Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

librbd: fix handling of proxied maintenance operations during shut down #8565

Merged
merged 4 commits into from Apr 14, 2016
Merged

librbd: fix handling of proxied maintenance operations during shut down #8565

merged 4 commits into from Apr 14, 2016

Commits on Apr 12, 2016

  1. librbd: gracefully handle blacklisted client in ImageWatcher 

    Signed-off-by: Jason Dillaman <dillaman@redhat.com>
    Jason Dillaman committed Apr 12, 2016
    Configuration menu
    Copy the full SHA
    1761a38 View commit details
    Browse the repository at this point in the history

  2. rbd: cleanly fail bench-write upon failure 

    Fixes: http://tracker.ceph.com/issues/15456
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
    Jason Dillaman committed Apr 12, 2016
    Configuration menu
    Copy the full SHA
    1e6fc45 View commit details
    Browse the repository at this point in the history

  3. librbd: delay invalidation of exclusive lock pointer 

    Certain IO and maintenance ops code paths have an expectation
that the exclusive lock pointer will be valid while in-flight.
Let the exclusive lock state machine clean up the pointer after
it has flushed all IO and canceled all ops.

Fixes: http://tracker.ceph.com/issues/15471
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
    Jason Dillaman committed Apr 12, 2016
    Configuration menu
    Copy the full SHA
    e90e3d9 View commit details
    Browse the repository at this point in the history

  4. librbd: cancel ops before blocking IO 

    Ops might have in-flight IO -- blocking IO after canceling the ops
will result in the in-flight IO being flushed.  Shutdown also requires
an intermediate state where is still acts like it downs the lock
until after all ops are canceled and all IO is flushed.

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
    Jason Dillaman committed Apr 12, 2016
    Configuration menu
    Copy the full SHA
    8451c0f View commit details
    Browse the repository at this point in the history