My 3 node ensemble produces the following type of warning fairly regularly:
WARN [SyncThread:3:FileTxnLog@334] - fsync-ing the write ahead log in SyncThread:3 took 6943ms which will adversely effect operation latency. See the ZooKeeper troubleshooting guide I understand that this is not a good thing. My questions: 1. How frequently is "too much" for this warning -- ie when should I start worrying? 2. Some of the reported latencies are as high as several minutes(!) (rarely) -- how bad is that? 3. What is the effect of this issue -- is it simply a performance issue or could we possibly see more serious side-effects such as inconsistent state, ephemeral znodes that remain after session expiration, or ??? Thanks for any advice. I suspect that a healthy ensemble would never report this condition .. is that true? Sam Galosh
