[
https://issues.apache.org/jira/browse/IMPALA-5662?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16487675#comment-16487675
]
ASF subversion and git services commented on IMPALA-5662:
---------------------------------------------------------
Commit 8d474a0170f60526e0b5611687e5166ae29ecae9 in impala's branch
refs/heads/master from [~bikram.sngh91]
[ https://git-wip-us.apache.org/repos/asf?p=impala.git;h=8d474a0 ]
IMPALA-3134: Support different proc mem limits among impalads for
admission control checks
Currently the admission controller assumes that all backends have the
same process mem limit as the impalad it itself is running on. With
this patch the proc mem limit for each impalad is available to the
admission controller and it uses it for making correct admisssion
decisions. It currently works under the assumption that the
per-process memory limit does not change dynamically.
Testing:
Added an e2e test.
IMPALA-5662: Log the queuing reason for a query
The queuing reason is now logged both while queuing for the first
time and while trying to dequeue.
Change-Id: Idb72eee790cc17466bbfa82e30f369a65f2b060e
Reviewed-on: http://gerrit.cloudera.org:8080/10396
Reviewed-by: Bikramjeet Vig <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Log all information relevant to admission control decision making
> -----------------------------------------------------------------
>
> Key: IMPALA-5662
> URL: https://issues.apache.org/jira/browse/IMPALA-5662
> Project: IMPALA
> Issue Type: Improvement
> Components: Backend
> Reporter: Balazs Jeszenszky
> Assignee: Bikramjeet Vig
> Priority: Major
> Labels: admission-control, observability, resource-management,
> supportability
>
> Currently, when making a decision whether to admit a query or not, the log
> has the following format:
> {code:java}
> I0705 14:43:04.031771 7388 admission-controller.cc:442] Stats:
> agg_num_running=1, agg_num_queued=0, agg_mem_reserved=486.74 MB,
> local_host(local_mem_admitted=0, num_admitted_running=0, num_queued=0,
> backend_mem_reserved=56.07 MB)
> {code}
> Since it's also possible to queue queries due to one node not being able to
> reserve the required memory, we should also log the max(backend_mem_reserved)
> across all nodes.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
