hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sergey Shelukhin (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-19916) master-txnstats branch - don't get write IDs from metastore when it's not safe
Date Wed, 27 Jun 2018 21:24:00 GMT

    [ https://issues.apache.org/jira/browse/HIVE-19916?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16525631#comment-16525631
] 

Sergey Shelukhin commented on HIVE-19916:
-----------------------------------------

Attached the master patch (whole branch + this patch) for HiveQA. If we see actual failures
in optimizer, then those would need to be fixed (because getting writeIds twice with the current
double compile scheme where optimizers run outside of lock could mean incorrect results).
If some DDL paths fail then we'd postpone... if none fail (compared to the branch itself)
we'd commit. 

> master-txnstats branch - don't get write IDs from metastore when it's not safe
> ------------------------------------------------------------------------------
>
>                 Key: HIVE-19916
>                 URL: https://issues.apache.org/jira/browse/HIVE-19916
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Sergey Shelukhin
>            Assignee: Sergey Shelukhin
>            Priority: Major
>         Attachments: HIVE-19916.patch, branch-19916.patch
>
>
> There's some code in original txn stats patch that may go to metastore to get write Ids.
This code should not go to metastore, it should fail instead. 
> HIVE-19382 should ensure that we have correct IDs already present during optimizer -
they are using by e.g. materialized view optimizer, so they should be there; if they are not
present, some integration might be needed so that txn stats optimizations also have access
to those write Ids.
> cc [~jcamachorodriguez]



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

Mime
View raw message