-
Notifications
You must be signed in to change notification settings - Fork 6k
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.
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
osd: make the all osd/filestore thread pool suicide timeouts separately configurable #5823
Conversation
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit df4e5de) Conflicts: src/common/config_opts.h place OPTION(osd_command_thread_suicide_timeout, ...) directly after OPTION(osd_command_thread_timeout, ...)
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit 85311b6) Conflicts: src/common/config_opts.h place OPTION(osd_command_thread_suicide_timeout, ...) directly after OPTION(osd_command_thread_timeout, ...)
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit e1073a4)
9d0d3d7
to
9ecac54
Compare
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit f2fbfa3) Conflicts: src/osd/OSD.cc No next_removal_seq(0) in master
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit 3f2946a) Conflicts: src/osd/OSD.cc adapt op_wq() and peering_wq() calls in OSD::OSD src/osd/OSD.h adapt OpWQ instead of ShardedOpWQ
Signed-off-by: Samuel Just <sjust@redhat.com> (cherry picked from commit 547a704) Conflicts: src/osd/OSD.cc firefly still has scrub_finalize_wq
Signed-off-by: Nathan Cutler <ncutler@suse.com>
9ecac54
to
bff2f47
Compare
Waiting for @athanatos review. The biggest anomaly is the last commit bff2f47 . |
@smithfarm could you explain what the anomaly is and why this pull request needs review from Sam ? |
@dachary The anomaly is that firefly has "scrub_finalize_wq" but hammer and above do not have this. On a hunch, I added a commit bff2f47 (which is not in the original master PR) that adds a configurable suicide timeout for this, in the same manner as the other suicide timeouts. However, I don't know enough about this feature to know for sure that this is the right thing to do. Maybe there is someone else who could review it in Sam's absence? |
…ide timeouts separately configurable Reviewed-by: Loic Dachary <ldachary@redhat.com>
@athanatos does this backport look good to merge ? It passed a run of the firefly rados suite ( see http://tracker.ceph.com/issues/11644#note-110 ). Note that there was a valgrind issues but it is unrelated and fixed by #6325. |
…ide timeouts separately configurable Reviewed-by: Loic Dachary <ldachary@redhat.com>
@athanatos ping ? |
Yep, that looks fine. |
make the all osd/filestore thread pool suicide timeouts separately configurable Reviewed-by: Samuel Just <sjust@redhat.com>
http://tracker.ceph.com/issues/11700