
}

   /*
* UserBareMeta

Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Alena Prokharchyk]

Thank you, Chip. If nobody else objects, I'm planning to commit CCP-CS db 
upgrade code to the master/4.2 builds tomorrow.

-Alena.

From: Chip Childers 
mailto:chip.child...@sungard.com>>
Reply-To: "dev@cloudstack.apache.org" 
mailto:dev@cloudstack.apache.org>>
Date: Tuesday, July 23, 2013 11:25 AM
To: "dev@cloudstack.apache.org" 
mailto:dev@cloudstack.apache.org>>
Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP

My only concern was testing.  Looks like that's handled.  Let's roll
with it.

On Tue, Jul 23, 2013 at 06:15:33PM +, Animesh Chaturvedi wrote:
Ilya, Chip
Are your concerns addressed? If so I would like to proceed with adding this 
upgrade path.
Animesh
> -Original Message-
> From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> Sent: Friday, July 19, 2013 3:54 PM
> To: dev@cloudstack.apache.org
> Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
>
>
>
> > -Original Message-
> > From: Musayev, Ilya [mailto:imusa...@webmd.net]
> > Sent: Friday, July 19, 2013 2:13 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> >
> > Animesh,
> >
> > Would this have any effect on non-CCP installations?
> [Animesh>] No CCP -> ACS upgrade path files will not affects ACS->ACS
> upgrade path
> >
> > I see this move as a positive thing and it would be ideal to have in
> > 4.2, but we don't want to risk stability, hence I'm curious on the
> > safety and possible issues for non-CCP users.
> >
> > Thanks
> > ilya
> >
> >
> > > -Original Message-
> > > From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> > > Sent: Friday, July 19, 2013 4:43 PM
> > > To: dev@cloudstack.apache.org
> > > Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > >
> > >
> > >
> > > > -Original Message-
> > > > From: Chip Childers [mailto:chip.child...@sungard.com]
> > > > Sent: Friday, July 19, 2013 9:39 AM
> > > > To: dev@cloudstack.apache.org
> > > > Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> > > >
> > > > On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi
> wrote:
> > > > > [Animesh>] Yes this seems late for 4.2, but it makes it easier
> > > > > to add
> > > > support now. If there is objection we can track it for 4.3.
> > > >
> > > > Are you sure it's easier to add support now?  That basically puts
> > > > us in a position where it's *possible* to attempt it.  That means
> > > > we have to test it, because if things go wrong then we are in a
> > > > mess
> > for 4.3.
> > > >
> > > > Am I getting it wrong?
> > > [Animesh>] Yes it is one upgrade path from 3.0.7 to 4.1 and follow
> > > through same upgrade path from 4.1 to 4.2 with couple of SQL files
> > > only. . Upgrade from 3.0,4-> 3.0.7 has already been tested in CCP
> > > releases and becomes the checkpoint. It was planned to be tested for
> > > CCP anyway. Sudha do you see it as a concern. Again if community
> > > thinks it is late, we can track for 4.3
> >

Ruby dev wanted

[Sebastien Goasguen]

Hi folks,

I committed to do an interoperability workshop in Madrid, Spain in september:
http://www.cloudplugfest.org/cloud-interoperability-week/detailed-program

All "four sisters" of cloud are represented.

The main goal is to show that we can talk to CloudStack with OCCI and even CIMI 
standards.

The problem is that currently we cannot :)

So I am looking for some help from Ruby developers who would be willing to work 
on:

https://github.com/gwdg/rOCCI

and also on delta cloud (even though the project may have stalled):

http://deltacloud.apache.org

A good start might be Chip's ruby client:
https://github.com/chipchilders/cloudstack_ruby_client

Help, any takers  ?



-Sebastien

Re: Using the ACS simulator for jclouds "live" tests - some questions. =)

[Sebastien Goasguen]

On Jul 22, 2013, at 10:58 AM, Andrew Bayer  wrote:

> Woot, making progress - unsurprisingly, all of the jclouds live tests that
> expect to be able to create a VM and then SSH into it fail, but that's
> entirely reasonable. =) I'm making progress through the other failing tests
> - catching some issues in the jclouds CloudStack implementation along the
> way. Thanks, everybody!
> 

any chance you can write a quick blog post or wiki entry ?

