Bug 1417801 - VMs snapshots should happen before setting the geo-replication checkpoint
Summary: VMs snapshots should happen before setting the geo-replication checkpoint
Alias: None
Product: ovirt-engine
Classification: oVirt
Component: BLL.Gluster
Version: 4.1.0
Hardware: x86_64
OS: Linux
Target Milestone: ---
: ---
Assignee: Sahina Bose
Depends On:
TreeView+ depends on / blocked
Reported: 2017-01-31 05:30 UTC by SATHEESARAN
Modified: 2017-01-31 06:01 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Last Closed: 2017-01-31 06:00:31 UTC
oVirt Team: Gluster

Attachments (Terms of Use)

Description SATHEESARAN 2017-01-31 05:30:34 UTC
Description of problem:
With the current feature of remote data sync setup for glusterfs data domain, geo-replication checkpoint is initiated first, then VM snapshots are triggered.

Setting up the geo-replication checkpoint only provides the guarantee that the data up to the checkpoint will be synced to remote volume. Geo-rep could sync little or less more than the checkpoint but with no guarantee.

With the above said property in mind, VM snapshots should be triggered initially followed by gluster geo-replication checkpoint. This provides the guaranteed sync  of base-image to the remote volume.

Version-Release number of selected component (if applicable):
oVirt Engine Version:

How reproducible:

Steps to Reproduce:
1. Setup the 'Remote Data sync Setup' for the glusterfs data domain

Actual results:
The events shows that the geo-rep checkpoint is done first, then followed by the VM snapshots.

Expected results:
VMs snapshot should happen first followed by geo-rep checkpoint creation.

Comment 1 SATHEESARAN 2017-01-31 06:00:31 UTC
I misread the events in the 'EVENTS' tab.
It states the snapshot deletion, I misread it for snapshot creation.

This issue is no longer a bug

Note You need to log in before you can comment on or make changes to this bug.