Skip to content

osd/PG: do not use approx_missing_objects pre-nautilus#27798

Merged
neha-ojha merged 1 commit intoceph:masterfrom
neha-ojha:wip-39441
Apr 26, 2019
Merged

osd/PG: do not use approx_missing_objects pre-nautilus#27798
neha-ojha merged 1 commit intoceph:masterfrom
neha-ojha:wip-39441

Conversation

@neha-ojha
Copy link
Member

We changed async recovery cost calculation in nautilus to also take
into account approx_missing_objects in ab241bf
This commit depends on #23663, hence
wasn't backported to mimic.

Mimic only uses the difference in length of logs as the cost. Due to this,
the same OSD might have different costs in a mixed mimic and nautilus(or above)
cluster. This can lead to choose_acting() cycling between OSDs, when trying
to select the acting set and async_recovery_targets.

Fixes: https://tracker.ceph.com/issues/39441
Signed-off-by: Neha Ojha nojha@redhat.com

  • References tracker ticket
  • Updates documentation if necessary
  • Includes tests for new functionality or reproducer for bug

We changed async recovery cost calculation in nautilus to also take
into account approx_missing_objects in ab241bf
This commit depends on ceph#23663, hence
wasn't backported to mimic.

Mimic only uses the difference in length of logs as the cost. Due to this,
the same OSD might have different costs in a mixed mimic and nautilus(or above)
cluster. This can lead to choose_acting() cycling between OSDs, when trying
to select the acting set and async_recovery_targets.

Fixes: https://tracker.ceph.com/issues/39441
Signed-off-by: Neha Ojha <nojha@redhat.com>
@neha-ojha
Copy link
Member Author

Copy link
Member

@xiexingguo xiexingguo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for catching this, Neha!

@neha-ojha
Copy link
Member Author

No related failures in 20 runs of the reproducer: http://pulpito.ceph.com/nojha-2019-04-25_05:40:23-rados:upgrade:mimic-x-singleton-wip-39441-distro-basic-smithi/

Dead jobs:
3891898 - PGs are not stuck in remapped+peering or peering
3891903 - infra

@neha-ojha neha-ojha requested review from jdurgin and liewegas April 26, 2019 16:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants