You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Eugene Koifman (JIRA)" <ji...@apache.org> on 2016/04/21 02:46:25 UTC
[jira] [Updated] (HIVE-11550) ACID queries pollute HiveConf
[ https://issues.apache.org/jira/browse/HIVE-11550?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Eugene Koifman updated HIVE-11550:
----------------------------------
Description:
HiveConf is a SessionState level object. Some ACID related logic makes changes to it (which are meant to be per query) but become per SessionState.
See SemanticAnalyzer.checkAcidConstraints()
Also note HiveConf.setVar(conf, HiveConf.ConfVars.DYNAMICPARTITIONINGMODE, "nonstrict");
in UpdateDeleteSemancitAnalzyer
[~alangates], do you know of other cases or ideas on how to deal with this differently?
_SortedDynPartitionOptimizer.process()_ is the place to have the logic to do _conf.setBoolVar(ConfVars.HIVEOPTSORTDYNAMICPARTITION, false);_ on per query basis
was:
HiveConf is a SessionState level object. Some ACID related logic makes changes to it (which are meant to be per query) but become permanent.
See SemanticAnalyzer.checkAcidConstraints()
Also note HiveConf.setVar(conf, HiveConf.ConfVars.DYNAMICPARTITIONINGMODE, "nonstrict");
in UpdateDeleteSemancitAnalzyer
[~alangates], do you know of other cases or ideas on how to deal with this differently?
> ACID queries pollute HiveConf
> -----------------------------
>
> Key: HIVE-11550
> URL: https://issues.apache.org/jira/browse/HIVE-11550
> Project: Hive
> Issue Type: Bug
> Components: Transactions
> Affects Versions: 1.0.0
> Reporter: Eugene Koifman
> Assignee: Eugene Koifman
>
> HiveConf is a SessionState level object. Some ACID related logic makes changes to it (which are meant to be per query) but become per SessionState.
> See SemanticAnalyzer.checkAcidConstraints()
> Also note HiveConf.setVar(conf, HiveConf.ConfVars.DYNAMICPARTITIONINGMODE, "nonstrict");
> in UpdateDeleteSemancitAnalzyer
> [~alangates], do you know of other cases or ideas on how to deal with this differently?
> _SortedDynPartitionOptimizer.process()_ is the place to have the logic to do _conf.setBoolVar(ConfVars.HIVEOPTSORTDYNAMICPARTITION, false);_ on per query basis
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)