Details
-
Bug
-
Status: Resolved
-
Major
-
Resolution: Won't Fix
-
None
-
None
-
None
Description
I have a Master that crashed on a large cluster with hundreds of outstanding Procedure WALs and (probably --TBD) a few million Procedures to load. It is taking a long time (two hours)...
There were STUCK procedures that were preventing clean-up of the old WALs.
I can tell we are making progress by enabling TRACE on the Procedure Store. Better would be an emission as we made progress through the files with an output after every so many procedures loaded.
Then, post-load, there is a long time spent sorting-out the Procedure image... We are in finish for ages doing stuff like:
"master/vc0207:22001:becomeActiveMaster" #98 daemon prio=5 os_prio=0 tid=0x0000000000d31800 nid=0x1efc0 runnable [0x00007f0a3c17d000] java.lang.Thread.State: RUNNABLE at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader$WalProcedureMap.removeFromMap(ProcedureWALFormatReader.java:837) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader$WalProcedureMap.fetchReady(ProcedureWALFormatReader.java:614) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader.finish(ProcedureWALFormatReader.java:201) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormat.load(ProcedureWALFormat.java:94) at org.apache.hadoop.hbase.procedure2.store.wal.WALProcedureStore.load(WALProcedureStore.java:426) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.load(ProcedureExecutor.java:382) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.init(ProcedureExecutor.java:663) at org.apache.hadoop.hbase.master.HMaster.createProcedureExecutor(HMaster.java:1335) at org.apache.hadoop.hbase.master.HMaster.finishActiveMasterInitialization(HMaster.java:878) at org.apache.hadoop.hbase.master.HMaster.startActiveMasterManager(HMaster.java:2119) at org.apache.hadoop.hbase.master.HMaster.lambda$run$0(HMaster.java:567) at org.apache.hadoop.hbase.master.HMaster$$Lambda$42/1930759883.run(Unknown Source) at java.lang.Thread.run(Thread.java:748)
and....
"master/vc0207:22001:becomeActiveMaster" #98 daemon prio=5 os_prio=0 tid=0x0000000000d31800 nid=0x1efc0 runnable [0x00007f0a3c17d000] java.lang.Thread.State: RUNNABLE at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:423) at java.security.Provider$Service.newInstance(Provider.java:1595) at sun.security.jca.GetInstance.getInstance(GetInstance.java:236) at sun.security.jca.GetInstance.getInstance(GetInstance.java:164) at java.security.Security.getImpl(Security.java:695) at java.security.MessageDigest.getInstance(MessageDigest.java:167) at org.apache.hadoop.hbase.util.MD5Hash.getMD5AsHex(MD5Hash.java:59) at org.apache.hadoop.hbase.client.RegionInfo.createRegionName(RegionInfo.java:560) at org.apache.hadoop.hbase.client.RegionInfo.createRegionName(RegionInfo.java:490) at org.apache.hadoop.hbase.client.RegionInfoBuilder$MutableRegionInfo.<init>(RegionInfoBuilder.java:243) at org.apache.hadoop.hbase.client.RegionInfoBuilder.build(RegionInfoBuilder.java:120) at org.apache.hadoop.hbase.shaded.protobuf.ProtobufUtil.toRegionInfo(ProtobufUtil.java:3132) at org.apache.hadoop.hbase.master.procedure.ServerCrashProcedure.deserializeStateData(ServerCrashProcedure.java:335) at org.apache.hadoop.hbase.procedure2.ProcedureUtil.convertToProcedure(ProcedureUtil.java:283) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader$Entry.convert(ProcedureWALFormatReader.java:359) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader$EntryIterator.next(ProcedureWALFormatReader.java:410) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.loadProcedures(ProcedureExecutor.java:460) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.access$200(ProcedureExecutor.java:76) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor$1.load(ProcedureExecutor.java:391) at org.apache.hadoop.hbase.procedure2.store.wal.WALProcedureStore$2.load(WALProcedureStore.java:441) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormatReader.finish(ProcedureWALFormatReader.java:202) at org.apache.hadoop.hbase.procedure2.store.wal.ProcedureWALFormat.load(ProcedureWALFormat.java:94) at org.apache.hadoop.hbase.procedure2.store.wal.WALProcedureStore.load(WALProcedureStore.java:426) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.load(ProcedureExecutor.java:382) at org.apache.hadoop.hbase.procedure2.ProcedureExecutor.init(ProcedureExecutor.java:663) at org.apache.hadoop.hbase.master.HMaster.createProcedureExecutor(HMaster.java:1335) at org.apache.hadoop.hbase.master.HMaster.finishActiveMasterInitialization(HMaster.java:878) at org.apache.hadoop.hbase.master.HMaster.startActiveMasterManager(HMaster.java:2119) at org.apache.hadoop.hbase.master.HMaster.lambda$run$0(HMaster.java:567) at org.apache.hadoop.hbase.master.HMaster$$Lambda$42/1930759883.run(Unknown Source)
In this case, it took 1.25 hours before we starting showing progress in the log. Let me add progress output (Separately am looking at how to do speedup).
Attachments
Attachments
Issue Links
- links to