[jira] [Work logged] (HIVE-24289) RetryingMetaStoreClient should not retry connecting to HMS on genuine errors

["ASF GitHub Bot (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HIVE-24289?focusedWorklogId=693808&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-693808 ]

ASF GitHub Bot logged work on HIVE-24289:
-----------------------------------------

                Author: ASF GitHub Bot
            Created on: 10/Dec/21 09:35
            Start Date: 10/Dec/21 09:35
    Worklog Time Spent: 10m 
      Work Description: kgyrtkirk commented on a change in pull request #2675:
URL: https://github.com/apache/hive/pull/2675#discussion_r766511427



##########
File path: standalone-metastore/metastore-common/src/main/java/org/apache/hadoop/hive/metastore/RetryingMetaStoreClient.java
##########
@@ -333,4 +352,9 @@ private boolean hasConnectionLifeTimeReached(Method method) {
     return shouldReconnect;
   }
 
+  @VisibleForTesting
+  public int getRetriesMade(){
+    return this.retriesMade;
+  }
+
 }

Review comment:
       please undo this `retriesMade` to field variable conversion; 
   instead you could probably try to:
   * check the exception being thrown - if that should be different
   * or introduce a specialized `HiveMetaStoreClient` in your test
   
   

##########
File path: standalone-metastore/metastore-common/src/main/java/org/apache/hadoop/hive/metastore/RetryingMetaStoreClient.java
##########
@@ -226,6 +239,11 @@ public Object run() throws MetaException {
         Throwable t = e.getCause();
         if (t instanceof TApplicationException) {
           TApplicationException tae = (TApplicationException)t;
+          if(tae.getMessage().contains("Internal error processing")){

Review comment:
       I think we should not retry any `InvocationTargetException` - or do we have some which should be retried?
   
   or probably you may want to add `TApplicationException.INTERNAL_ERROR` to the `switch` below?




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: gitbox-unsubscribe@hive.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Issue Time Tracking
-------------------

    Worklog Id:     (was: 693808)
    Time Spent: 20m  (was: 10m)

> RetryingMetaStoreClient should not retry connecting to HMS on genuine errors
> ----------------------------------------------------------------------------
>
>                 Key: HIVE-24289
>                 URL: https://issues.apache.org/jira/browse/HIVE-24289
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Rajesh Balamohan
>            Assignee: Harshit Gupta
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> When there is genuine error from HMS, it should not be retried in RetryingMetaStoreClient. 
> For e.g, following query would be retried multiple times (~20+ times) in HMS causing huge delay in processing, even though this constraint is available in HMS. 
> It should just throw exception to client and stop retrying in such cases.
> {noformat}
> alter table web_sales add constraint tpcds_bin_partitioned_orc_10000_ws_s_hd foreign key  (ws_ship_hdemo_sk) references household_demographics (hd_demo_sk) disable novalidate rely;
> org.apache.hadoop.hive.ql.metadata.HiveException: org.apache.thrift.TApplicationException: Internal error processing add_foreign_key
> 	at org.apache.hadoop.hive.ql.metadata.Hive.addForeignKey(Hive.java:5914)
> ..
> ...
> Caused by: org.apache.thrift.TApplicationException: Internal error processing add_foreign_key
>        at org.apache.thrift.TApplicationException.read(TApplicationException.java:111)
>        at org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:79)
>        at org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$Client.recv_add_foreign_key(ThriftHiveMetastore.java:1872)
> {noformat}
> https://github.com/apache/hive/blob/master/standalone-metastore/metastore-common/src/main/java/org/apache/hadoop/hive/metastore/RetryingMetaStoreClient.java#L256
> For e.g, if exception contains "Internal error processing ", it could stop retrying all over again.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)