> A.
> 
> On Sun, Jul 21, 2013 at 9:25 PM, Prasanna Santhanam  wrote:
> 
>> Hi - the advanced zone setup should work without troubles. The current
>> master is broken and a couple of bugs are pending for this after the
>> baremetal merge done recently.
>> 
>> CLOUDSTACK-3481 and CLOUDSTACK-1812.
>> 
>> On Sun, Jul 21, 2013 at 08:32:42PM -0700, Andrew Bayer wrote:
>>> Finally got a chance to work on this - trying to get the basic networking
>>> setup with Marvin seems to be barfing - I get the following exception in
>>> the CS logs when I run mvn -Pdeveloper,marvin.setup
>>> -Dmarvin.config=setup/dev/basic.cfg -pl :cloud-marvin integration-test:
>>> 
>>> WARN  [cloud.network.NetworkServiceImpl] (1384506397@qtp-455967038-0:)
>>> Exception:
>>> com.cloud.exception.InvalidParameterValueException: Invalid Network
>> Service
>>> Provider=BaremetalDhcpProvider
>>> at
>>> 
>> com.cloud.network.NetworkServiceImpl.addProviderToPhysicalNetwork(NetworkServiceImpl.java:3239)
>>> at
>>> 
>> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
>>> at
>>> 
>> com.cloud.network.NetworkServiceImpl.addDefaultBaremetalProvidersToPhysicalNetwork(NetworkServiceImpl.java:3747)
>>> at
>>> 
>> com.cloud.network.NetworkServiceImpl.createPhysicalNetwork(NetworkServiceImpl.java:2517)
>>> at
>>> 
>> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
>>> at
>>> 
>> org.apache.cloudstack.api.command.admin.network.CreatePhysicalNetworkCmd.create(CreatePhysicalNetworkCmd.java:160)
>>> at com.cloud.api.ApiDispatcher.dispatchCreateCmd(ApiDispatcher.java:104)
>>> at com.cloud.api.ApiServer.queueCommand(ApiServer.java:476)
>>> at com.cloud.api.ApiServer.handleRequest(ApiServer.java:372)
>>> at com.cloud.api.ApiServlet.processRequest(ApiServlet.java:305)
>>> at com.cloud.api.ApiServlet.doGet(ApiServlet.java:66)
>>> at javax.servlet.http.HttpServlet.service(HttpServlet.java:707)
>>> at javax.servlet.http.HttpServlet.service(HttpServlet.java:820)
>>> at org.mortbay.jetty.servlet.ServletHolder.handle(ServletHolder.java:511)
>>> at
>> org.mortbay.jetty.servlet.ServletHandler.handle(ServletHandler.java:401)
>>> at
>>> 
>> org.mortbay.jetty.security.SecurityHandler.handle(SecurityHandler.java:216)
>>> at
>> org.mortbay.jetty.servlet.SessionHandler.handle(SessionHandler.java:182)
>>> at
>> org.mortbay.jetty.handler.ContextHandler.handle(ContextHandler.java:766)
>>> at org.mortbay.jetty.webapp.WebAppContext.handle(WebAppContext.java:450)
>>> at
>>> 
>> org.mortbay.jetty.handler.ContextHandlerCollection.handle(ContextHandlerCollection.java:230)
>>> at
>>> 
>> org.mortbay.jetty.handler.HandlerCollection.handle(HandlerCollection.java:114)
>>> at
>> org.mortbay.jetty.handler.HandlerWrapper.handle(HandlerWrapper.java:152)
>>> at org.mortbay.jetty.Server.handle(Server.java:326)
>>> at
>> org.mortbay.jetty.HttpConnection.handleRequest(HttpConnection.java:542)
>>> at
>>> 
>> org.mortbay.jetty.HttpConnection$RequestHandler.headerComplete(HttpConnection.java:928)
>>> at org.mortbay.jetty.HttpParser.parseNext(HttpParser.java:549)
>>> at org.mortbay.jetty.HttpParser.parseAvailable(HttpParser.java:212)
>>> at org.mortbay.jetty.HttpConnection.handle(HttpConnection.java:404)
>>> at
>>> 
>> org.mortbay.io.nio.SelectChannelEndPoint.run(SelectChannelEndPoint.java:410)
>>> at
>>> 
>> org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582)
>>> ERROR [cloud.api.ApiServer] (1384506397@qtp-455967038-0:) unhandled
>>> exception executing api command: createPhysicalNetwork
>>> com.cloud.utils.exception.CloudRuntimeException: Fail to create a
>> physical
>>> network
>>> at
>>> 
>> com.cloud.network.NetworkServiceImpl.createPhysicalNetwork(NetworkServiceImpl.java:2526)
>>> at
>>> 
>> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
>>> at
>>> 
>> org.apache.cloudstack.api.command.admin.network.CreatePhysicalNetworkCmd.create(CreatePhysicalNetworkCmd.java:160)
>>> at com.cloud.api.ApiDispatcher.dispatchCreateCmd(ApiDispatcher.java:104)
>>> at com.cloud.api.ApiServer.queueCommand(ApiServer.java:476)
>>> at com.cloud.api.ApiServer.handleRequest(ApiServer.java:372)
>>> at com.cloud.api.ApiServlet.processRequest(ApiServlet.java:305)
>>> at com.cloud.api.ApiServlet.doGet(ApiServlet.java:66)
>>> at javax.servlet.http.HttpServlet.service(HttpServlet.java:707)
>>> 

Re: Using the ACS simulator for jclouds "live" tests - some questions. =)

[Andrew Bayer]

Once I get it a bit more refined, yeah - probably gonna talk about it a bit
at the ACS meetup in Santa Clara next month too.

A.

On Tue, Jul 23, 2013 at 11:47 AM, Sebastien Goasguen wrote:

