[infinispan-dev] Transaction table cleanup

Mircea Markus mircea.markus at jboss.com
Thu Oct 18 21:44:31 EDT 2012


On 18 Oct 2012, at 18:21, Manik Surtani wrote:

> On 19 Oct 2012, at 02:19, Mircea Markus <mircea.markus at jboss.com> wrote:
> 
>> 
>> On 18 Oct 2012, at 16:41, Manik Surtani wrote:
>> 
>>> 
>>> On 18 Oct 2012, at 17:45, Vladimir Blagojevic <vblagoje at redhat.com> wrote:
>>> 
>>>> Hey guys,
>>>> 
>>>> Investigating why EmbeddedCacheManager#cacheRemove hiccups 
>>>> DistributedTwoNodesMapReduceTest. As you might recall upon end of 
>>>> MapReduceTask there is EmbeddedCacheManager#cacheRemove call to remove 
>>>> intermediate caches across the cluster.
>> I think you meant ECM#removeCache?
>>>> Very often, almost every test 
>>>> run, execution of cache remove is blocked by ongoing transactions that 
>>>> have not completed - more specifically TransactionTable shows pending 
>>>> remote transaction. This in turn prevents cache stop call, which in turn 
>>>> causes a timeout on cache remove thus failing the test.
>>>> 
>>>> MapReduce uses txs in MapReduceManagerImpl#combine - it might well be 
>>>> that there is smth wrong there or it could be that somehow 
>>> 
>>> ^^^ Does this mean that to use Map/Reduce now, we *require* that a transaction manager is available, configured and present?
>> Not really. It's only that the problem Vladimir describes only occurs when you run the MR against transactional caches (there is a tx leak which prevents caches to stop, that should be fixed by a pull request I sent).
> 
> Ok.  I was just wondering, since Vladimir did say MR uses txs in MapReduceManagerImpl#combine.  Vladimir, could you pls clarify?
You're right actually, the temporary cache created is transactional. it is built in the CreateCacheCommand and relies on the DummyTransactionManager, might be better to use batching perhaps? Or even not require for this cache to be transactional?

Cheers,
-- 
Mircea Markus
Infinispan lead (www.infinispan.org)




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/infinispan-dev/attachments/20121018/5b18556c/attachment-0001.html 


More information about the infinispan-dev mailing list