[HBASE-8778] Region assigments scan table directory making them slow for huge tables

• Type: Improvement
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Fix Version/s: 0.98.0, 0.95.2
• Hadoop Flags: Reviewed

Release Note

Table descriptors are now moved inside hdfs from residing directly in the table directory (alongside region directories) to being in a well known subdirectory called ".tabledesc". For example, instead of /hbase/exampleTable/.tableinfo.0000000003 the file would be /hbase/exampleTable/.tabledesc/.tableinfo.0000000003 after this release. The same will be true for snapshots. The first active master to be started up will move these files for existing tables and snapshots.

• Tags: 0.96notable
• Assignee: Dave Latham
• Reporter: Dave Latham
• Created: 20/Jun/13 23:35
• Updated: 31/Jul/15 06:58
• Resolved: 06/Aug/13 14:44

Description

On a table with 130k regions it takes about 3 seconds for a region server to open a region once it has been assigned.

Watching the threads for a region server running 0.94.5 that is opening many such regions shows the thread opening the reigon in code like this:

"PRI IPC Server handler 4 on 60020" daemon prio=10 tid=0x00002aaac07e9000 nid=0x6566 runnable [0x000000004c46d000]
   java.lang.Thread.State: RUNNABLE
        at java.lang.String.indexOf(String.java:1521)
        at java.net.URI$Parser.scan(URI.java:2912)
        at java.net.URI$Parser.parse(URI.java:3004)
        at java.net.URI.<init>(URI.java:736)
        at org.apache.hadoop.fs.Path.initialize(Path.java:145)
        at org.apache.hadoop.fs.Path.<init>(Path.java:126)
        at org.apache.hadoop.fs.Path.<init>(Path.java:50)
        at org.apache.hadoop.hdfs.protocol.HdfsFileStatus.getFullPath(HdfsFileStatus.java:215)
        at org.apache.hadoop.hdfs.DistributedFileSystem.makeQualified(DistributedFileSystem.java:252)
        at org.apache.hadoop.hdfs.DistributedFileSystem.listStatus(DistributedFileSystem.java:311)
        at org.apache.hadoop.fs.FilterFileSystem.listStatus(FilterFileSystem.java:159)
        at org.apache.hadoop.fs.FileSystem.listStatus(FileSystem.java:842)
        at org.apache.hadoop.fs.FileSystem.listStatus(FileSystem.java:867)
        at org.apache.hadoop.hbase.util.FSUtils.listStatus(FSUtils.java:1168)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.getTableInfoPath(FSTableDescriptors.java:269)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.getTableInfoPath(FSTableDescriptors.java:255)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.getTableInfoModtime(FSTableDescriptors.java:368)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.get(FSTableDescriptors.java:155)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.get(FSTableDescriptors.java:126)
        at org.apache.hadoop.hbase.regionserver.HRegionServer.openRegion(HRegionServer.java:2834)
        at org.apache.hadoop.hbase.regionserver.HRegionServer.openRegion(HRegionServer.java:2807)
        at sun.reflect.GeneratedMethodAccessor64.invoke(Unknown Source)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at org.apache.hadoop.hbase.ipc.WritableRpcEngine$Server.call(WritableRpcEngine.java:320)
        at org.apache.hadoop.hbase.ipc.HBaseServer$Handler.run(HBaseServer.java:1426)

To open the region, the region server first loads the latest HTableDescriptor. Since HBASE-4553 HTableDescriptor's are stored in the file system at "/hbase/<tableDir>/.tableinfo.<sequenceNum>". The file with the largest sequenceNum is the current descriptor. This is done so that the current descirptor is updated atomically. However, since the filename is not known in advance FSTableDescriptors it has to do a FileSystem.listStatus operation which has to list all files in the directory to find it. The directory also contains all the region directories, so in our case it has to load 130k FileStatus objects. Even using a globStatus matching function still transfers all the objects to the client before performing the pattern matching. Furthermore HDFS uses a default of transferring 1000 directory entries in each RPC call, so it requires 130 roundtrips to the namenode to fetch all the directory entries.

Consequently, to reassign all the regions of a table (or a constant fraction thereof) requires time proportional to the square of the number of regions.

In our case, if a region server fails with 200 such regions, it takes 10+ minutes for them all to be reassigned, after the zk expiration and log splitting.

Issue Links

• is related to: HBASE-8348 Polish the migration to 0.96
• is related to: HBASE-9132 Use table dir modtime to avoid scanning table dir to check cached table descriptor in 0.94