FAQ
Balancer can hang in PendingBlockMove
-------------------------------------

Key: HDFS-871
URL: https://issues.apache.org/jira/browse/HDFS-871
Project: Hadoop HDFS
Issue Type: Bug
Components: balancer
Affects Versions: 0.20.1
Environment: Yahoo 0.20
Reporter: Andrew Ryan


We started the balancer, with default options (-threshold 10), and it ran fine for a few hours, then hung. The process was still alive but no balancing was taking place.

At the time of the hang, jstack showed there were three threads in RUNNABLE status. Subsequent jstacks taken minutes and hours later showed the same three threads running in the same place, so I don't think this was a case where requests were being restarted, it looks like hangs. My best guess is, there's no timeout in the request to the namenode for these requests, and there needs to be.

I'll attach the full jstack output, but here's a sample thread, they are all stuck in the same place.

"pool-1-thread-972" prio=10 tid=0x00002aaafc23a800 nid=0x27a8 runnable [0x00002a
ab0a9a2000]
java.lang.Thread.State: RUNNABLE
at java.net.SocketInputStream.socketRead0(Native Method)
at java.net.SocketInputStream.read(SocketInputStream.java:129)
at java.io.BufferedInputStream.fill(BufferedInputStream.java:218)
at java.io.BufferedInputStream.read(BufferedInputStream.java:237)
- locked <0x00002aaaebdbe158> (a java.io.BufferedInputStream)
at java.io.DataInputStream.readShort(DataInputStream.java:295)
at org.apache.hadoop.hdfs.server.balancer.Balancer$PendingBlockMove.receiveResponse(Balancer.java:371)
at org.apache.hadoop.hdfs.server.balancer.Balancer$PendingBlockMove.dispatch(Balancer.java:326)
at org.apache.hadoop.hdfs.server.balancer.Balancer$PendingBlockMove.access$1800(Balancer.java:232)
at org.apache.hadoop.hdfs.server.balancer.Balancer$PendingBlockMove$1.run(Balancer.java:393)
at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
at java.lang.Thread.run(Thread.java:619)

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Search Discussions

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouphdfs-dev @
categorieshadoop
postedJan 6, '10 at 5:02p
activeJan 6, '10 at 5:02p
posts1
users1
websitehadoop.apache.org...
irc#hadoop

1 user in discussion

Andrew Ryan (JIRA): 1 post

People

Translate

site design / logo © 2022 Grokbase