Agent monitoring status.json of OVSvApp agent always shows as BAD

Bug #1500352 reported by Maruti Kamat
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
networking-vsphere
Fix Committed
High
Maruti Kamat

Bug Description

The content of the file /var/log/neutron/ovsvapp-agent/monitor.log always shows

openvswitch-switch process is not responding. Going to restart
openvswitch-switch process is restarted. Going to restart
ovs: broken

Therefore, the content of the file /var/log/neutron/ovsvapp-agent/status.json is always “{“ovs”: “BAD”}”.

Revision history for this message
Maruti Kamat (marutik) wrote :

This occurs even when ovs-vswitchd process is running.

Changed in networking-vsphere:
assignee: nobody → Maruti Kamat (marutik)
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix proposed to networking-vsphere (master)

Fix proposed to branch: master
Review: https://review.openstack.org/228373

Changed in networking-vsphere:
status: New → In Progress
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to networking-vsphere (master)

Reviewed: https://review.openstack.org/228373
Committed: https://git.openstack.org/cgit/openstack/networking-vsphere/commit/?id=ed865aec49a9c02ac2eef84ae60cc2a7d152664d
Submitter: Jenkins
Branch: master

commit ed865aec49a9c02ac2eef84ae60cc2a7d152664d
Author: Maruti <email address hidden>
Date: Mon Sep 28 02:44:02 2015 -0700

    Fix the status of OVS in agent monitoring

    The patch set fixes the problem where the status of
    OVS is shown as BAD in
    /var/log/neutron/ovsvapp-agent/status.json file even
    when the openvswitch-switch process is actually alive.
    Secondly, ovsvapp-agent-monitor.sh script does not
    restart the openvswitch-switch process if it is not
    running.
    Closes-Bug: 1500352

    Change-Id: I6784e79ebc81401b28331dbe8746ecd54f5db549

Changed in networking-vsphere:
status: In Progress → Fix Committed
Changed in networking-vsphere:
importance: Undecided → High
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Related fix merged to networking-vsphere (master)

Reviewed: https://review.openstack.org/232457
Committed: https://git.openstack.org/cgit/openstack/networking-vsphere/commit/?id=fb45ffd3ec9fc8a91aa02e1824fd354cf330309b
Submitter: Jenkins
Branch: master

commit fb45ffd3ec9fc8a91aa02e1824fd354cf330309b
Author: Romil Gupta <email address hidden>
Date: Thu Oct 8 03:25:25 2015 -0700

    Fix monitor status logs on OVS restart

    When we are mitigating an OVS restart, the agent logs the datapath
    status wrongly when there are not VMs in the cluster.
    This patch addresses the issue by making sure that we mark the status
    as BAD only when there are some ports to be processed.

    Related-Bug: #1500352

    Change-Id: Ie318296a299526bf276e4fee84468f8516308b73
    Co-Authored-By: Sudhakar Babu Gariganti <email address hidden>

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.