Collecting findbugs analysis results occasionally causes ssh slave to go offline causing job to abort
Some of our legacy builds have a large number of Findbugs warnings.
Parsing these results in the post build action sometimes causes the slave to go offline, and jenkins quickly brings the slave back online but it causes the job to have failed.
I notice that findbugs outputs a lot of information to the slave.log on the jenkins master
By the time I check, the slave.log has rotated to slave.log.1 but a new slave.log has not been created.
From tailing slave.log on Jenkins master:
list of slave logs: notice how slave.log no longer exists
slave.log has rolled over to slave.log.1
From Jenkins log:
Slave Reconnect from jenkins.log
Log from build: