[ 
https://issues.apache.org/jira/browse/ACCUMULO-4469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15517280#comment-15517280
 ] 

Sean Busbey commented on ACCUMULO-4469:
---------------------------------------

{quote}
The fixture clean up is the only thing that can run in parallel, since the 
randomwalk executor only has one thread. We should make a copy of the list, but 
that also might get a concurrent modification exception (e.g. if the 
waitTermination timedout). It should be rare, so I think we should just retry 
making a deep copy. (using CopyOnWriteArrayList would avoid the concurrent 
modification exception but we might miss a table to delete if the final 
operation was one of those that adds to the list.)
{quote}

I reflected on this over lunch, and I think my reasoning is flawed. If we're 
making a deep copy, we're just as likely to miss a final table addition as if 
we used CopyOnWriteArrayList in the first place. I think the COWAL approach 
will be simpler, so we should do it. However, we should add a follow on JIRA to 
improve this module to make use of a namespace that we just delete at the end 
to account for stragglers.

> ConcurrentModificationException while running MultiTable.xml node in Random 
> Walk 
> ---------------------------------------------------------------------------------
>
>                 Key: ACCUMULO-4469
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-4469
>             Project: Accumulo
>          Issue Type: Bug
>          Components: test
>    Affects Versions: 1.7.2
>            Reporter: Dima Spivak
>            Assignee: Dima Spivak
>
> After the resolution of ACCUMULO-4467, I got back to playing with Random Walk 
> and had a failure caused by a {{ConcurrentModificationException}}:
> {code}
> 23 01:03:04,316 [randomwalk.Framework] ERROR: Error during random walk
> java.lang.Exception: Error running node MultiTable.xml
>         at org.apache.accumulo.test.randomwalk.Module.visit(Module.java:346)
>         at 
> org.apache.accumulo.test.randomwalk.Framework.run(Framework.java:59)
>         at 
> org.apache.accumulo.test.randomwalk.Framework.main(Framework.java:119)
>         at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>         at 
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
>         at 
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>         at java.lang.reflect.Method.invoke(Method.java:606)
>         at org.apache.accumulo.start.Main$2.run(Main.java:157)
>         at java.lang.Thread.run(Thread.java:745)
> Caused by: java.util.ConcurrentModificationException
>         at java.util.ArrayList$Itr.checkForComodification(ArrayList.java:859)
>         at java.util.ArrayList$Itr.next(ArrayList.java:831)
>         at 
> org.apache.accumulo.test.randomwalk.multitable.MultiTableFixture.tearDown(MultiTableFixture.java:64)
>         at org.apache.accumulo.test.randomwalk.Module.visit(Module.java:365)
>         at org.apache.accumulo.test.randomwalk.Module$1.call(Module.java:283)
>         at org.apache.accumulo.test.randomwalk.Module$1.call(Module.java:278)
>         at java.util.concurrent.FutureTask.run(FutureTask.java:262)
>         at 
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
>         at java.util.concurrent.FutureTask.run(FutureTask.java:262)
>         at 
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
>         at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
>         at 
> org.apache.accumulo.fate.util.LoggingRunnable.run(LoggingRunnable.java:35)
>         ... 1 more
> {code}
> [This section of 
> code|https://github.com/apache/accumulo/blob/master/test/src/main/java/org/apache/accumulo/test/randomwalk/multitable/MultiTableFixture.java#L61-L71]
>  seems to be at fault. In particular, it looks like we're getting the table 
> list, but then instead of doing a deep copy to a new {{ArrayList<String>}} 
> from which we choose tables to delete, we're looping through and deleting 
> tables while referring to the changing list, which has the effect of 
> modifying it and making Java unhappy. Am I missing something more complex or 
> can I fix this one myself by just doing the aforementioned deep copy of the 
> table list? Or is a better way to use the {{TableOperations.list()}} method 
> and iterate through the {{SortedSet<String>}} it provides?



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to