yahoo-eng-team team mailing list archive

[OpenStack Infra <2011590@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.opendev.org/c/openstack/neutron/+/877383
Committed: https://opendev.org/openstack/neutron/commit/dea48cfc0a491c63e5a3f899e545b3e8768cbb80
Submitter: "Zuul (22348)"
Branch:    master

commit dea48cfc0a491c63e5a3f899e545b3e8768cbb80
Author: Terry Wilson <twilson@xxxxxxxxxx>
Date:   Tue Mar 14 09:39:32 2023 -0500

    Only create a frozen  Row on matching events
    
    It is very wasteful to create a frozen row copy for every event
    that we process. It can dramatically increase the time to process
    the initial events from connecting to the database.
    
    Closes-bug: #2011590
    Change-Id: Ic4bf26d9b1f937073ddc6d0c3e9d22a777912ebf


** Changed in: neutron
       Status: In Progress => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/2011590

Title:
  Startup times for large OVN dbs is greatly increased by frozen_row()
  calls

Status in neutron:
  Fix Released

Bug description:
  On a large OVN DB, frozen_row() calls for the event notification
  processing can add 10ms up to 40ms or more depending on the size of
  the row (as compared to 4usec for just passing the Row object). This
  can easily make startup times w/ large DBs increase by a factor of 3
  or more.

  We are currently calling frozen_row() for every event, but we really
  only need it for the events that we have registered.

To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/2011590/+subscriptions