>
> On Jul 22, 2013, at 10:58 AM, Andrew Bayer  wrote:
>
> > Woot, making progress - unsurprisingly, all of the jclouds live tests
> that
> > expect to be able to create a VM and then SSH into it fail, but that's
> > entirely reasonable. =) I'm making progress through the other failing
> tests
> > - catching some issues in the jclouds CloudStack implementation along the
> > way. Thanks, everybody!
> >
>
> any chance you can write a quick blog post or wiki entry ?
>
> > A.
> >
> > On Sun, Jul 21, 2013 at 9:25 PM, Prasanna Santhanam 
> wrote:
> >
> >> Hi - the advanced zone setup should work without troubles. The current
> >> master is broken and a couple of bugs are pending for this after the
> >> baremetal merge done recently.
> >>
> >> CLOUDSTACK-3481 and CLOUDSTACK-1812.
> >>
> >> On Sun, Jul 21, 2013 at 08:32:42PM -0700, Andrew Bayer wrote:
> >>> Finally got a chance to work on this - trying to get the basic
> networking
> >>> setup with Marvin seems to be barfing - I get the following exception
> in
> >>> the CS logs when I run mvn -Pdeveloper,marvin.setup
> >>> -Dmarvin.config=setup/dev/basic.cfg -pl :cloud-marvin integration-test:
> >>>
> >>> WARN  [cloud.network.NetworkServiceImpl] (1384506397@qtp-455967038-0:)
> >>> Exception:
> >>> com.cloud.exception.InvalidParameterValueException: Invalid Network
> >> Service
> >>> Provider=BaremetalDhcpProvider
> >>> at
> >>>
> >>
> com.cloud.network.NetworkServiceImpl.addProviderToPhysicalNetwork(NetworkServiceImpl.java:3239)
> >>> at
> >>>
> >>
> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
> >>> at
> >>>
> >>
> com.cloud.network.NetworkServiceImpl.addDefaultBaremetalProvidersToPhysicalNetwork(NetworkServiceImpl.java:3747)
> >>> at
> >>>
> >>
> com.cloud.network.NetworkServiceImpl.createPhysicalNetwork(NetworkServiceImpl.java:2517)
> >>> at
> >>>
> >>
> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
> >>> at
> >>>
> >>
> org.apache.cloudstack.api.command.admin.network.CreatePhysicalNetworkCmd.create(CreatePhysicalNetworkCmd.java:160)
> >>> at
> com.cloud.api.ApiDispatcher.dispatchCreateCmd(ApiDispatcher.java:104)
> >>> at com.cloud.api.ApiServer.queueCommand(ApiServer.java:476)
> >>> at com.cloud.api.ApiServer.handleRequest(ApiServer.java:372)
> >>> at com.cloud.api.ApiServlet.processRequest(ApiServlet.java:305)
> >>> at com.cloud.api.ApiServlet.doGet(ApiServlet.java:66)
> >>> at javax.servlet.http.HttpServlet.service(HttpServlet.java:707)
> >>> at javax.servlet.http.HttpServlet.service(HttpServlet.java:820)
> >>> at
> org.mortbay.jetty.servlet.ServletHolder.handle(ServletHolder.java:511)
> >>> at
> >> org.mortbay.jetty.servlet.ServletHandler.handle(ServletHandler.java:401)
> >>> at
> >>>
> >>
> org.mortbay.jetty.security.SecurityHandler.handle(SecurityHandler.java:216)
> >>> at
> >> org.mortbay.jetty.servlet.SessionHandler.handle(SessionHandler.java:182)
> >>> at
> >> org.mortbay.jetty.handler.ContextHandler.handle(ContextHandler.java:766)
> >>> at
> org.mortbay.jetty.webapp.WebAppContext.handle(WebAppContext.java:450)
> >>> at
> >>>
> >>
> org.mortbay.jetty.handler.ContextHandlerCollection.handle(ContextHandlerCollection.java:230)
> >>> at
> >>>
> >>
> org.mortbay.jetty.handler.HandlerCollection.handle(HandlerCollection.java:114)
> >>> at
> >> org.mortbay.jetty.handler.HandlerWrapper.handle(HandlerWrapper.java:152)
> >>> at org.mortbay.jetty.Server.handle(Server.java:326)
> >>> at
> >> org.mortbay.jetty.HttpConnection.handleRequest(HttpConnection.java:542)
> >>> at
> >>>
> >>
> org.mortbay.jetty.HttpConnection$RequestHandler.headerComplete(HttpConnection.java:928)
> >>> at org.mortbay.jetty.HttpParser.parseNext(HttpParser.java:549)
> >>> at org.mortbay.jetty.HttpParser.parseAvailable(HttpParser.java:212)
> >>> at org.mortbay.jetty.HttpConnection.handle(HttpConnection.java:404)
> >>> at
> >>>
> >>
> org.mortbay.io.nio.SelectChannelEndPoint.run(SelectChannelEndPoint.java:410)
> >>> at
> >>>
> >>
> org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582)
> >>> ERROR [cloud.api.ApiServer] (1384506397@qtp-455967038-0:) unhandled
> >>> exception executing api command: createPhysicalNetwork
> >>> com.cloud.utils.exception.CloudRuntimeException: Fail to create a
> >> physical
> >>> network
> >>> at
> >>>
> >>
> com.cloud.network.NetworkServiceImpl.createPhysicalNetwork(NetworkServiceImpl.java:2526)
> >>> at
> >>>
> >>
> com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)
> >>> at
> >>>
> >>
> org.apache.cloudstack.api.command.admin.network.CreatePhysicalNetworkCmd.create(CreatePhysicalN

Re: [GSOC]cannot add host with the lastest master

[Han,Meng]

Hi Wei,

Thank you for your reply.

On Wed, 24 Jul 2013 02:08:36 +0800, Wei ZHOU wrote:

I suggest updating the agent on host as well



I run yum install cloudstack-agent on the kvm host, it told me that 
it's already the latest version.  I found the post 
here:https://issues.apache.org/jira/browse/CLOUDSTACK-3118
Rajesh Battala  mentioned that he installed the latest agent from the 
build CloudStack-non-OSS-MASTER-499-rhel6.3.


Do you know where can I get that build and how to start the agent if I 
have the build(will service cloudstack-agent start do) ?
I didn't find the build in 
https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;a=summary



Any help is appreciated.

Cheers,
Meng

Review Request 12877: Support HTTP to HTTPS redirect for UCS

[Amogh Vasekar]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12877/
---

Review request for cloudstack and Frank Zhang.


Bugs: https://issues.apache.org/jira/browse/CLOUDSTACK-3285


Repository: cloudstack-git


Description
---

The patch uses EasySSLProtocolFactory from httpclient.contrib.ssl package.
The implementation accepts self-signed certificates. This is similar to 
cloud-plugin-network-cisco-vnmc 's behavior.
As a part of the patch, the contrib code has been moved to cloud-utils to avoid 
duplication.
Since there is no maven published jar for contrib, the source currently has the 
actual code instead. This needs to be revisited.


Diffs
-

  plugins/hypervisors/ucs/src/com/cloud/ucs/manager/UcsHttpClient.java 66d2ed2 
  
plugins/network-elements/cisco-vnmc/src/org/apache/commons/httpclient/contrib/ssl/EasySSLProtocolSocketFactory.java
 52f0ea6 
  
plugins/network-elements/cisco-vnmc/src/org/apache/commons/httpclient/contrib/ssl/EasyX509TrustManager.java
 ae9f938 
  
utils/src/org/apache/commons/httpclient/contrib/ssl/EasySSLProtocolSocketFactory.java
 PRE-CREATION 
  utils/src/org/apache/commons/httpclient/contrib/ssl/EasyX509TrustManager.java 
PRE-CREATION 

Diff: https://reviews.apache.org/r/12877/diff/


Testing
---

Works on local environment + UCS
cisco-vnmc not tested after refactoring


Thanks,

Amogh Vasekar

Re: Reformatting UI code

[Mike Tutkowski]

Yeah, I agree, as well, that ideally such a change would go into the
younger branch (master, in this case) as opposed to being placed in a
branch that is about to be frozen.


