[jira] [Updated] (IGNITE-16290) Non-informative warning on query retry - Query run was already removed

["Luchnikov Alexander (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IGNITE-16290?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Luchnikov Alexander updated IGNITE-16290:
-----------------------------------------
    Labels: ise.lts  (was: )

> Non-informative warning on query retry - Query run was already removed
> ----------------------------------------------------------------------
>
>                 Key: IGNITE-16290
>                 URL: https://issues.apache.org/jira/browse/IGNITE-16290
>             Project: Ignite
>          Issue Type: Bug
>          Components: sql
>            Reporter: Andrey Mashenkov
>            Assignee: Andrey Mashenkov
>            Priority: Major
>              Labels: ise.lts
>             Fix For: 2.13
>
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> The non-informative warning is printed when SQL query is restarted.
> {noformat}
> [WARN ][pub-#3460][GridReduceQueryExecutor] Query run was already removed: 1467
> 2[2021-11-28T05:30:25,813][WARN ][pub-#3460][GridReduceQueryExecutor] Query run was already removed: 1468
> 3[2021-11-28T05:30:25,874][WARN ][pub-#3460][GridReduceQueryExecutor] Query run was already removed: 1469{noformat}
> Theis happens because we remove query context (ReduceQueryRun) from a map twice: when decide to retry the query 
> {code:java}
>                     if (retry) {
>                         lastRun = runs.remove(qryReqId);
>                         assert lastRun != null;
>                     } else { {code}
> then when releasing resources in a finally block.
> {code:java}
> finally {
>     if (release) {
>         releaseRemoteResources(nodes, r, qryReqId, qry.distributedJoins(), mvccTracker); {code}
> Thus, the second try is failed with the log message.



--
This message was sent by Atlassian Jira
(v8.20.7#820007)