Red Hat Bugzilla – Bug 1475305
migration/postcopy: Handle network failures (qemu-kvm-rhev)
Last modified: 2017-12-04 07:37:09 EST
Description of problem:
A failure of the migration network during the postcopy phase of migration is fatal; we can't restart the source since the current state is spread between the two hosts. We want to be able to reconnect the migration stream and complete the migration after someone has fixed the network
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Start a migration
2. switch to postcopy mode
3. take an axe to the network cable
4. wait for TCP timeout
5. Replace network cable
Migration fails after the TCP timeout, the destination VM fails, and can't be restarted.
Some way to recover.
To allow network recovery to happen, we need both the mechanism to do the recover, and also we need to solve the existing monitor hang problem (we need QMP to trigger the recovery on destination side, however existing QMP may hang due to main thread page fault).
The latest work for the migration recovery (there will be a new version very soon though):
The lastest work for the monitor hang solution:
Both series are still discussing upstream. It's hard to make it for 7.5 on both (the recovery part is relatively easier, the monitor part would be harder).
After discussion with Hai, postponing this bug to 7.6 for better luck.
(In reply to Peter Xu from comment #3)
> Both series are still discussing upstream. It's hard to make it for 7.5 on
> both (the recovery part is relatively easier, the monitor part would be
> After discussion with Hai, postponing this bug to 7.6 for better luck.
Also removing from the RPL.