[
https://issues.apache.org/jira/browse/HIVE-18457?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16327931#comment-16327931
]
Prasanth Jayachandran commented on HIVE-18457:
----------------------------------------------
For debugging, pasting the state of metastore tables here
{code}
mysql> select * from wm_resourceplan;
+-------+--------+-------------------+---------+-----------------+
| RP_ID | NAME | QUERY_PARALLELISM | STATUS | DEFAULT_POOL_ID |
+-------+--------+-------------------+---------+-----------------+
| 1 | global | NULL | ACTIVE | 1 |
| 2 | llap | NULL | ENABLED | 2 |
+-------+--------+-------------------+---------+-----------------+
2 rows in set (0.00 sec)
mysql> select * from wm_pool;
+---------+-------+---------+----------------+-------------------+-------------------+
| POOL_ID | RP_ID | PATH | ALLOC_FRACTION | QUERY_PARALLELISM |
SCHEDULING_POLICY |
+---------+-------+---------+----------------+-------------------+-------------------+
| 1 | 1 | default | 1 | 4 | NULL
|
| 2 | 2 | default | 1 | 4 | NULL
|
+---------+-------+---------+----------------+-------------------+-------------------+
2 rows in set (0.01 sec)
mysql> select * from wm_pool_to_trigger;
Empty set (0.01 sec)
mysql> select * from wm_mapping;
Empty set (0.00 sec)
mysql> select * from wm_trigger;
+------------+-------+-----------------+------------------------------+-------------------+-----------------+
| TRIGGER_ID | RP_ID | NAME | TRIGGER_EXPRESSION |
ACTION_EXPRESSION | IS_IN_UNMANAGED |
+------------+-------+-----------------+------------------------------+-------------------+-----------------+
| 29 | 1 | highly_parallel | TOTAL_TASKS > 40 | KILL
| |
| 33 | 1 | big_hdfs_read | HDFS_BYTES_READ > 3000000000 | KILL
| |
| 34 | 1 | slow_query | EXECUTION_TIME > 100000 | KILL
| |
| 35 | 1 | some_spills | SPILLED_RECORDS > 10 | KILL
| |
+------------+-------+-----------------+------------------------------+-------------------+-----------------+
4 rows in set (0.00 sec)
{code}
> Triggers in unmanaged pools are not shown
> -----------------------------------------
>
> Key: HIVE-18457
> URL: https://issues.apache.org/jira/browse/HIVE-18457
> Project: Hive
> Issue Type: Sub-task
> Affects Versions: 3.0.0
> Reporter: Prasanth Jayachandran
> Assignee: Sergey Shelukhin
> Priority: Major
>
> Did the following sequence to add triggers to UNMANAGED. I can see the
> triggers added to metastore by IS_IN_UNAMANGED flag is not set in metastore.
> Also show resource plans does not show triggers in unmanaged pool.
> {code}
> 0: jdbc:hive2://localhost:10000> show resource plans;
> +----------+----------+--------------------+
> | rp_name | status | query_parallelism |
> +----------+----------+--------------------+
> | global | ACTIVE | NULL |
> | llap | ENABLED | NULL |
> +----------+----------+--------------------+
> 0: jdbc:hive2://localhost:10000>ALTER RESOURCE PLAN llap ACTIVATE;
> 0: jdbc:hive2://localhost:10000>ALTER RESOURCE PLAN global DISABLE;
> 0: jdbc:hive2://localhost:10000>CREATE TRIGGER global.highly_parallel WHEN
> TOTAL_TASKS > 40 DO KILL;
> 0: jdbc:hive2://localhost:10000>ALTER TRIGGER global.highly_parallel ADD TO
> UNMANAGED;
> 0: jdbc:hive2://localhost:10000>CREATE TRIGGER global.big_hdfs_read WHEN
> HDFS_BYTES_READ > 3000000000 DO KILL;
> 0: jdbc:hive2://localhost:10000>ALTER TRIGGER global.big_hdfs_read ADD TO
> UNMANAGED;
> 0: jdbc:hive2://localhost:10000>CREATE TRIGGER global.slow_query WHEN
> EXECUTION_TIME > 100000 DO KILL;
> 0: jdbc:hive2://localhost:10000>ALTER TRIGGER global.slow_query ADD TO
> UNMANAGED;
> 0: jdbc:hive2://localhost:10000>CREATE TRIGGER global.some_spills WHEN
> SPILLED_RECORDS > 10 DO KILL;
> 0: jdbc:hive2://localhost:10000>ALTER TRIGGER global.some_spills ADD TO
> UNMANAGED;
> 0: jdbc:hive2://localhost:10000>ALTER RESOURCE PLAN global ENABLE;
> 0: jdbc:hive2://localhost:10000>ALTER RESOURCE PLAN global ACTIVATE;
> 0: jdbc:hive2://localhost:10000> show resource plan global;
> +----------------------------------------------------+
> | line |
> +----------------------------------------------------+
> | global[status=ACTIVE,parallelism=null,defaultPool=default] |
> | default[allocFraction=1.0,schedulingPolicy=null,parallelism=4] |
> +----------------------------------------------------+
> {code}
> {code:title=mysql}
> mysql> select * from wm_trigger;
> +------------+-------+-----------------+------------------------------+-------------------+-----------------+
> | TRIGGER_ID | RP_ID | NAME | TRIGGER_EXPRESSION |
> ACTION_EXPRESSION | IS_IN_UNMANAGED |
> +------------+-------+-----------------+------------------------------+-------------------+-----------------+
> | 29 | 1 | highly_parallel | TOTAL_TASKS > 40 | KILL
> | |
> | 33 | 1 | big_hdfs_read | HDFS_BYTES_READ > 3000000000 | KILL
> | |
> | 34 | 1 | slow_query | EXECUTION_TIME > 100000 | KILL
> | |
> | 35 | 1 | some_spills | SPILLED_RECORDS > 10 | KILL
> | |
> +------------+-------+-----------------+------------------------------+-------------------+-----------------+
> {code}
> From the above mysql table, IS_IN_UNMANAGED is not set and 'show resource
> plan global' is not showing triggers defined in unmanaged pool.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
