swift-proxy goes into leader-only mode if swift-storage node added but no devices are detected

Bug #1820003 reported by Paul Goins
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
OpenStack Swift Proxy Charm
Triaged
High
Unassigned

Bug Description

As a side effect of lp#1820000 (swift-storage), after adding 2 new swift storage nodes but having them fail to detect disks for use with swift, swift-proxy fell into a state where only the leader node remained active; the other nodes shut down their swift-proxy services as a response to swift-storage-relation-joined hooks, however the cluster-relation-changed hook never got triggered, resulting in those 2 nodes never restarting their swift-proxy services.

Revision history for this message
David Ames (thedac) wrote :

Apologies for the cut and paste fail above. I don't seem to have edit rights.

Guarantee a *cluster-relation-changed hook* gets executed after swift-storage-relation hooks.

Changed in charm-swift-proxy:
status: New → Triaged
importance: Undecided → High
milestone: none → 19.07
David Ames (thedac)
Changed in charm-swift-proxy:
milestone: 19.07 → 19.10
David Ames (thedac)
Changed in charm-swift-proxy:
milestone: 19.10 → 20.01
James Page (james-page)
Changed in charm-swift-proxy:
milestone: 20.01 → 20.05
David Ames (thedac)
Changed in charm-swift-proxy:
milestone: 20.05 → 20.08
James Page (james-page)
Changed in charm-swift-proxy:
milestone: 20.08 → none
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.