If the private IP address is dropped from an interface (or the interface is simply taken down) then CTDB does not recover. This is because a bind failure for an outgoing connection does not trigger a retry. There are some other less likely possibly transient failures that are also not retried.
Created attachment 15787 [details] Patch for 4.12, 4.11, 4.10 Commit cherry-picks cleanly into 4.12 and the resulting patch then applied to 4.11 and 4.10.
Hi Karolin, This is ready for v4-12, v4-11 and v4-10. Thanks.
(In reply to Amitay Isaacs from comment #2) Pushed to autobuild-v4-{12,11,10}-test.
(In reply to Karolin Seeger from comment #3) Pushed to all branches. Closing out bug report. Thanks!