Charan,
After reading the code more carefully, I think your observation is correct.
That is an issue when exceptions are thrown on flushing.
Can you file a JIRA for this? I think the fix should be straightforward by
catching this exception and adding the unflushed logs back to the list.
On Fri,
Charan, will check later today
On Apr 20, 2017 10:55 PM, "Charan Reddy G" wrote:
> Hey,
>
> SyncThread.checkpoint(Checkpoint checkpoint) (which is called periodically
> by SyncThread's executor for every flushInterval) ultimately calls
> EntryLogger.flushRotatedLogs.
>
> In EntryLogger.flushRota
+Sijie, Matteo, Enrico
On Thu, Apr 20, 2017 at 10:55 PM, Charan Reddy G
wrote:
> Hey,
>
> SyncThread.checkpoint(Checkpoint checkpoint) (which is called
> periodically by SyncThread's executor for every flushInterval) ultimately
> calls EntryLogger.flushRotatedLogs.
>
> In EntryLogger.flushRotate
Hey,
SyncThread.checkpoint(Checkpoint checkpoint) (which is called periodically
by SyncThread's executor for every flushInterval) ultimately calls
EntryLogger.flushRotatedLogs.
In EntryLogger.flushRotatedLogs, first we set 'logChannelsToFlush' to null
and then we try to flush and close individual
