-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/63890/#review191387
-----------------------------------------------------------




itests/hive-unit/src/test/java/org/apache/hive/jdbc/TestTriggersNoTezSessionPool.java
Line 59 (original), 58 (patched)
<https://reviews.apache.org/r/63890/#comment269182>

    doesn't look like pool is used



ql/src/java/org/apache/hadoop/hive/ql/exec/DDLTask.java
Line 716 (original), 717 (patched)
<https://reviews.apache.org/r/63890/#comment269183>

    nit: WorkloadManager.getInstance is enough to determine if it's enabled. 
    WM is initialized based on queue name only



ql/src/java/org/apache/hadoop/hive/ql/exec/DDLTask.java
Lines 743 (patched)
<https://reviews.apache.org/r/63890/#comment269184>

    hmm... why is this an either or case? it's possible to have both in the 
same HS2. Should it call the else in both cases?



ql/src/java/org/apache/hadoop/hive/ql/exec/tez/TezSessionPoolManager.java
Line 511 (original), 511 (patched)
<https://reviews.apache.org/r/63890/#comment269186>

    openSessions appears to be synchronized elsewhere, accessing it here may 
not be thread safe. Might make sense to call this under the same sync block 
that makes the changes to the list.



ql/src/java/org/apache/hadoop/hive/ql/exec/tez/TezSessionPoolManager.java
Lines 517 (patched)
<https://reviews.apache.org/r/63890/#comment269185>

    this seems to get all triggers for the pools... I can file a follow-up 
jira, I think we need to make addition of triggers to Tez more explicit



service/src/java/org/apache/hive/service/server/HiveServer2.java
Line 218 (original), 220 (patched)
<https://reviews.apache.org/r/63890/#comment269187>

    same about else



standalone-metastore/src/gen/thrift/gen-javabean/org/apache/hadoop/hive/metastore/api/WMFullResourcePlan.java
Line 329 (original), 329 (patched)
<https://reviews.apache.org/r/63890/#comment269189>

    this looks like a change to generated code without a thrift change. The 
null check (or isSet check?) should be in the caller


- Sergey Shelukhin


On Nov. 17, 2017, 2:20 a.m., Prasanth_J wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/63890/
> -----------------------------------------------------------
> 
> (Updated Nov. 17, 2017, 2:20 a.m.)
> 
> 
> Review request for hive and Sergey Shelukhin.
> 
> 
> Bugs: HIVE-18025
>     https://issues.apache.org/jira/browse/HIVE-18025
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> HIVE-18025: Push resource plan changes to tez/unmanaged sessions
> 
> 
> Diffs
> -----
> 
>   
> itests/hive-unit/src/test/java/org/apache/hive/jdbc/TestTriggersNoTezSessionPool.java
>  bcce3dc 
>   
> itests/hive-unit/src/test/java/org/apache/hive/jdbc/TestTriggersTezSessionPoolManager.java
>  b377275 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/DDLTask.java e7af5e0 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/tez/TezSessionPoolManager.java 
> 4e48f15 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/tez/TezSessionPoolSession.java 
> 769b24a 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/tez/WorkloadManager.java 3990f95 
>   
> ql/src/java/org/apache/hadoop/hive/ql/wm/MetastoreGlobalTriggersFetcher.java 
> 87c007f 
>   service/src/java/org/apache/hive/service/server/HiveServer2.java 5a6d0c8 
>   
> standalone-metastore/src/gen/thrift/gen-javabean/org/apache/hadoop/hive/metastore/api/WMFullResourcePlan.java
>  1d21908 
> 
> 
> Diff: https://reviews.apache.org/r/63890/diff/2/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Prasanth_J
> 
>

Reply via email to