hi??all?? I use the HBase version is hbase-0.96.0.This afternoon,I met a problem. dead loop small compact in the region server log.
Regionserver on the log is as follows?? 2015-08-07 16:03:24,406 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HRegion: Starting compaction on info in region qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. 2015-08-07 16:03:24,406 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HStore: Starting compaction of 1 file(s) in info of qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. into tmpdir=hdfs://l-hbase26.data.cn8:8020/hbase/data/default/qproxy_request_info2/fc5207e028a3f5e549bf79c089561793/.tmp, totalSize=555 2015-08-07 16:03:24,417 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.StoreFile: HFile Bloom filter type for f1b919a67a5a4af591c054299329fc88: NONE, but ROW specified in column family configuration 2015-08-07 16:03:24,421 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.StoreFile: HFile Bloom filter type for f1b919a67a5a4af591c054299329fc88: NONE, but ROW specified in column family configuration 2015-08-07 16:03:24,477 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HStore: Completed compaction of 1 file(s) in info of qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. into f1b919a67a5a4af591c054299329fc88(size=555), total size for store is 320.5 M. This selection was in queue for 0sec, and took 0sec to execute. 2015-08-07 16:03:24,477 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.CompactSplitThread: Completed compaction: Request = regionName=qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793., storeName=info, fileCount=1, fileSize=555, priority=-57, time=5089609138060339; duration=0sec 2015-08-07 16:03:24,477 INFO [regionserver60020-smallCompactions-1433845182674] compactions.RatioBasedCompactionPolicy: Deleting the expired store file by compaction: hdfs://l-hbase26.data.cn8:8020/hbase/data/default/qproxy_request_info2/fc5207e028a3f5e549bf79c089561793/info/f1b919a67a5a4af591c054299329fc88 whose maxTimeStamp is -1 while the max expired timestamp is 1438329804477 2015-08-07 16:03:24,477 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HRegion: Starting compaction on info in region qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. 2015-08-07 16:03:24,478 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HStore: Starting compaction of 1 file(s) in info of qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. into tmpdir=hdfs://l-hbase26.data.cn8:8020/hbase/data/default/qproxy_request_info2/fc5207e028a3f5e549bf79c089561793/.tmp, totalSize=555 2015-08-07 16:03:24,488 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.StoreFile: HFile Bloom filter type for 529bd70c10994d09bad8f0298be9ff59: NONE, but ROW specified in column family configuration 2015-08-07 16:03:24,491 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.StoreFile: HFile Bloom filter type for 529bd70c10994d09bad8f0298be9ff59: NONE, but ROW specified in column family configuration 2015-08-07 16:03:24,546 INFO [regionserver60020-smallCompactions-1433845182674] regionserver.HStore: Completed compaction of 1 file(s) in info of qproxy_request_info2,\x00\x00\x01_\x00\xC8\x01g\xA26,1415980522820.fc5207e028a3f5e549bf79c089561793. into 529bd70c10994d09bad8f0298be9ff59(size=555), total size for store is 320.5 M. This selection was in queue for 0sec, and took 0sec to execute. Is this a bug? Any body know how to avoid this situation?
