[jira] [Resolved] (HBASE-24126) Up the container nproc uplimit from 10000 to 12500

["Michael Stack (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-24126?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Michael Stack resolved HBASE-24126.
-----------------------------------
    Fix Version/s: 2.3.0
                   3.0.0
     Hadoop Flags: Reviewed
     Release Note: Start docker with upped ulimit for nproc passing '--ulimit nproc=12500'. It was 10000, the default, but made it 12500. Then, set PROC_LIMIT in hbase-personality so when yetus runs, it is w/ the new 12500 value.  (was: Start docker with upped ulimit for nproc. It was 10000 but made it 12500. Then, set proclimit when we run yetus to be the new 12500 value.)
         Assignee: Michael Stack
       Resolution: Fixed

Merged master PR to master branch and to 2.3. On branch-2, applied a fixup to make branch-2 look like branch-2.3 and master (it had had a bunch of addendums applied against this JIRA but final solution undid some of the extra insurance the original put in place; i.e. passing --proclimit=12500 when invoking yetus).

> Up the container nproc uplimit from 10000 to 12500
> --------------------------------------------------
>
>                 Key: HBASE-24126
>                 URL: https://issues.apache.org/jira/browse/HBASE-24126
>             Project: HBase
>          Issue Type: Sub-task
>          Components: build
>            Reporter: Michael Stack
>            Assignee: Michael Stack
>            Priority: Major
>             Fix For: 3.0.0, 2.3.0
>
>         Attachments: 0001-HBASE-24126-Up-the-container-nproc-uplimit-from-1000.patch
>
>
> Let me up the container nproc count from 10000. If many tests running in parallel, container could breach the 10000 limit: i.e. if a bunch of heavy-duty long tests run at same time (0.5C on a 16CPU machine means 8 possible concurrents) and we spin up 1-2000 threads in each test, then could hit the 10000 limit. Move up the limit some expecially after INFRA just upped our limit to 30000. 



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