tx aborted errors in AlertConditionCacheCoordinator

Ian Springer ian.springer at redhat.com
Mon Oct 17 16:32:42 UTC 2011


Should I create a BZ? I can see users hitting this if they took their 
Server down over a weekend to do a DB upgrade or some other sort of 
maintenance.

On 10/17/2011 12:05 PM, John Mazzitelli wrote:
> On Mon, 2011-10-17 at 11:42 -0400, Ian Springer wrote:
>> Notice, these 9 merges all completed within 20 seconds of each other,
>> and they each took 15-20 seconds to complete, which means all 9 of them
>> were running simultaneously. We should consider adding some throttling
>> on the Server side that only allows one measurement report from a given
>> Agent to be processed at a time. Perhaps the Server could reject any
>> other requests that come in, and the Agent would have to retry them later.
>>
> I actually think we have the ability to do this. IIRC, me and Joe put in
> some kind of latch or semaphore framework that lets you prohibit more
> than one report of a particular type from a particular agent to be
> processed (but can concurrently process other agent reports).
>
>
> _______________________________________________
> rhq-devel mailing list
> rhq-devel at lists.fedorahosted.org
> https://fedorahosted.org/mailman/listinfo/rhq-devel


-- 
Ian Springer
Principal Software Developer
JBoss Operations Network
Red Hat
ian.springer at redhat.com



More information about the rhq-devel mailing list