Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
-
None
-
None
-
Reviewed
Description
Was trying to do a perf eval on asyncfswal. I ran w/ these args:
Performance counter stats for '/home/stack/hbase/bin/hbase --config /home/stack/conf_hbase org.apache.hadoop.hbase.wal.WALPerformanceEvaluation -path /user/stack/logs/ -verify -threads 25 -iterations 1000000 -keySize 50 -valueSize 100 -syncInterval 10':
The verify fails on all runs:
Exception in thread "main" java.lang.IllegalStateException: Counted=12390228, expected=25000000
at org.apache.hadoop.hbase.wal.WALPerformanceEvaluation.run(WALPerformanceEvaluation.java:368) at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:76)
at org.apache.hadoop.hbase.wal.WALPerformanceEvaluation.innerMain(WALPerformanceEvaluation.java:597) at org.apache.hadoop.hbase.wal.WALPerformanceEvaluation.main(WALPerformanceEvaluation.java:601)
I need to fix test or figure what is wrong in asyncfswal.
Also seeing these when I run w/ one thread only:
2017-11-28 21:25:49,952 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
Log has a spew of them.
Has stuff like this too:
2017-11-28 21:25:40,065 WARN [Close-WAL-Writer-3] wal.AsyncProtobufLogWriter: normal close failed, try recover java.io.IOException: stream already broken at org.apache.hadoop.hbase.io.asyncfs.FanOutOneBlockAsyncDFSOutput.endBlock(FanOutOneBlockAsyncDFSOutput.java:510) at org.apache.hadoop.hbase.io.asyncfs.FanOutOneBlockAsyncDFSOutput.lambda$close$12(FanOutOneBlockAsyncDFSOutput.java:550) at org.apache.hadoop.hbase.shaded.io.netty.util.concurrent.AbstractEventExecutor.safeExecute(AbstractEventExecutor.java:163)
at org.apache.hadoop.hbase.shaded.io.netty.util.concurrent.SingleThreadEventExecutor.runAllTasks(SingleThreadEventExecutor.java:403) at org.apache.hadoop.hbase.shaded.io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:462)
at org.apache.hadoop.hbase.shaded.io.netty.util.concurrent.SingleThreadEventExecutor$5.run(SingleThreadEventExecutor.java:858) at org.apache.hadoop.hbase.shaded.io.netty.util.concurrent.DefaultThreadFactory$DefaultRunnableDecorator.run(DefaultThreadFactory.java:138)
at java.lang.Thread.run(Thread.java:748)
Starts out spewing EMPTY here:
2017-11-28 21:16:52,051 INFO [main] regionserver.HRegion: Setting FlushNonSloppyStoresFirstPolicy for the region=WALPerformanceEvaluation:0,,1511932610787.deca03e0ca447fa25d02fe9cd6e31aa4.
2017-11-28 21:16:52,058 DEBUG [main] regionserver.HRegion: Found 0 recovered edits file(s) under hdfs://ve0524.halxg.cloudera.com:8020/user/stack/logs/data/WALPerformanceEvaluation/0/deca03e0ca447fa25d02fe9cd6e31aa4
2017-11-28 21:16:52,068 DEBUG [main] regionserver.FlushLargeStoresPolicy: No hbase.hregion.percolumnfamilyflush.size.lower.bound set in description of table WALPerformanceEvaluation:0, use config (134217728) instead
2017-11-28 21:16:52,084 DEBUG [main] wal.WALSplitter: Wrote file=hdfs://ve0524.halxg.cloudera.com:8020/user/stack/logs/data/WALPerformanceEvaluation/0/deca03e0ca447fa25d02fe9cd6e31aa4/recovered.edits/2.seqid, newSeqId=2, maxSeqId=0
2017-11-28 21:16:52,084 INFO [main] regionserver.HRegion: Onlined deca03e0ca447fa25d02fe9cd6e31aa4; next sequenceid=2
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
2017-11-28 21:16:52,185 WARN [AsyncFSWAL-1-1] wal.AsyncFSWAL: RingBufferTruck with unexpected type: EMPTY
...
This is tip of branch-2 minus the commit of the change to the ringbuffer (I'd not pulled in that change).
Apache9 in case you've seen it before else I'll dig in sir.
Attachments
Attachments
Issue Links
- is blocked by
-
HBASE-19372 Remove the Span object in SyncFuture as it is useless now
- Closed
- is broken by
-
HBASE-18601 Update Htrace to 4.2
- Closed
- is related to
-
HBASE-16890 Analyze the performance of AsyncWAL and fix the same
- Closed