On Tue, Jul 23, 2013 at 2:25 PM, Animesh Chaturvedi <
animesh.chaturv...@citrix.com> wrote:

>
>
> > -Original Message-
> > From: Chip Childers [mailto:chip.child...@sungard.com]
> > Sent: Tuesday, July 23, 2013 9:36 AM
> > To: dev@cloudstack.apache.org
> > Cc: Jessica Wang; Brian Federle; Pranav Saxena; Ian Duffy
> > (i...@ianduffy.ie)
> > Subject: Re: Reformatting UI code
> >
> > On Tue, Jul 23, 2013 at 09:37:26AM -0600, Mike Tutkowski wrote:
> > > One other thought here is that we might want to wait until, say, the
> > > day after code freeze - if allowable - and then check in these kinds
> > > of changes (basically give them special permission to go in after code
> > freeze).
> > >
> > > I think they're valuable since they do make the code much more
> > > readable, but it might be better if they go in once development is
> > "frozen."
> >
> > The challenge with that, is that sweeping changes like this (or
> > architectural changes as well) are best done *early* in a release cycle.
> > The challenge we've run into here is that while 4.2 work is proceeding,
> > master is open for 4.3 changes (and there is a preference that if
> > something big is going to come in, nows the time to do it).
> >
> > Perhaps that model is what's broken?
> >
> [Animesh>] I concur with Chip, 4.2 is near code freeze and we have a ton
> of blocker and critical to deal with.
> > >
> > >
> > > On Tue, Jul 23, 2013 at 8:42 AM, Sebastien Goasguen
> > wrote:
> > >
> > > >
> > > > On Jul 22, 2013, at 6:54 PM, Jessica Wang 
> > wrote:
> > > >
> > > > > Sebastien,
> > > > >
> > > > > Brian and I are fix 4.2 bugs.
> > > > > Any fix for 4.2 bugs should go to both 4.2 branch and master
> > branch.
> > > > > That's why Brian and I still need to merge our check-in between
> > > > > master
> > > > branch and 4.2 branch.
> > > >
> > > > Ok got it .
> > > >
> > > > Ian next time you send a patch related to the UI maybe you can set
> > > > Jessica and Brian as reviewers.
> > > >
> > > > thanks,
> > > >
> > > > >
> > > > > Jessica
> > > > >
> > > > > -Original Message-
> > > > > From: Sebastien Goasguen [mailto:run...@gmail.com]
> > > > > Sent: Monday, July 22, 2013 3:53 PM
> > > > > To: Brian Federle
> > > > > Cc: Jessica Wang; Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > > > dev@cloudstack.apache.org
> > > > > Subject: Re: Reformatting UI code
> > > > >
> > > > >
> > > > >
> > > > > On 23 Jul 2013, at 00:34, Brian Federle 
> > > > wrote:
> > > > >
> > > > >> In this case it was fine, I could resolve the conflicts by git's
> > > > -Xignore-space-change and reformatting accordingly.
> > > > >>
> > > > >> The main issue was that I didn't see any indication that this was
> > > > >> being
> > > > committed from reading the discussion thread. Since we're doing a
> > > > lot of bugfixing right now for 4.2, a large sweeping commit like
> > > > this should at least be posted as part of thread, so that there is
> > > > heads-up so that we can prepare any pending changes we have to avoid
> > being blocked.
> > > > >>
> > > > >
> > > > > Ok but this is not for 4.2. It was committed to master. How is it
> > > > impacting your 4.2 fixes ?
> > > > >
> > > > >
> > > > >> -Brian
> > > > >>
> > > > >> -Original Message-
> > > > >> From: Sebastien Goasguen [mailto:run...@gmail.com]
> > > > >> Sent: Monday, July 22, 2013 3:26 PM
> > > > >> To: Jessica Wang
> > > > >> Cc: Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > > > dev@cloudstack.apache.org; Brian Federle
> > > > >> Subject: Re: Reformatting UI code
> > > > >>
> > > > >> Hi jessica,
> > > > >>
> > > > >> Sorry this is causing you trouble. Ian us developing his ldap
> > > > >> plugin in
> > > > a feature branch .
> > > > >> This was a UI cosmetic code change, i saw that pranav shipped it
> > > > >> in RB
> > > > but did not apply the patch. Since the three if us work in more or
> > > > less the same time zone i applied the patch quickly to master. I did
> > > > think about other UI work but since 4.2 is in feature freeze i did
> > > > not think committing to master would be a problem.
> > > > >>
> > > > >> Moreover i did not know u were working on UI feature branches.
> > > > >> Where
> > > > can we look at the feature descriptions ?
> > > > >>
> > > > >> I suppose u could revert the patch . But for features for 4.3 i
> > > > >> think
> > > > it should be rebase of your feature branches . I dont think this
> > > > change should be impacting 4.2 not sure why u say thats the case.
> > > > >>
> > > > >> Thoughts ?
> > > > >>
> > > > >> -Sebastien
> > > > >>
> > > > >> On 23 Jul 2013, at 00:07, Jessica Wang 
> > wrote:
> > > > >>
> > > > >>> Pranav, Ian, Sebastien,
> > > > >>>
> > > > >>> The problem is the duration between the time Ian brought it up
> > > > >>> on the
> > > > mailing list and the time Ian/Seba

Re: [GSOC]cannot add host with the lastest master

[Wido den Hollander]

Hi Meng,

On 07/23/2013 09:28 PM, Han,Meng wrote:

Hi Wei,

Thank you for your reply.

