yahoo-eng-team team mailing list archive

[OpenStack Infra <1655300@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.openstack.org/418409
Committed: https://git.openstack.org/cgit/openstack/horizon/commit/?id=46c4c3d4f993e793d32de65f0cc99ad2b00a2cf1
Submitter: Jenkins
Branch:    master

commit 46c4c3d4f993e793d32de65f0cc99ad2b00a2cf1
Author: Mateusz Kowalski <mateusz.kowalski@xxxxxxx>
Date:   Tue Jan 10 10:10:08 2017 +0100

    If the flavor_id not in full_flavors, don't make a new API request
    
    Flavors are cached from api.nova.flavor_list(), therefore if we have
    machine for which flavor is not in the obtained list, it means the
    machine has been created using flavor which is not available anymore.
    Additional API calls are just decreasing the performance.
    
    Change-Id: I190ae0bd70fc19a9d2d402a808f465c23c842fa7
    Closes-bug: #1655300


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

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Dashboard (Horizon).
https://bugs.launchpad.net/bugs/1655300

Title:
  Low performance when instances are using not available flavour

Status in OpenStack Dashboard (Horizon):
  Fix Released

Bug description:
  When user has a lot of machines using flavour which is not available
  anymore, performance of instance list panel is very bad. The reason is
  the redundant API call in instances/views.py for each machine like
  this.

  Horizon first grabs a list of all available flavours and then iterates
  through all machines. If for a particular one flavour is not on the
  cached list, it sends an API request to nova. I believe it's
  redundant, as if the flavour was not grabbed in the first call, it
  won't appear in any of the following.

To manage notifications about this bug go to:
https://bugs.launchpad.net/horizon/+bug/1655300/+subscriptions