You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Norbert Schultz (JIRA)" <ji...@apache.org> on 2018/04/12 13:08:00 UTC
[jira] [Commented] (CASSANDRA-12882) Deadlock in MemtableAllocator
[ https://issues.apache.org/jira/browse/CASSANDRA-12882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16435522#comment-16435522 ]
Norbert Schultz commented on CASSANDRA-12882:
---------------------------------------------
We have the same Problem on fairly huge machines with Cassandra 3.0.14 & DSE 5.0.11
A lot of Threads waiting for the MemTableAllocator, the whole instance is blocking and does not respond to anything.
{code:java}
"SharedPool-Worker-19" daemon prio=5 tid=1542 WAITING
at sun.misc.Unsafe.park(Native Method)
at java.util.concurrent.locks.LockSupport.park(LockSupport.java:304)
at org.apache.cassandra.utils.concurrent.WaitQueue$AbstractSignal.awaitUninterruptibly(WaitQueue.java:279)
Local Variable: org.apache.cassandra.utils.concurrent.WaitQueue$AnySignal#10
at org.apache.cassandra.utils.memory.MemtableAllocator$SubAllocator.allocate(MemtableAllocator.java:162)
at org.apache.cassandra.utils.memory.SlabAllocator.allocate(SlabAllocator.java:82)
at org.apache.cassandra.utils.memory.ContextAllocator.allocate(ContextAllocator.java:57)
at org.apache.cassandra.utils.memory.ContextAllocator.clone(ContextAllocator.java:47)
...{code}
> Deadlock in MemtableAllocator
> -----------------------------
>
> Key: CASSANDRA-12882
> URL: https://issues.apache.org/jira/browse/CASSANDRA-12882
> Project: Cassandra
> Issue Type: Bug
> Environment: Ubuntu 14.40
> Cassandra 3.5
> Reporter: Nimi Wariboko Jr.
> Priority: Major
> Fix For: 3.11.x
>
> Attachments: cassandra.yaml, threaddump.txt
>
>
> I'm seeing an issue where a node will eventually lock up and their thread pools - I looked into jstack, and a lot of threads are stuck in the Memtable Allocator
> {code}
> at java.util.concurrent.locks.LockSupport.park(LockSupport.java:304)
> at org.apache.cassandra.utils.concurrent.WaitQueue$AbstractSignal.awaitUninterruptibly(WaitQueue.java:279)
> at org.apache.cassandra.utils.memory.MemtableAllocator$SubAllocator.allocate(MemtableAllocator.java:198)
> at org.apache.cassandra.utils.memory.SlabAllocator.allocate(SlabAllocator.java:89)
> at org.apache.cassandra.utils.memory.ContextAllocator.allocate(ContextAllocator.java:57)
> at org.apache.cassandra.utils.memory.ContextAllocator.clone(ContextAllocator.java:47)
> at org.apache.cassandra.utils.memory.MemtableBufferAllocator.clone(MemtableBufferAllocator.java:41)
> {code}
> I looked into the code, and its not immediately apparent to me what thread might hold the relevant lock.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org