On Wed, 24 Jul 2013 02:08:36 +0800, Wei ZHOU wrote:

I suggest updating the agent on host as well



I run yum install cloudstack-agent on the kvm host, it told me that it's
already the latest version.  I found the post
here:https://issues.apache.org/jira/browse/CLOUDSTACK-3118
Rajesh Battala  mentioned that he installed the latest agent from the
build CloudStack-non-OSS-MASTER-499-rhel6.3.



This is what you run into: 
https://issues.apache.org/jira/browse/CLOUDSTACK-3714



Do you know where can I get that build and how to start the agent if I
have the build(will service cloudstack-agent start do) ?
I didn't find the build in
https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;a=summary



I see you use RHEL/CentOS, with the packaging script in packaging/rhel 
you should be able to build RPM packages from master.


Wido



Any help is appreciated.

Cheers,
Meng

RE: Reformatting UI code

[Animesh Chaturvedi]

> -Original Message-
> From: Chip Childers [mailto:chip.child...@sungard.com]
> Sent: Tuesday, July 23, 2013 9:36 AM
> To: dev@cloudstack.apache.org
> Cc: Jessica Wang; Brian Federle; Pranav Saxena; Ian Duffy
> (i...@ianduffy.ie)
> Subject: Re: Reformatting UI code
> 
> On Tue, Jul 23, 2013 at 09:37:26AM -0600, Mike Tutkowski wrote:
> > One other thought here is that we might want to wait until, say, the
> > day after code freeze - if allowable - and then check in these kinds
> > of changes (basically give them special permission to go in after code
> freeze).
> >
> > I think they're valuable since they do make the code much more
> > readable, but it might be better if they go in once development is
> "frozen."
> 
> The challenge with that, is that sweeping changes like this (or
> architectural changes as well) are best done *early* in a release cycle.
> The challenge we've run into here is that while 4.2 work is proceeding,
> master is open for 4.3 changes (and there is a preference that if
> something big is going to come in, nows the time to do it).
> 
> Perhaps that model is what's broken?
> 
[Animesh>] I concur with Chip, 4.2 is near code freeze and we have a ton of 
blocker and critical to deal with.
> >
> >
> > On Tue, Jul 23, 2013 at 8:42 AM, Sebastien Goasguen
> wrote:
> >
> > >
> > > On Jul 22, 2013, at 6:54 PM, Jessica Wang 
> wrote:
> > >
> > > > Sebastien,
> > > >
> > > > Brian and I are fix 4.2 bugs.
> > > > Any fix for 4.2 bugs should go to both 4.2 branch and master
> branch.
> > > > That's why Brian and I still need to merge our check-in between
> > > > master
> > > branch and 4.2 branch.
> > >
> > > Ok got it .
> > >
> > > Ian next time you send a patch related to the UI maybe you can set
> > > Jessica and Brian as reviewers.
> > >
> > > thanks,
> > >
> > > >
> > > > Jessica
> > > >
> > > > -Original Message-
> > > > From: Sebastien Goasguen [mailto:run...@gmail.com]
> > > > Sent: Monday, July 22, 2013 3:53 PM
> > > > To: Brian Federle
> > > > Cc: Jessica Wang; Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > > dev@cloudstack.apache.org
> > > > Subject: Re: Reformatting UI code
> > > >
> > > >
> > > >
> > > > On 23 Jul 2013, at 00:34, Brian Federle 
> > > wrote:
> > > >
> > > >> In this case it was fine, I could resolve the conflicts by git's
> > > -Xignore-space-change and reformatting accordingly.
> > > >>
> > > >> The main issue was that I didn't see any indication that this was
> > > >> being
> > > committed from reading the discussion thread. Since we're doing a
> > > lot of bugfixing right now for 4.2, a large sweeping commit like
> > > this should at least be posted as part of thread, so that there is
> > > heads-up so that we can prepare any pending changes we have to avoid
> being blocked.
> > > >>
> > > >
> > > > Ok but this is not for 4.2. It was committed to master. How is it
> > > impacting your 4.2 fixes ?
> > > >
> > > >
> > > >> -Brian
> > > >>
> > > >> -Original Message-
> > > >> From: Sebastien Goasguen [mailto:run...@gmail.com]
> > > >> Sent: Monday, July 22, 2013 3:26 PM
> > > >> To: Jessica Wang
> > > >> Cc: Pranav Saxena; Ian Duffy (i...@ianduffy.ie);
> > > dev@cloudstack.apache.org; Brian Federle
> > > >> Subject: Re: Reformatting UI code
> > > >>
> > > >> Hi jessica,
> > > >>
> > > >> Sorry this is causing you trouble. Ian us developing his ldap
> > > >> plugin in
> > > a feature branch .
> > > >> This was a UI cosmetic code change, i saw that pranav shipped it
> > > >> in RB
> > > but did not apply the patch. Since the three if us work in more or
> > > less the same time zone i applied the patch quickly to master. I did
> > > think about other UI work but since 4.2 is in feature freeze i did
> > > not think committing to master would be a problem.
> > > >>
> > > >> Moreover i did not know u were working on UI feature branches.
> > > >> Where
> > > can we look at the feature descriptions ?
> > > >>
> > > >> I suppose u could revert the patch . But for features for 4.3 i
> > > >> think
> > > it should be rebase of your feature branches . I dont think this
> > > change should be impacting 4.2 not sure why u say thats the case.
> > > >>
> > > >> Thoughts ?
> > > >>
> > > >> -Sebastien
> > > >>
> > > >> On 23 Jul 2013, at 00:07, Jessica Wang 
> wrote:
> > > >>
> > > >>> Pranav, Ian, Sebastien,
> > > >>>
> > > >>> The problem is the duration between the time Ian brought it up
> > > >>> on the
> > > mailing list and the time Ian/Sebastien checked in the change to
> > > master branch is too short (less than 4 hours).
> > > >>>
> > > >>> Ian brought it up on the mailing list at Thu 7/18/2013 5:44 AM
> > > >>> (email
> > > subject is "Auto format javascript").
> > > >>> Sebastien checked in Ian's change to master branch at 7/18/2013
> > > >>> 9:34
> > > AM (Commit hash: ad69bc8da3244b783dd003ddf3184fca2762c514).
> > > >>>
> > > >>> This is a big change of UI code.
> > > >>> In GIT's view, every line in JS files has been changed (If you
> > > >>> lo

