-
Notifications
You must be signed in to change notification settings - Fork 3.8k
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
rpc: panic in per-peer metrics #105335
Comments
Hi @tbg, please add a C-ategory label to your issue. Check out the label system docs. While you're here, please consider adding an A- label to help keep our repository tidy. 🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is dev-inf. |
cc @cockroachdb/replication |
It looks like it was trying to reach n3 on 2 different IP addresses, fwiw. It paniced when it was trying the second one.
|
Hi @erikgrinaker, please add branch-* labels to identify which branch(es) this release-blocker affects. 🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is dev-inf. |
I think the problem is here: Lines 860 to 867 in caaa7d8
The following can happen:
Should be easy peasy to fix. |
107391: roachtest: include link to testeng grafana in issue posts r=smg260,tbg a=annrpom This adds a link, populated with relevant cluster name and test timeframe, to the testeng grafana instance for failed roachtests. Fixes: #105894 Release note: None 107659: serverutils: provide SQLConn/SQLConnE in ApplicationLayerInterface r=stevendanna a=knz Fixes #107672. Part of solving #107058. Informs #106772. Epic: CRDB-18499 107697: rpc: avoid crash in newPeer r=erikgrinaker a=tbg It was previously possible to make a new peer while the old one was in the middle of being deleted, which caused a crash due to to acquiring child metrics when they still existed. Luckily, this is easy enough to fix: just remove some premature optimization where I had tried to be too clever. Fixes #105335. Epic: CRDB-21710 Release note: None (bug never released) 107721: asim: skip TestAllocatorSimulatorDeterministic and example_fulldisk r=wenyihu6 a=wenyihu6 We found some non-deterministic behavior in the allocator simulator (see #105904 for more details). For now, we are skipping these potentially flaky tests. Release Note: None Epic: None 107728: persistedsqlstats: specify background qos for compaction job r=xinhaoz a=xinhaoz The compaction job can be an expensive operation so we should de-prioritize it with the `UserLow` qos setting. Fixes: #99949 Release note: None 107750: ui: fix app = empty string filter on stmts page r=xinhaoz a=xinhaoz The filter on app name = empty string was not working on the stmts page. This was due to the fact that we use (unset) as the option in the filter to represent selecting the empty string app name. However when filtering statements, the empty string app name on the stmt was not changed accordingly. this commit fixes this and also adds testing for the unset case. Epic: none Fixes: #107748 Release note (bug fix): Filter on stmts page works for app name = empty string (represented as 'unset'). https://www.loom.com/share/2fee4f0fb7b04208803e0dac1d9694ab?sid=5cabecf9-1c2a-406b-89a8-b378ed07d329 107753: backupccl: deflake TestBackupAndRestoreJobDescription r=stevendanna a=adityamaru This change sorts the jobs based on when they were created to ensure we get a stable sort of job descriptions. Fixes: #107684 Release note: None Co-authored-by: Annie Pompa <[email protected]> Co-authored-by: Raphael 'kena' Poss <[email protected]> Co-authored-by: Tobias Grieger <[email protected]> Co-authored-by: wenyihu6 <[email protected]> Co-authored-by: Xin Hao Zhang <[email protected]> Co-authored-by: adityamaru <[email protected]>
cockroach/pkg/rpc/peer.go
Line 178 in caaa7d8
@tbg made some recent changes here in #99191.
Originally posted by @erikgrinaker in #105260 (comment)
Jira issue: CRDB-28987
The text was updated successfully, but these errors were encountered: