> On May 23, 2018, 9:43 p.m., Ashutosh Chauhan wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUpdaterThread.java
> > Lines 557 (patched)
> > <https://reviews.apache.org/r/67125/diff/1/?file=2022628#file2022628line557>
> >
> >     This will compute basic and column stats. I assume this is what you 
> > want.

Yes.


> On May 23, 2018, 9:43 p.m., Ashutosh Chauhan wrote:
> > standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/MetaStoreDirectSql.java
> > Lines 2424 (patched)
> > <https://reviews.apache.org/r/67125/diff/1/?file=2022633#file2022633line2424>
> >
> >     Returning concatenated table Name + Dbname is error prone. Lets make 
> > this return List <catName, dbName, tablename>

This seems to be a standard approach in Hive to pass around the full name(s). 
In fact some existing APIs for metastore return catalog+db in one string, and 
Hive uses full table name all over the place.
I don't think either is error-prone and requires String[]...


> On May 23, 2018, 9:43 p.m., Ashutosh Chauhan wrote:
> > standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/MetaStoreDirectSql.java
> > Lines 2425 (patched)
> > <https://reviews.apache.org/r/67125/diff/1/?file=2022633#file2022633line2425>
> >
> >     Currently it will fetch all tables (acid or not). Is that intentional?

Yes


> On May 23, 2018, 9:43 p.m., Ashutosh Chauhan wrote:
> > standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/RawStore.java
> > Lines 1642 (patched)
> > <https://reviews.apache.org/r/67125/diff/1/?file=2022635#file2022635line1642>
> >
> >     Return value should be List<List<CatName, DBname, TblName>>

Same as above... it's ok for Hive to use fq table name everywhere as a string, 
so it should be ok here


> On May 23, 2018, 9:43 p.m., Ashutosh Chauhan wrote:
> > standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/conf/MetastoreConf.java
> > Lines 738 (patched)
> > <https://reviews.apache.org/r/67125/diff/1/?file=2022638#file2022638line738>
> >
> >     I think this timeunit should be in minutes since this task need to run 
> > more frequently than that. That will make this config less error-prone.
> >     Default value : 1 hour.

Changed the setting. Note that this wait is the wait until we check the next 
time if the check was a noop... I don't think it's a good idea to wait for an 
hour.
It will produce counterintuitive behavior where if tables are updated steadily, 
after each non-noop run it will find yet another table to update and do the 
full check, but one small gap will result at nothing being updated for an hour.
This is a throttle for checks, not analyze queries. We may add separate one for 
queries if desired.


- Sergey


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/67125/#review203694
-----------------------------------------------------------


On May 15, 2018, 4:55 a.m., Sergey Shelukhin wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/67125/
> -----------------------------------------------------------
> 
> (Updated May 15, 2018, 4:55 a.m.)
> 
> 
> Review request for hive, Ashutosh Chauhan and Seong (Steve) Yeom.
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> see jira. This should eventually integrate with ACID stats to determine what 
> stats are out of date, when that is done. Probably in separate jira if this 
> goes in first.
> 
> 
> Diffs
> -----
> 
>   
> itests/hcatalog-unit/src/test/java/org/apache/hive/hcatalog/listener/DummyRawStoreFailEvent.java
>  3d6fda6bd4 
>   ql/src/java/org/apache/hadoop/hive/ql/DriverUtils.java PRE-CREATION 
>   ql/src/java/org/apache/hadoop/hive/ql/session/SessionState.java 89129f99fe 
>   ql/src/java/org/apache/hadoop/hive/ql/stats/StatsUpdaterThread.java 
> PRE-CREATION 
>   ql/src/java/org/apache/hadoop/hive/ql/txn/compactor/CompactorMR.java 
> b698c84080 
>   ql/src/test/org/apache/hadoop/hive/ql/stats/TestStatsUpdaterThread.java 
> PRE-CREATION 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/HiveAlterHandler.java
>  0be0aaa10c 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/HiveMetaStore.java
>  92d2e3f368 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/MetaStoreDirectSql.java
>  48f77b9878 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/ObjectStore.java
>  264fdb9db9 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/RawStore.java
>  ce7d2861dd 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/cache/CachedStore.java
>  b223920e82 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/conf/EnumValidator.java
>  PRE-CREATION 
>   
> standalone-metastore/src/main/java/org/apache/hadoop/hive/metastore/conf/MetastoreConf.java
>  114d5da205 
>   
> standalone-metastore/src/test/java/org/apache/hadoop/hive/metastore/DummyRawStoreControlledCommit.java
>  f6899be750 
>   
> standalone-metastore/src/test/java/org/apache/hadoop/hive/metastore/DummyRawStoreForJdoConnection.java
>  98a85cc758 
> 
> 
> Diff: https://reviews.apache.org/r/67125/diff/1/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Sergey Shelukhin
> 
>

Reply via email to