You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by Peter Vary <pv...@cloudera.com> on 2016/10/28 13:56:05 UTC
Review Request 53261: HIVE-15090 Temporary DB failure can stop
ExpiredTokenRemover thread
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53261/
-----------------------------------------------------------
Review request for hive, Aihua Xu, Piotr Wikie?, and Thejas Nair.
Bugs: HIVE-15090
https://issues.apache.org/jira/browse/HIVE-15090
Repository: hive-git
Description
-------
Moved the try/catch into the while cycle, so the thread will keep running after the exception.
Added an extra sleep when there is an exception so we do not end up busy waiting for the solution of the problem which caused the exception
Diffs
-----
shims/common/src/main/java/org/apache/hadoop/hive/thrift/TokenStoreDelegationTokenSecretManager.java abe8cc2
Diff: https://reviews.apache.org/r/53261/diff/
Testing
-------
I have tested with the following configuration:
<property>
<!-- So the tokens are stored in database -->
<name>hive.cluster.delegation.token.store.class</name>
<value>org.apache.hadoop.hive.thrift.DBTokenStore</value>
</property>
<property>
<!-- So the removal happens often -->
<name>hive.cluster.delegation.token.gc-interval</name>
<value>10</value>
</property>
Started the Metastore, and after a while I have stopped the database. When the database was turned off the metastore logged the ExpiredTokenRemover thread received unexpected exception message, after the database restarted the messages are stopped, and everything become normal. In the debugger I have verified the thread is running.
Thanks,
Peter Vary
Re: Review Request 53261: HIVE-15090 Temporary DB failure can stop
ExpiredTokenRemover thread
Posted by Aihua Xu <ax...@cloudera.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53261/#review154126
-----------------------------------------------------------
Ship it!
Ship It!
- Aihua Xu
On Oct. 28, 2016, 2:31 p.m., Peter Vary wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/53261/
> -----------------------------------------------------------
>
> (Updated Oct. 28, 2016, 2:31 p.m.)
>
>
> Review request for hive, Aihua Xu, Piotr Wikie?, and Thejas Nair.
>
>
> Bugs: HIVE-15090
> https://issues.apache.org/jira/browse/HIVE-15090
>
>
> Repository: hive-git
>
>
> Description
> -------
>
> Moved the try/catch into the while cycle, so the thread will keep running after the exception.
> Added an extra sleep when there is an exception so we do not end up busy waiting for the solution of the problem which caused the exception
>
>
> Diffs
> -----
>
> shims/common/src/main/java/org/apache/hadoop/hive/thrift/TokenStoreDelegationTokenSecretManager.java abe8cc2
>
> Diff: https://reviews.apache.org/r/53261/diff/
>
>
> Testing
> -------
>
> I have tested with the following configuration:
>
> <property>
> <!-- So the tokens are stored in database -->
> <name>hive.cluster.delegation.token.store.class</name>
> <value>org.apache.hadoop.hive.thrift.DBTokenStore</value>
> </property>
> <property>
> <!-- So the removal happens often -->
> <name>hive.cluster.delegation.token.gc-interval</name>
> <value>10</value>
> </property>
>
> Started the Metastore, and after a while I have stopped the database. When the database was turned off the metastore logged the ExpiredTokenRemover thread received unexpected exception message, after the database restarted the messages are stopped, and everything become normal. In the debugger I have verified the thread is running.
>
>
> Thanks,
>
> Peter Vary
>
>
Re: Review Request 53261: HIVE-15090 Temporary DB failure can stop
ExpiredTokenRemover thread
Posted by Peter Vary <pv...@cloudera.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53261/
-----------------------------------------------------------
(Updated Oct. 28, 2016, 2:31 p.m.)
Review request for hive, Aihua Xu, Piotr Wikie?, and Thejas Nair.
Changes
-------
Spelling issues pointed out by Zsombor, and fixed
Bugs: HIVE-15090
https://issues.apache.org/jira/browse/HIVE-15090
Repository: hive-git
Description
-------
Moved the try/catch into the while cycle, so the thread will keep running after the exception.
Added an extra sleep when there is an exception so we do not end up busy waiting for the solution of the problem which caused the exception
Diffs (updated)
-----
shims/common/src/main/java/org/apache/hadoop/hive/thrift/TokenStoreDelegationTokenSecretManager.java abe8cc2
Diff: https://reviews.apache.org/r/53261/diff/
Testing
-------
I have tested with the following configuration:
<property>
<!-- So the tokens are stored in database -->
<name>hive.cluster.delegation.token.store.class</name>
<value>org.apache.hadoop.hive.thrift.DBTokenStore</value>
</property>
<property>
<!-- So the removal happens often -->
<name>hive.cluster.delegation.token.gc-interval</name>
<value>10</value>
</property>
Started the Metastore, and after a while I have stopped the database. When the database was turned off the metastore logged the ExpiredTokenRemover thread received unexpected exception message, after the database restarted the messages are stopped, and everything become normal. In the debugger I have verified the thread is running.
Thanks,
Peter Vary
Re: Review Request 53261: HIVE-15090 Temporary DB failure can stop
ExpiredTokenRemover thread
Posted by Barna Zsombor Klara <zs...@cloudera.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53261/#review154125
-----------------------------------------------------------
Ship it!
LGTM.
Thanks for the patch.
- Barna Zsombor Klara
On Oct. 28, 2016, 1:56 p.m., Peter Vary wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/53261/
> -----------------------------------------------------------
>
> (Updated Oct. 28, 2016, 1:56 p.m.)
>
>
> Review request for hive, Aihua Xu, Piotr Wikie?, and Thejas Nair.
>
>
> Bugs: HIVE-15090
> https://issues.apache.org/jira/browse/HIVE-15090
>
>
> Repository: hive-git
>
>
> Description
> -------
>
> Moved the try/catch into the while cycle, so the thread will keep running after the exception.
> Added an extra sleep when there is an exception so we do not end up busy waiting for the solution of the problem which caused the exception
>
>
> Diffs
> -----
>
> shims/common/src/main/java/org/apache/hadoop/hive/thrift/TokenStoreDelegationTokenSecretManager.java abe8cc2
>
> Diff: https://reviews.apache.org/r/53261/diff/
>
>
> Testing
> -------
>
> I have tested with the following configuration:
>
> <property>
> <!-- So the tokens are stored in database -->
> <name>hive.cluster.delegation.token.store.class</name>
> <value>org.apache.hadoop.hive.thrift.DBTokenStore</value>
> </property>
> <property>
> <!-- So the removal happens often -->
> <name>hive.cluster.delegation.token.gc-interval</name>
> <value>10</value>
> </property>
>
> Started the Metastore, and after a while I have stopped the database. When the database was turned off the metastore logged the ExpiredTokenRemover thread received unexpected exception message, after the database restarted the messages are stopped, and everything become normal. In the debugger I have verified the thread is running.
>
>
> Thanks,
>
> Peter Vary
>
>