Jon,
Re: "RocksDBException: IO error ... No locks available". We have nailed a couple of rebalance issues that could cause the rocksDB lock file to be released not in time recently in trunk, and by checking from your logs I suspect you were hitting one of these issues. Could you try out the current trunk in Kafka and see if they can still be re-produced? Re: "Log end offset should not change while restoring". Unfortunately the stack trace does not reveal which state store's restoration has this issue, could you apply this patch and rerun your application (from trunk + this patch) and send me the stack trace if you encounter this issue again? https://github.com/apache/kafka/pull/2276 Guozhang On Thu, Dec 15, 2016 at 6:54 AM, Jon Yeargers <[email protected]> wrote: > Attached is a debug log showing this exception. > > Question: is it typical to have so many disconnections from brokers? > > This log also includes the exception "Log end offset should not change > while restoring" > > > -- -- Guozhang
