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.

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

Performance: Only scan for omap corruption once #17705

Merged
merged 4 commits into from Sep 15, 2017
Merged

Conversation

dzafman
Copy link
Contributor

@dzafman dzafman commented Sep 13, 2017

Fix hard-coded "leveldb" backend.  The command is broken in Luminous
now that "rocksdb" is the default.

Signed-off-by: David Zafman <dzafman@redhat.com>
Signed-off-by: David Zafman <dzafman@redhat.com>
Available for testing and user downgrade.

Signed-off-by: David Zafman <dzafman@redhat.com>
Before
 	state 2: Can have complete tables (some may be bad)
	state 3: Never had complete tables
After
 	state 2: Can have complete tables (some may be bad)
	state 3 with legacy: Can have complete tables (bad ones are cleared)
	state 3: Never had complete tables

Once OSDs boot with this change you can't downgrade to a previous release.
If someone does downgrade they could have unstable OSDs that hit assert(state.v < 3).
The following command run after shutting down the cluster but before downgrading
ceph packages would be a way to fix this.

ceph-osdomap-tool --omap-path ... --command resetv2

Fixes: http://tracker.ceph.com/issues/21328

Signed-off-by: David Zafman <dzafman@redhat.com>
Copy link
Member

@jdurgin jdurgin left a comment

Choose a reason for hiding this comment

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

nice catch!

@smithfarm
Copy link
Contributor

It seems this was merged after the luminous split. Opened luminous backport PR: #18301

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
4 participants