cloudstack-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF subversion and git services (JIRA)" <>
Subject [jira] [Commented] (CLOUDSTACK-8545) Unneeded VR reboot after successful live migration
Date Thu, 18 Jun 2015 06:06:00 GMT


ASF subversion and git services commented on CLOUDSTACK-8545:

Commit 85e2d9611f90d00e0d61b2cfd6a1779d400eb60b in cloudstack's branch refs/heads/4.5 from
[;h=85e2d96 ]

CLOUDSTACK-8545 alert when out of band migration is detected and not
acted upon

> Unneeded VR reboot after successful live migration
> --------------------------------------------------
>                 Key: CLOUDSTACK-8545
>                 URL:
>             Project: CloudStack
>          Issue Type: Bug
>      Security Level: Public(Anyone can view this level - this is the default.) 
>    Affects Versions: 4.4.3, 4.5.1
>            Reporter: René Moser
>            Priority: Blocker
>             Fix For: 4.6.0
> The changes committed to fix #7994 are causing a lot of downtimes due unneeded VR reboots,
it fixes an edge case:
> In case of live migration fails and the VR will be rebooted from VMware HA, it VR network
config will get lost.
> This edge case seems to be fixed in a more reliable way in 4.6 with persistent router
config. Live migration on VMware rarely fails with such a small VM like the VR is, with low
RAM and no I/O. The current implementation does a router restart in every case just to make
"sure" config is there. This causes downtime up to 30 minutes in production environments.
> Therefore I suggest to revert the changes in 4.5.2 and 4.6.
> Also see ML thread in dev

This message was sent by Atlassian JIRA

View raw message