[ 
https://issues.apache.org/jira/browse/SOLR-15644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17419020#comment-17419020
 ] 

Mark Robert Miller commented on SOLR-15644:
-------------------------------------------

Sure, follow the issues need it, I won’t be firing it around in this issue for 
no reason and I’m only working on a dozen test issues at most.

It’s applicable in dozens and dozens of cases, there won’t be much to follow in 
the few out of the dozen I’ll do here.

Right now it matters little - most tests are super inefficient and slow, the 
universal 10 second linger gives lots of time, other waits like the up to 60 
second wait for closed objects is in affect, god knows what else I’m 
forgetting. And so long shutdowns, pushing to the end of lingering, and then 
interrupt whatever remains, maybe retry a little, it’s all in the same noise.

You don’t see all the issues and problems Solr actually throws against it until 
you make any of that reasonable, turn off the ridiculous exceptions and waits 
and lingering, and still have to deal with a plethora of behavior that you want 
as fast as if it actually behaved.

I’m not actually after any of that here, but point is, the control needed is 
not there. This is not Lucene. Here I need it to deal with Poot behavior of 
limited items from limited tests where if I fix them to be reasonable and I 
don’t want to linger and wait forever, they will trip thread leaks and create 
fails and or other none sense. 

In an ideal world, interrupts would almost never make things slower, more fine 
interrupts and / or waits would not be needed. But java programmers don’t 
really work that way. The test framework hardly sees a lot of it because a 
bunch of exceptions and sponges that are even more forgiving than the 
exceptions we have in it eat up first. 

> Add the ability to interrupt and wait for threads for problematic tests.
> ------------------------------------------------------------------------
>
>                 Key: SOLR-15644
>                 URL: https://issues.apache.org/jira/browse/SOLR-15644
>             Project: Solr
>          Issue Type: Test
>      Security Level: Public(Default Security Level. Issues are Public) 
>          Components: Tests
>            Reporter: Mark Robert Miller
>            Assignee: Mark Robert Miller
>            Priority: Major
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> The stuff in the test framework is slow and lacks control. For problematic 
> tests, you don't want to linger first and you want fine control around 
> interrupting - interrupting with a sledgehammer approach can actually make 
> things take longer.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to