If you get an IOException like in the description of the jira while persisting a txn in SyncRequestProcessor, then the leader should experience the issue in the jira:
2015-08-14 15:41:18,556 [myid:100] - ERROR [SyncThread:100:ZooKeeperCriticalThread@48] - Severe unrecoverable error, from thread : SyncThread:100 java.io.IOException: Input/output error -Flavio > On 03 Aug 2016, at 17:08, Luciano Afranllie <[email protected]> wrote: > > Hi > > In this issue https://issues.apache.org/jira/browse/ZOOKEEPER-2247 > > There is a comment that says: > > In test environment, after the zookeeper server start, *disk is made > read-only using an internal tool which executes commands similar to mount > -o ro,remount /dev/sda3. yes, it is not a disk full scenario* > > > What I would like to know if this issue happens in disk full scenarios. > > > Regards > > Luciano
