|
If you were logged in you would be able to see more operations.
|
|
|
Original Estimate:
|
Unknown
|
Remaining Estimate:
|
Unknown
|
Time Spent:
|
Unknown
|
|
Environment:
|
MySQL 5.x
|
|
The Tungsten replicator periodically seems to miss heartbeat events. The problem appears to be that the wait call on the master site (JMX API ReplicatorManagerMBean.waitForAppliedSequenceNumber() fails to note that we have reached the sequence number, causing a prolonged wait.
The workaround in this case is to send traffic through the master, in which case we recognize that the wait has occurred.
This looks like a race condition related to heartbeat event processing. We also see it in switch operations from the Tungsten Manager. It needs to be fixed as it blocks convenient failover.
|
|
|
Bug
Resolved
Critical
Intermittent issue with master side hang while waiting for heartbeat