Re: Review Request 12877: Support HTTP to HTTPS redirect for UCS

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12877/#review23712
---


Review 12877 PASSED the build test
The url of build cloudstack-master-with-patch #26 is : 
http://jenkins.cloudstack.org/job/cloudstack-master-with-patch/26/

- Jenkins Cloudstack.org


On July 23, 2013, 7:56 p.m., Amogh Vasekar wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12877/
> ---
> 
> (Updated July 23, 2013, 7:56 p.m.)
> 
> 
> Review request for cloudstack and Frank Zhang.
> 
> 
> Bugs: https://issues.apache.org/jira/browse/CLOUDSTACK-3285
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> The patch uses EasySSLProtocolFactory from httpclient.contrib.ssl package.
> The implementation accepts self-signed certificates. This is similar to 
> cloud-plugin-network-cisco-vnmc 's behavior.
> As a part of the patch, the contrib code has been moved to cloud-utils to 
> avoid duplication.
> Since there is no maven published jar for contrib, the source currently has 
> the actual code instead. This needs to be revisited.
> 
> 
> Diffs
> -
> 
>   plugins/hypervisors/ucs/src/com/cloud/ucs/manager/UcsHttpClient.java 
> 66d2ed2 
>   
> plugins/network-elements/cisco-vnmc/src/org/apache/commons/httpclient/contrib/ssl/EasySSLProtocolSocketFactory.java
>  52f0ea6 
>   
> plugins/network-elements/cisco-vnmc/src/org/apache/commons/httpclient/contrib/ssl/EasyX509TrustManager.java
>  ae9f938 
>   
> utils/src/org/apache/commons/httpclient/contrib/ssl/EasySSLProtocolSocketFactory.java
>  PRE-CREATION 
>   
> utils/src/org/apache/commons/httpclient/contrib/ssl/EasyX509TrustManager.java 
> PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/12877/diff/
> 
> 
> Testing
> ---
> 
> Works on local environment + UCS
> cisco-vnmc not tested after refactoring
> 
> 
> Thanks,
> 
> Amogh Vasekar
> 
>

Re: Inclusion of commons-httpclient-contrib jar

[Amogh Vasekar]

In the absence of an officially published jar on maven for contrib-ssl,
the current idea is to put this package in cloud-utils (it is a dependency
for both cisco-vnmc and ucs) for easier maintenance.
Please let me know if this sounds good.

Thanks,
Amogh

On 7/22/13 5:26 PM, "Frank Zhang"  wrote:

>I prefer to just use httpclient-contribs.jar as it has been proven by
>lots of users.
>Reimplementing such code brings no gain but maintenance effort.
>
>> -Original Message-
>> From: Amogh Vasekar [mailto:amogh.vase...@citrix.com]
>> Sent: Monday, July 22, 2013 5:17 PM
>> To: dev@cloudstack.apache.org
>> Subject: Inclusion of commons-httpclient-contrib jar
>> 
>> Hi All,
>> 
>> I was working on a HTTPS related fix, and found this package in the
>>codebase :
>> 
>> https://github.com/apache/cloudstack/tree/master/plugins/network-
>> elements/c
>> isco-vnmc/src/org/apache/commons/httpclient/contrib/ssl
>> 
>> I need to use the same classes, which are already a part of
>>httpclient-contribs
>> jar
>> (http://www.java2s.com/Code/Jar/c/Downloadcommonshttpclientcontrib31jar.
>> htm
>> )
>> 
>> Was wondering if there is a specific reason the jar was not included
>>instead
>> (probably since it is a contrib jar?)
>> 
>> Thanks,
>> Amogh
>

Re: Review Request 12810: DnsMasqConfigurator need to be rewrite in bash script

[Sheng Yang]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12810/#review23713
---



patches/systemvm/debian/config/root/dnsmasq.sh


dns2 can be null.



patches/systemvm/debian/config/root/dnsmasq.sh


dhcp-option=tag.* is not in the log



patches/systemvm/debian/config/root/dnsmasq.sh


Backup the failure configures would also help to investigate in the future.



patches/systemvm/debian/config/root/dnsmasq.sh


No, don't override the original result. It failed, then we need to fail 
loudly. The overriding here would cover the fact that it's failed to apply the 
new config.


It's in much better shape than previous edition. Please do more testing, e.g. 
VPC, multiple guest network for user VM. It's likely we can get it in tomorrow.

Thanks!

- Sheng Yang


On July 23, 2013, 1:07 p.m., bharat kumar wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12810/
> ---
> 
> (Updated July 23, 2013, 1:07 p.m.)
> 
> 
> Review request for cloudstack, Alena Prokharchyk and Sheng Yang.
> 
> 
> Bugs: CLOUDSTACK-3694
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> https://issues.apache.org/jira/browse/CLOUDSTACK-3694
> 
> wrote the dnsmasq config in bash instead of creating the config file in java 
> and overwriting. 
> 
> 
> Diffs
> -
> 
>   core/src/com/cloud/agent/api/routing/DnsMasqConfigCommand.java 521ad70 
>   
> core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
> 0b26220 
>   core/src/com/cloud/network/DnsMasqConfigurator.java 3fc61df 
>   patches/systemvm/debian/config/etc/init.d/cloud-early-config c04ff90 
>   patches/systemvm/debian/config/root/dnsmasq.sh b70e2d3 
>   
> plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/resource/VmwareResource.java
>  c7f487e 
>   
> plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
>  f80d4b6 
>   scripts/vm/hypervisor/xenserver/vmops f8c0253 
>   server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java 
> 6c0f7a1 
> 
> Diff: https://reviews.apache.org/r/12810/diff/
> 
> 
> Testing
> ---
> 
> Tested on old master using xenserver.
>  Could not test on the latest one as it is broken.
> 
> 
> Thanks,
> 
> bharat kumar
> 
>

Re: Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[Sheng Yang]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/#review23715
---

Ship it!


