yahoo-eng-team team mailing list archive

[OpenStack Infra <1884596@xxxxxxxxxxxxxxxxxx>]

Reviewed:  https://review.opendev.org/743597
Committed: https://git.openstack.org/cgit/openstack/glance/commit/?id=3f6e349d0853a9746d0d744bc3eb0b2baa1ddff9
Submitter: Zuul
Branch:    master

commit 3f6e349d0853a9746d0d744bc3eb0b2baa1ddff9
Author: Dan Smith <dansmith@xxxxxxxxxx>
Date:   Tue Jul 28 09:02:13 2020 -0700

    Implement time-limited import locking
    
    This attempts to provide a time-based import lock that is dependent
    on the task actually making progress. While the task is copying
    data, the task message is updated, which in turn touches the task
    updated_at time. The API will break any lock after 30 minutes of
    no activity on a stalled or dead task. The import taskflow will
    check to see if it has lost the lock at any point, and/or if its
    task status has changed and abort if so.
    
    The logic in more detail:
    
    1. API locks the image by task-id before we start the task thread, but
       before we return
    2. Import thread will check the task-id lock on the image every time it
       tries to modify the image, and if it has changed, will abort
    3. The data pipeline will heartbeat the task every minute by updating
       the task.message (bonus: we get some status)
    4. If the data pipeline heartbeat ever finds the task state to be changed
       from the expected 'processing' it will abort
    5. On task revert or completion, we drop the task-id lock from the image
    6. If something ever gets stuck or dies, the heartbeating will stop
    7. If the API gets a request for an import where the lock is held, it
       will grab the task by id (in the lock) and check the state and age.
       If the age is sufficiently old (no heartbeating) and the state is
       either 'processing' or terminal, it will mark the task as failed,
       steal the lock, and proceed.
    
    Lots of logging throughout any time we encounter unexpected situations.
    
    Closes-Bug: #1884596
    Change-Id: Icb3c1d27e9a514d96fca7c1d824fd2183f69d8b3


** Changed in: glance
       Status: In Progress => 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/1884596

Title:
  image import copy-image will start multiple importing threads due to
  race condition

Status in Glance:
  Fix Released

Bug description:
  I'm filing this bug a little prematurely because Abhi and I didn't get
  a chance to fully discuss it. However, looking at the code and the
  behavior I'm seeing due to another bug (1884587), I feel rather
  confident.

  Especially in a situation where glance is running on multiple control
  plane nodes (i.e. any real-world situation), I believe there is a race
  condition whereby two closely-timed requests to copy an image to a
  store will result in two copy operations in glance proceeding in
  parallel. I believe this to be the case due to a common "test-and-set
  that isn't atomic" error.

  In the API layer, glance checks that an import copy-to-store operation
  isn't already in progress here:

  https://github.com/openstack/glance/blob/e6db0b10a703037f754007bef6f56451086850cd/glance/api/v2/images.py#L167

  And if that passes, it proceeds to setup the task as a thread here:

  https://github.com/openstack/glance/blob/e6db0b10a703037f754007bef6f56451086850cd/glance/api/v2/images.py#L197

  which may start running immediately or sometime in the future. Once
  running, that code updates a property on the image to indicate that
  the task is running here:

  https://github.com/openstack/glance/blob/e6db0b10a703037f754007bef6f56451086850cd/glance/async_/flows/api_image_import.py#L479-L484

  Between those two events, if another API user makes the same request,
  glance will not realize that a thread is already running to complete
  the initial task and will start another. In a situation where a user
  spawns a thousand new instances to a thousand compute nodes in a
  single operation where the image needs copying first, it's highly
  plausible to have _many_ duplicate glance operations going, impacting
  write performance on the rbd cluster at the very least.

  As evidence that this can happen, we see an abnormally extended race
  window because of the aforementioned bug (1884587) where we fail to
  update the property that indicates the task is running. In a test we
  see a large number of them get started, followed by a cascade of
  failures when they fail to update that image property, implying that
  many such threads are running. If this situation is allowed to happen
  when the property does *not* fail to update, I believe we would end up
  with glance copying the image to the destination in multiple threads
  simultaneously. That is much harder to simulate in practice in a
  development environment, but the other bug makes it happen every time
  since we never update the image property to prevent it and thus the
  window is long.

  Abhi also brought up the case where if this race occurs on the same
  node, the second attempt *may* actually start copying the partial
  image in the staging directory to the destination, finish early, and
  then mark the image as "copied to $store" such that nova will attempt
  to use the partial image immediately, resulting in a corrupted disk
  and various levels of failure after that. Note that it's not clear if
  that's really possible or not, but I'm putting it here so the glance
  gurus can validate.

  The use of the os_glance_importing_to_stores property to "lock" a copy
  to a particular store is good, except that updating that list
  atomically means that the above mentioned race will not have anything
  to check after the update to see if it was the race loser. I don't see
  any checks in the persistence layer to ensure that an UPDATE to the
  row with this property doesn't already have a given store in it, or do
  any kind of merge. This also leads me to worry that two parallel
  requests to copy an image to two different stores may result in
  clobbering the list of stores-in-progress and potentially also the
  final list of stores at rest. This is just conjecture at this point, I
  just haven't seen anywhere that situation is accounted for.

To manage notifications about this bug go to:
https://bugs.launchpad.net/glance/+bug/1884596/+subscriptions