[
https://issues.apache.org/jira/browse/CLOUDSTACK-3887?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13729479#comment-13729479
]
Rajesh Battala commented on CLOUDSTACK-3887:
--------------------------------------------
figured out the root cause,
hostPoolRecords is holding all the hosts need to be processed.
when the command is sent to agent to delete the host , its getting processed
for the first host and entering the
if (answer != null && answer.getResult()) {
deleteFlag = true;
break;
}
its breaking from the loop after deleting the storage pool from one host.
"break" is creating issue not getting other hosts id gets processed.
Need to investigate on impact of why "break" is present and how it can handle
other VMM providers.
issue is with the code
for (StoragePoolHostVO host : hostPoolRecords) {
DeleteStoragePoolCommand deleteCmd = new
DeleteStoragePoolCommand(pool);
final Answer answer = agentMgr.easySend(host.getHostId(),
deleteCmd);
if (answer != null && answer.getResult()) {
deleteFlag = true;
break;
} else {
if (answer != null) {
s_logger.debug("Failed to delete storage pool: " +
answer.getResult());
}
}
}
> [KVM] [PrimaryStorage] deleteStoragePool is not removing the storage pool
> information from KVM agent/host.
> ----------------------------------------------------------------------------------------------------------
>
> Key: CLOUDSTACK-3887
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-3887
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Management Server
> Affects Versions: 4.2.0
> Environment: commit # ca474d0e09f772cb22abf2802a308a2da5351592
> Reporter: venkata swamybabu budumuru
> Assignee: Rajesh Battala
> Priority: Blocker
> Fix For: 4.2.0
>
> Attachments: logs.tgz
>
>
> Steps to reproduce:
> 1. Have at least one advanced zone with KVM cluster of 2 hosts.
> 2. make sure everything is working fine.
> 3. add an additional primary storage with scope set to "Cluster"
> 4. verify and make sure that "#virsh pool-list & #virsh dump-xml <pool-id>"
> are showing the newly added one.
> 5. delete the above added primary storage from cloudstack.
> Observations:
> (i) I can see the deleteStoragePool API fired and went fine but, on the KVM
> host side, "virsh pool-list" still shows the primary storage that was added
> in 3rd step.
> (ii) I see that the HeartBeat is continuously written to the primary and it
> still considers that this primary is up.
> Attaching all the mgmt server, agent logs and db dump to the bug.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
