yahoo-eng-team team mailing list archive

[Steve Lewis <steve.lewis@xxxxxxxxxxxxx>]

Small revision: I'm gonna go with this issue is fully fixed rather than just mostly fixed.
Change-Id as noted prior.

** Changed in: glance
       Status: Triaged => Fix Released

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to Glance.
https://bugs.launchpad.net/bugs/1179777

Title:
  glance-cache-cleaner does not respect image_cache_stall_time

Status in OpenStack Image Registry and Delivery Service (Glance):
  Fix Released

Bug description:
  We're running a Folsom deployment within the NeCTAR Research Cloud.
  Currently the cloud has a central glance-registry and we run a couple
  of glance-api servers for local API handling and caching. I recently
  configured cron on the glance-api servers to do periodic cache
  cleaning, initially at a high frequency to see how it works/ed.

  I was very surprised to see the cleaner remove an image as it was
  actively caching into $image_cache_dir/incomplete/. Upon further
  inspection I found the sqlite driver for the cleaner code calculates
  an expiry time but never uses it when looping over the list of
  incomplete images and unlinking them. This bug appears to be present
  in the current trunk too (and presumably Grizzly).

  I have a fix (which compares the older_than parameter to the atime on
  each returned path) that I'll be happy to contribute as my first basic
  contribution...

To manage notifications about this bug go to:
https://bugs.launchpad.net/glance/+bug/1179777/+subscriptions