Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/24637 )
Change subject: KUDU-3751: master steps down instead of crashing on catch-up timeout ...................................................................... Patch Set 6: > > (2 comments) > > That's a fair point, I hadn't fully considered the "runs forever > but never actually catches up" failure mode, especially for the > newly-added-master scenario you're describing. I agree that's > genuinely worse than the original crash in the way you describe: it > looks healthy while silently never doing its job. > > I don't think reverting to the original LOG(FATAL) abort is the > right fix either, though, since that behavior is what the ticket > was filed against in the first place. Would a controlled, graceful > shutdown after the catch-up timeout (clear log message, clean exit, > rather than LOG(FATAL)'s abort/crash-dump) address the concern? > That gives operators/process supervisors an unambiguous "this > master stopped and needs attention" signal, without either the > crash-dump noise of the original behavior or the silent-zombie > problem of my current patch. Yes -- graceful shutdown in that case is the best course of action, I guess. Quick and dirty exit (crashing with SIGABRT) was a shortcut: it was non-ambiguous and safe regarding adding a new master to a cluster, but it was too rough as for user/operator experience -- that's why the TODO, IIUC. -- To view, visit http://gerrit.cloudera.org:8080/24637 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I4c9c5a3a6ec43c5174f8d8e92e2421029c51ea01 Gerrit-Change-Number: 24637 Gerrit-PatchSet: 6 Gerrit-Owner: Jeison Romero <[email protected]> Gerrit-Reviewer: Abhishek Chennaka <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Gabriella Lotz <[email protected]> Gerrit-Reviewer: Jeison Romero <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Marton Greber <[email protected]> Gerrit-Comment-Date: Wed, 29 Jul 2026 23:46:57 +0000 Gerrit-HasComments: No
