incubator-cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ilya musayev (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CLOUDSTACK-1411) Issues with VMWare Hypervisor host_ids not updated when ESX(i) crashes in instance table
Date Thu, 28 Feb 2013 17:11:12 GMT

    [ https://issues.apache.org/jira/browse/CLOUDSTACK-1411?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13589670#comment-13589670
] 

ilya musayev commented on CLOUDSTACK-1411:
------------------------------------------

Mice,

That is correct, issue most likely does not exists on 4.2 due to the work on CLOUDSTACK-669,
which does better vmware sync and is currently commited to a master (i suppose). The CLOUDSTACK-669
code missed the review and submission of January 31 deadline and hence wont be making it into
4.1

However, current production release 4.0 and soon to be released 4.1 are not getting the CS-669
patch (since it missed the review and commit deadline). We need a fix to address this issue
only.

I do see that CS is trying to look for a VM on VCenter (MS log has something like "Looking
for VM ..." , the DB gets updated *sometimes*, since last_host_id and host_id in instance
table sometimes are different. However, in the case outlined above when hypervisor fails,
CS fails to detect the new hypervisor host_id eventhough lookup is performed.

Please retry this on 4.1

Thanks
ilya




                
> Issues with VMWare Hypervisor host_ids not updated when ESX(i) crashes in instance table
> ----------------------------------------------------------------------------------------
>
>                 Key: CLOUDSTACK-1411
>                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-1411
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the default.) 
>          Components: Hypervisor Controller
>    Affects Versions: 4.0.0, 4.1.0
>         Environment: CS 4.0 and 4.1, VMWare vSphere 5.x
>            Reporter: ilya musayev
>            Priority: Blocker
>              Labels: esxi, hypervisor, vsphere
>
> As noted previously by several users on cloudstack-users mailing list, the host_id are
not updated when VM is moved from one hypervisor to another during hypervisor crash. The subsequent
tasks that require proper host_id then fail going forward. 
> Please reference this thread for more details:
> http://markmail.org/message/byfdmxt3vf2txt5d

--
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

Mime
View raw message