Ship It!

- Sheng Yang


On July 23, 2013, 4:01 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12864/
> ---
> 
> (Updated July 23, 2013, 4:01 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3697
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> LB is already deleted when project is cleaned up. Removed it from cleanup 
> queue
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_resources.py 378238e 
> 
> Diff: https://reviews.apache.org/r/12864/diff/
> 
> 
> Testing
> ---
> 
> Tested script with fix locally and it passes now.
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12865: CLOUDSTACK-3693 - Fix more account objects

[Sheng Yang]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12865/#review23716
---

Ship it!


Ship It!

- Sheng Yang


On July 23, 2013, 4:58 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12865/
> ---
> 
> (Updated July 23, 2013, 4:58 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3693
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed more account objects. Fixes CLOUDSTACK-3693
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_egress_rules.py c8b3eee 
> 
> Diff: https://reviews.apache.org/r/12865/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/#review23718
---


Commit cfffe63afb60be77d43448ddcad18baf5f76d781 in branch refs/heads/4.2 from 
Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=cfffe63 ]

CLOUDSTACK-3697 LB rule clean up already done in cleanup project


- ASF Subversion and Git Services


On July 23, 2013, 4:01 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12864/
> ---
> 
> (Updated July 23, 2013, 4:01 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3697
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> LB is already deleted when project is cleaned up. Removed it from cleanup 
> queue
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_resources.py 378238e 
> 
> Diff: https://reviews.apache.org/r/12864/diff/
> 
> 
> Testing
> ---
> 
> Tested script with fix locally and it passes now.
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12865: CLOUDSTACK-3693 - Fix more account objects

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12865/#review23717
---


Commit 7551d8b1a96426a85c35f6b377cd3024feff96c5 in branch refs/heads/4.2 from 
Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=7551d8b ]

CLOUDSTACK-3693 Fix more account objects


- ASF Subversion and Git Services


On July 23, 2013, 4:58 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12865/
> ---
> 
> (Updated July 23, 2013, 4:58 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3693
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed more account objects. Fixes CLOUDSTACK-3693
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_egress_rules.py c8b3eee 
> 
> Diff: https://reviews.apache.org/r/12865/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[Sheng Yang]

> On July 23, 2013, 8:55 p.m., Sheng Yang wrote:
> > Ship It!

Pushed to 4.2 and master.


- Sheng


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/#review23715
---


On July 23, 2013, 4:01 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12864/
> ---
> 
> (Updated July 23, 2013, 4:01 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3697
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> LB is already deleted when project is cleaned up. Removed it from cleanup 
> queue
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_resources.py 378238e 
> 
> Diff: https://reviews.apache.org/r/12864/diff/
> 
> 
> Testing
> ---
> 
> Tested script with fix locally and it passes now.
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12864: CLOUDSTACK-3697 Fix cleanup LB failure

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12864/#review23720
---


Commit c6ea21866e008c467afdcc8b5d505c1c7978e164 in branch refs/heads/master 
from Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=c6ea218 ]

CLOUDSTACK-3697 LB rule clean up already done in cleanup project


- ASF Subversion and Git Services


On July 23, 2013, 4:01 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12864/
> ---
> 
> (Updated July 23, 2013, 4:01 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3697
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> LB is already deleted when project is cleaned up. Removed it from cleanup 
> queue
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_resources.py 378238e 
> 
> Diff: https://reviews.apache.org/r/12864/diff/
> 
> 
> Testing
> ---
> 
> Tested script with fix locally and it passes now.
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

Re: Review Request 12865: CLOUDSTACK-3693 - Fix more account objects

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12865/#review23719
---


Commit 1898fa5046b6b73c85d9d0ed9490156d2278ece8 in branch refs/heads/master 
from Sowmya Krishnan
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=1898fa5 ]

CLOUDSTACK-3693 Fix more account objects


- ASF Subversion and Git Services


On July 23, 2013, 4:58 p.m., Sowmya Krishnan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12865/
> ---
> 
> (Updated July 23, 2013, 4:58 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3693
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Fixed more account objects. Fixes CLOUDSTACK-3693
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_egress_rules.py c8b3eee 
> 
> Diff: https://reviews.apache.org/r/12865/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Sowmya Krishnan
> 
>

RE: DNS Network Domain

[Soheil Eizadi]

Created Jira bug against this, was going to submit a UI patch:
https://issues.apache.org/jira/browse/CLOUDSTACK-3754

-Soheil

From: Soheil Eizadi [seiz...@infoblox.com]
Sent: Tuesday, July 02, 2013 6:36 PM
To: dev@cloudstack.apache.org
Subject: DNS Network Domain

I found that the DNS Network Domain set for Basic Networking is not getting 
passed down to Network Plugins. I am trying to figure out how DNS Network 
Domain is suppose to work for my plugin integration.

I was getting a NULL value for Basic Networking for the DNS Domain even though 
I had set a value for it in the Wizard. After I set the value in the database, 
it seems to be working but wanted to understand why it was not set,I found code 
in the system that enforces this behavior, when I try to set it from the API.
-Soheil

More detail logs .

 The Domain is set in the Zone Setting to acme.com:
> list zones
count = 1
zone:
name = testzone
..
domain = acme.com
..

It does not get passed down to the attached network.

API call "list networks" missing networkdomain field as it is NULL.

