[ 
https://issues.apache.org/jira/browse/IGNITE-9754?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16644634#comment-16644634
 ] 

Oleg Ignatenko commented on IGNITE-9754:
----------------------------------------

hi [~NSAmelchev] this is known problem and it is expected to be addressed in a 
separate ticket that is linked to this one (IGNITE-9512). The point of the 
change I made here is to only get rid from unreasonable timeout and nothing 
else, we have already discussed this with [~ascherbakov], and the reason we do 
it separately is in particular to help us get more informative failure reports 
from Teamcity (because currently these are polluted by senseless timeout 
failures).

If you insist on further loading this change with fixes that were intended to 
be made per another ticket, please let me know and I'll re-check with 
[~ascherbakov].

> Increase timeout waiting for commitLatch in test 
> TxRollbackAsyncWithPersistenceTest.testRollbackOnTopologyLockPessimistic
> -------------------------------------------------------------------------------------------------------------------------
>
>                 Key: IGNITE-9754
>                 URL: https://issues.apache.org/jira/browse/IGNITE-9754
>             Project: Ignite
>          Issue Type: Bug
>    Affects Versions: 2.6
>            Reporter: Oleg Ignatenko
>            Assignee: Oleg Ignatenko
>            Priority: Major
>              Labels: MakeTeamcityGreenAgain
>             Fix For: 2.8
>
>
> Current timeout for {{commitLatch}} in test 
> {{TxRollbackAsyncWithPersistenceTest.testRollbackOnTopologyLockPessimistic}} 
> is 10 seconds.
> This looks insufficient to make a compelling reason that test failure 
> indicates a problem in the implementation. For example log of this [recent 
> failure at 
> Teamcity|https://ci.ignite.apache.org/viewLog.html?buildId=1970040&tab=buildResultsDiv&buildTypeId=IgniteTests24Java8_Cache7]
>  suggests that there were just a couple of non-critical delays for a few 
> seconds that made test fail.
> Suggest to increase timeout from 10 seconds to 60, so that in case if test 
> fails waiting on the latch one could assume with reasonable certainty that 
> something is wrong with the implementation or infrastructure.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to