Merge stable/3.0 branch #1
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Our corosync.conf.v2 template was based on a version from SLE HA which had some tunings accidentally dropped in the migration from SLE11. In particular, some of the timeouts were too aggressive. https://bugzilla.suse.com/show_bug.cgi?id=1001164 (cherry picked from commit c1d9c8f)
…able [stable/3.0] fix corosync.conf values (bsc#1001164)
It seems new versions requires a newer version of rack, that requires ruby 2.2.2. See https://travis-ci.org/crowbar/crowbar-ha/jobs/142499249 (cherry picked from commit 5191cfd)
[3.0] pacemaker: Force use of old rack version
Rake 12.0.0 got released on Dec 6 but our rspec version still uses removed code. (cherry picked from commit d7dc20f)
[3.0] travis: Use rake < 12.0.0
It happened after reinstallation one of the compute nodes (which acts as pacemaker-remote). The pacemaker barclamp couldn't apply anymore and it throws the error "Failed to apply the proposal: exception before calling chef (undefined method `sort' for nil:NilClass)" (cherry picked from commit 4c82aa6)
[3.0] Fix undefined method `sort' for nil:NilClass
If the config parameters are changed, it's too risky to just restart the cluster - this could happen on all cluster nodes at a similar time and cause a significant outage. Fortunately it's possible to instead reload the config whilst keeping corosync running, via corosync-cfgtool -R. https://bugzilla.suse.com/show_bug.cgi?id=1001164 (cherry picked from commit a446fe9)
…kport [3.0] don't restart corosync when corosync.conf changes (bsc#1001164)
https://bugzilla.suse.com/show_bug.cgi?id=971771 We only want DRBD started by Pacemaker, with the possible exception of during initial DRBD setup. If it gets started by systemd, systemd will believe it owns the service, in which case during system shutdown it will prematurely shut down DRBD without regard for any of the other services and resources depending on it. Instead we want Pacemaker to stop things in the correct order, even taking care of inter-node dependencies. This fix will not be sufficient by itself, since we still need to address: https://bugzilla.suse.com/show_bug.cgi?id=980341 Thanks to Adam Spiers <[email protected]> (for the fix & for writing the commit message) (cherry picked from commit 52d11b7)
It should only be started by pacemaker (to avoid ownership conflict with systemd), and therefore we should not require drbd to be running to do the setup. This is a followup of 52d11b7 and is required to properly fix https://bugzilla.suse.com/show_bug.cgi?id=971771 (cherry picked from commit fbd12df)
drbd-overview has changed in SP2 and it will not give us the previous "Unconfigured" status We cherry-pick this not because of SP2, but because we don't want to depend on drbd-overview at this point in time (since it requires drbd running). (cherry picked from commit a2a08dc)
Partial cherry-pick from: Adapt to the latest (SP2-based) drbd-overview output (cherry picked from commit c8fa28c)
Due to fbd12df, the drbd daemon may not run (as we let pacemaker start it, and for the first resource, it will therefore not run yet), so drbd-overview will not work. We were using drbd-overview to make sure that all steps are working; without it, we will just fail a bit later. If there's a need to wait for drbd to be ready, there's a new :wait action for the resource. (cherry picked from commit 220caa4)
It's unclear how things could work before without this as it's a mandatory step, see https://www.suse.com/documentation/sle-ha-12/book_sleha/data/sec_ha_drbd_configure.html I can only assume that the drbd service restart was triggering this somehow, and we were lucky to not hit any issue until now. (cherry picked from commit 0525631)
(cherry picked from commit ef5963d)
crowbar-pacemaker: Do not start or restart drbd service (bsc#971771)
vuntz
force-pushed
the
staging-merge-3.0
branch
from
17ee8a3 to
0cf2e65
Compare
matelakat
approved these changes
Feb 22, 2017
toabctl
approved these changes
Sep 8, 2017
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
8 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The only changes compared to the current PTF are DRBD-related (so not relevant for us) and a small bug fix in the rails app.