> list networks
count = 1
network:
id = 1065fff0-a6e2-410a-94d5-50bc72f879db
name = defaultGuestNetwork
acltype = Domain
broadcastdomaintype = Native
.


I also tried to set it using API to set the value but get exception:

ERROR [cloud.async.AsyncJobManagerImpl] (Job-Executor-20:job-20) Unexpected 
exception while executing 
org.apache.cloudstack.api.command.user.network.UpdateNetworkCmd
com.cloud.exception.InvalidParameterValueException: NetworkOffering and domain 
suffix upgrade can be perfomed for Isolated networks only
at 
com.cloud.network.NetworkServiceImpl.updateGuestNetwork(NetworkServiceImpl.java:1982)
at 
com.cloud.utils.component.ComponentInstantiationPostProcessor$InterceptorDispatcher.intercept(ComponentInstantiationPostProcessor.java:125)

Exception from here:


// network offering and domain suffix can be updated for Isolated 
networks only in 3.0

if ((networkOfferingId != null || domainSuffix != null) && 
network.getGuestType() != GuestType.Isolated) {

throw new InvalidParameterValueException("NetworkOffering and 
domain suffix upgrade can be perfomed for Isolated networks only");

}


I've patched the database and it works for me, trying to understand why we have 
code to prevent it?


> list networks

count = 1

network:

id = 1065fff0-a6e2-410a-94d5-50bc72f879db

name = defaultGuestNetwork

...

networkdomain = acme.com




INFO  [network.element.InfobloxElement] (consoleproxy-1:) InfobloxDeviceElement 
called to prepare Host Name soheil-test with DNS Domain acme.com Gateway null 
Netmask null with MAC 06:97:84:00:00:16 with IPv4 172.16.197.161

Review Request 12883: DNS value in Zone is not getting set in the Network created for the Zone

[Soheil Eizadi]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12883/
---

Review request for cloudstack.


Bugs: https://issues.apache.org/jira/browse/CLOUDSTACK-3754


Repository: cloudstack-git


Description
---

DNS value in Zone is not getting set in the Network created for the Zone. Due 
to this the prepare and release calls to the Network Plugins are missing this 
information for the VMs.

see Jira bug https://issues.apache.org/jira/browse/CLOUDSTACK-3754 for 
additional details.

Administrators-MacBook-Pro-7:cloudstack seizadi$ git diff
diff --git a/ui/scripts/zoneWizard.js b/ui/scripts/zoneWizard.js
index cb92c4e..9885b1d 100755
--- a/ui/scripts/zoneWizard.js
+++ b/ui/scripts/zoneWizard.js
@@ -3084,7 +3084,9 @@
  zoneid: args.data.returnedZone.id,
name: 'defaultGuestNetwork',
displaytext: 
'defaultGuestNetwork',
-   networkofferingid: 
args.data.zone.networkOfferingId
+   networkofferingid: 
args.data.zone.networkOfferingId,
+   networkDomain: 
args.data.returnedZone.domain  
};

//Advanced zone with SG


Diffs
-


Diff: https://reviews.apache.org/r/12883/diff/


Testing
---

Created Basic Network with Wizard and validated with FireFox debugger that 
value was properly set and validated with CloudMonkey that the 
defaultGuestNetwork created had the DNS Domain value set.

I noted that there are other Ajax calls in UI that createNetwork() and don't 
set this field, also there is a use case where the zone value can be modified 
and this value would need to be updated in the attached Networks. So I would 
not close the Jira bug until those additional cases are fixed in the UI.


Thanks,

Soheil Eizadi

RE: [GSOC]cannot add host with the lastest master

[Donal Lafferty]

Hi Meng,

You may also be having problems with your system VM.  Let me explain:

The issue you see is caused by a change in JSON serialisation that happened in 
4.2.  

Previously, we recorded an object's type using it's class' simple name.  Now, 
we are using the full class name, which include the package.  E.g. the 
serializer used to write { "StartupRoutingCommand" : ... }, but now it records 
{"com.cloud.agent.api.com.cloud.agent.api.StartupRoutingCommand":...}

A lot of code compile against 4.1 will not be able to communicate with the 
latest management server.

This means you also have to update your system VM templates.  If you don't, you 
get an error like https://issues.apache.org/jira/browse/CLOUDSTACK-3377

Unfortunately, the default download URLs for system VMs are for 4.1.

If you have not already, download the latest system VMs from 
http://jenkins.cloudstack.org/view/master/job/build-systemvm-master/ 

Cheers,

DL

> -Original Message-
> From: Wido den Hollander [mailto:w...@widodh.nl]
> Sent: 23 July 2013 9:27 PM
> To: dev@cloudstack.apache.org
> Subject: Re: [GSOC]cannot add host with the lastest master
> 
> Hi Meng,
> 
> On 07/23/2013 09:28 PM, Han,Meng wrote:
> > Hi Wei,
> >
> > Thank you for your reply.
> >
> > On Wed, 24 Jul 2013 02:08:36 +0800, Wei ZHOU wrote:
> >> I suggest updating the agent on host as well
> >
> >
> > I run yum install cloudstack-agent on the kvm host, it told me that
> > it's already the latest version.  I found the post
> > here:https://issues.apache.org/jira/browse/CLOUDSTACK-3118
> > Rajesh Battala  mentioned that he installed the latest agent from the
> > build CloudStack-non-OSS-MASTER-499-rhel6.3.
> >
> 
> This is what you run into:
> https://issues.apache.org/jira/browse/CLOUDSTACK-3714
> 
> > Do you know where can I get that build and how to start the agent if I
> > have the build(will service cloudstack-agent start do) ?
> > I didn't find the build in
> > https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;a=summary
> >
> 
> I see you use RHEL/CentOS, with the packaging script in packaging/rhel you
> should be able to build RPM packages from master.
> 
> Wido
> 
> >
> > Any help is appreciated.
> >
> > Cheers,
> > Meng
> >