WSREP: Failed to report last committed (Interrupted system call)

[Laimonas Anusauskas]

Every node in the cluster is logging warnings similar to this:

2015-03-19 01:11:43 18351 [Warning] WSREP: Failed to report last committed 211651504, -4 (Interrupted system call)
2015-03-19 01:11:50 18351 [Warning] WSREP: Failed to report last committed 211652461, -4 (Interrupted system call)
2015-03-19 01:11:54 18351 [Warning] WSREP: Failed to report last committed 211652631, -4 (Interrupted system call)
2015-03-19 02:22:09 18351 [Warning] WSREP: Failed to report last committed 212016107, -4 (Interrupted system call)
2015-03-19 03:55:23 18351 [Warning] WSREP: Failed to report last committed 212530989, -4 (Interrupted system call)
2015-03-19 03:55:32 18351 [Warning] WSREP: Failed to report last committed 212531158, -4 (Interrupted system call)
2015-03-19 06:52:06 18351 [Warning] WSREP: Failed to report last committed 213732197, -4 (Interrupted system call)
2015-03-19 11:01:12 18351 [Warning] WSREP: Failed to report last committed 215646000, -4 (Interrupted system call)

Happens intermittently, about 40-60 times a day on each node. Cluster seems to be operating fine.

CentOS release 6.6 (Final)
Percona-XtraDB-Cluster-server-56-5.6.22-25.8.978.el6.x86_64
Percona-XtraDB-Cluster-galera-3-3.9-1.3494.rhel6.x86_64

Limus

[Erik Lundin (e-erik)]

We are also affected by this

[Przemek (pmalkowski)]

Are you experiencing any real problems when these messages are logged?
Reporting last committed transaction is just a part of certification index purge process. In case it fails for some reason occasionally, the cert index purge may be little delayed. But it does not mean the transaction was not applied successfully. This is a warning after all.

[Laimonas Anusauskas (lanusauskas)]

No, no real problems. But its annoying to see this constantly in the error log.

[Launchpad Janitor (janitor)]

[Expired for Percona XtraDB Cluster because there has been no activity for 60 days.]

[Chris Calender (chriscalender)]

I'm seeing this same error and the cluster is not entirely down, but largely unresponsive until the mysqld node with the problem is killed.

[Walter Heck (walterheck)]

If it's not a problem then either the message should be changed to reflect that, or t should not be logged unless we run in debug mode. I think right now it just scares people unnecessarily because it's a warning (not a notice) and from the message itself it is not at all clear what is up.

2015-09-04 04:17:36 31519 [Warning] WSREP: Failed to report last committed 1670998208, -4 (Interrupted system call)
2015-09-04 04:17:38 31519 [Warning] WSREP: Failed to report last committed 1670998213, -4 (Interrupted system call)
2015-09-04 04:17:41 31519 [Warning] WSREP: Failed to report last committed 1670998600, -4 (Interrupted system call)
2015-09-04 04:17:45 31519 [Warning] WSREP: Failed to report last committed 1670998725, -4 (Interrupted system call)
2015-09-04 04:18:11 31519 [Warning] WSREP: Failed to report last committed 1671001162, -4 (Interrupted system call)
2015-09-04 04:18:14 31519 [Warning] WSREP: Failed to report last committed 1671001284, -4 (Interrupted system call)

[Antonio Falzarano (antonio-falzarano)]

I'm seeing this same error and the cluster is not entirely down, but largely unresponsive until the mysqld node with the problem is killed.

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PXC-1810