MySQL Bugs: #44430: Slave reports 'Lost connection to MyS QL server during query' and IO reconnects

Bug #44430	Slave reports 'Lost connection to MyS QL server during query' and IO reconnects
Submitted:	23 Apr 14:20
Reporter:	Elena Stepanova
Status:	Analyzing
Category:	Server: Replication	Severity:	S3 (Non-critical)
Version:	5.0.81	OS:	Linux
Assigned to:	Elena Stepanova	Target Version:

Description:
After a few hours of stress test, slave started writing batches of 3 messages into error
log, and did so every few minutes until activities on master finished and slave caught up
with master (the latter took a few hours, during which error messages still appeared, even
although nothing was happening on master).

Time interval between consequent error messages was different.

090421 21:51:56 [ERROR] Error reading packet from server: Lost connection to MySQL server
during query ( server_errno=2013)
090421 21:51:56 [Note] Slave I/O thread: Failed reading log event, reconnecting to retry,
log 'master-bin.000026' position 76539832
090421 21:51:56 [Note] Slave: connected to master 'root@127.0.0.1:10730',replication
resumed in log 'master-bin.000026' at position 76539832
090421 21:58:55 [ERROR] Error reading packet from server: Lost connection to MySQL server
during query ( server_errno=2013)
090421 21:58:55 [Note] Slave I/O thread: Failed reading log event, reconnecting to retry,
log 'master-bin.000026' position 474822009
090421 21:58:55 [Note] Slave: connected to master 'root@127.0.0.1:10730',replication
resumed in log 'master-bin.000026' at position 474822009

etc.

The test finished successfully -- that is, result of replication apparently was not
affected by these reconnects.

How to repeat:
no repeatable test case yet

Got these errors while stress testing with 5.4.1-beta build too.
090615 11:31:19 [ERROR] Error reading packet from server: Lost connection to MySQL server
during query ( server_errno=2013)
090615 11:31:19 [Note] Slave I/O thread: Failed reading log event, reconnecting to retry,
log 'master-bin.000030' at postion 2602890
090615 21:11:59 [ERROR] Error reading packet from server: Lost connection to MySQL server
during query ( server_errno=2013)
090615 21:11:59 [Note] Slave I/O thread: Failed reading log event, reconnecting to retry,
log 'master-bin.000136' at postion 3205540

Though it doesn't say 'reconnected and replication resumed' here.

See also bug #44152