HAHT - Fail to restart datasources node

Bug #1604845 reported by Masahito Muroi
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
congress
Fix Released
High
Anusha

Bug Description

Restarting datasource node fails because datasource policy is in used when policy engine and datasource driver are deployed on separate nodes.

Policy engine has datasource type policy after starting subscribe datasource table. The policy remains in policy engine event though the datasource node goes down. Then the datasource node fails to re-start since policy engine reports the name of datasource policy is in used.

This bug affects datasource node HA. When datasource node fails over warm-standby node, it fails to
start.

How to reproduce:

1. deploy 2 node, one has policy engine and another has datasource
$ congress-server --api --engine --node_id=engine-node
$ congress-server --datasources --node_id=ds-node

2. stop datasource node
3. restart datasource node

Changed in congress:
importance: Undecided → High
Anusha (anusha-iiitm)
Changed in congress:
assignee: nobody → Anusha (anusha-iiitm)
Revision history for this message
Anusha (anusha-iiitm) wrote :

I think this bug would be resolved by synchronizer patch as datasource policies creation is handled by respective PE's now. I'll double check it to verify it solves the issue. Thanks

Changed in congress:
status: New → In Progress
Revision history for this message
Masahito Muroi (muroi-masahito) wrote :

I check the synchronizer patch. It looks the patch can almost resolve this issue.

There is still one corner case as following. If datasource node restarts faster than the time window of synchronizer it could happen. But I think the patch don't need to resolve the corner case as well in one patch.

Possible situation:
 1. synchronizer in PE works
 2. datasource node fails
 3. datasource node re-start

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to congress (master)

Reviewed: https://review.openstack.org/324328
Committed: https://git.openstack.org/cgit/openstack/congress/commit/?id=0c6aafec9c5014398d4ec50ca06ec9f68e094d05
Submitter: Jenkins
Branch: master

commit 0c6aafec9c5014398d4ec50ca06ec9f68e094d05
Author: Anusha Ramineni <email address hidden>
Date: Thu Jun 2 11:40:49 2016 +0530

    Migrate synchronizer to DSE2

    Closes-Bug: #1585576
    Closes-Bug: #1604845
    Change-Id: I0717b28b1578ef281c430814b32cd5ff5e86746f

Changed in congress:
status: In Progress → Fix Released
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix included in openstack/congress 4.0.0.0b3

This issue was fixed in the openstack/congress 4.0.0.0b3 development milestone.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.