Bug #79771 | found a zombie dump thread with the same UUID. Master is killing the zombie dum | ||
---|---|---|---|
Submitted: | 25 Dec 2015 2:26 | Modified: | 3 Jan 2017 7:06 |
Reporter: | lee zhenhua | Email Updates: | |
Status: | Can't repeat | Impact on me: | |
Category: | MySQL Server: Replication | Severity: | S1 (Critical) |
Version: | 5.6.24 | OS: | CentOS (6.6 x86_64) |
Assigned to: | CPU Architecture: | Any |
[25 Dec 2015 2:26]
lee zhenhua
[25 Dec 2015 7:20]
lee zhenhua
on the master host i found this 2015-09-14 14:31:03 3585 [Note] Recovering after a crash using /export/data/binlog/mysql-bin 2015-09-14 14:31:03 3585 [ERROR] Error in Log_event::read_log_event(): 'Event too small', data_len: 0, event_type: 0 2015-09-14 14:31:03 3585 [Note] Starting crash recovery... 2015-09-14 14:31:03 3585 [Note] Crash recovery finished. 2015-09-14 14:31:03 3585 [Note] Crashed binlog file /export/data/binlog/mysql-bin.000008 size is 31975, but recovered up to 31580. B inlog trimmed to 31580 bytes.
[29 Dec 2016 9:58]
MySQL Verification Team
Thank you for the report. Post Bug #72578 fix, a NOTE is added on master's error log file with the following information when log_warnings is greater than 1. "While initializing dump thread for slave with UUID <slave_uuid>, found a zombie dump thread with the same UUID. Master is killing the zombie dump thread." Marking this as duplicate of Bug #84358 for now as Bug #84358 is in verified state.
[3 Jan 2017 7:06]
MySQL Verification Team
This is just to update you that our internal discussions concluded that it is not a duplicate of BUG#84358(which is sort of enhancement to the reported note to fix the confusing error message) this issue is just the expected result of recovery after a crash, and maybe what really needs to know here is that the reason for the crash. This is currently impossible to debug because it is missing lots of information, e.g. replication topology, configuration, and binlogs. Also, this is an expected if you are using bidirectional replication and the master is using sync-binlog=0 and crashes. I would suggest you to reopen the issue if you are seeing crash on latest GA builds i.e 5.6.35/5.7.16 along with the missing detail. If you can provide more information, feel free to add it to this bug and change the status back to 'Open'. Thank you for your interest in MySQL. Thanks, Umesh