[
https://issues.apache.org/jira/browse/HIVE-26414?focusedWorklogId=794071&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-794071
]
ASF GitHub Bot logged work on HIVE-26414:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 22/Jul/22 06:07
Start Date: 22/Jul/22 06:07
Worklog Time Spent: 10m
Work Description: SourabhBadhya commented on code in PR #3457:
URL: https://github.com/apache/hive/pull/3457#discussion_r927295389
##########
ql/src/java/org/apache/hadoop/hive/ql/lockmgr/DbTxnManager.java:
##########
@@ -485,6 +480,26 @@ private void clearLocksAndHB() {
stopHeartbeat();
}
+ private void cleanupDirForCTAS() {
Review Comment:
Currently `Context ctx` is not available during `rollbackTxn` which is why I
chose to store the object.
However I agree passing `Context ctx` is better.
There are multiple ways of doing this -
First would be to involve passing `Context ctx` to `rollbackTxn` method
which would change the HiveTxnManager API itself (I particularly dont like this
since this would be a breaking change).
Or we could create a new function in the `HiveTxnManager` interface of the
same name and call it from the driver when rollback conditions are satisfied.
My idea was to avoid both and initialise the `destinationTable` in one of
the existing APIs but I am open for any other suggestions.
Issue Time Tracking
-------------------
Worklog Id: (was: 794071)
Time Spent: 1h 50m (was: 1h 40m)
> Aborted/Cancelled CTAS operations must initiate cleanup of uncommitted data
> ---------------------------------------------------------------------------
>
> Key: HIVE-26414
> URL: https://issues.apache.org/jira/browse/HIVE-26414
> Project: Hive
> Issue Type: Improvement
> Reporter: Sourabh Badhya
> Assignee: Sourabh Badhya
> Priority: Major
> Labels: pull-request-available
> Time Spent: 1h 50m
> Remaining Estimate: 0h
>
> When a CTAS query fails before creation of table and after writing the data,
> the data is present in the directory and not cleaned up currently by the
> cleaner or any other mechanism currently. This is because the cleaner
> requires a table corresponding to what its cleaning. In order surpass such a
> situation, we can directly pass the relevant information to the cleaner so
> that such uncommitted data is deleted.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
