[
https://issues.apache.org/jira/browse/HAWQ-1372?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15903678#comment-15903678
]
ASF GitHub Bot commented on HAWQ-1372:
--------------------------------------
Github user dyozie commented on a diff in the pull request:
https://github.com/apache/incubator-hawq-docs/pull/97#discussion_r105213169
--- Diff: markdown/admin/ambari-admin.html.md.erb ---
@@ -206,44 +208,22 @@ First ensure that the new node(s) has been configured
per the instructions found
23. (Optional.) If you enabled temporary password-based authentication
while preparing/configuring your HAWQ host systems, turn off password-based
authentication as described in [Apache HAWQ System
Requirements](../requirements/system-requirements.html#topic_pwdlessssh).
-#### <a id="manual-config-steps"></a>Manually Updating the HAWQ
Configuration
-If you need to expand your HAWQ cluster without restarting the HAWQ
service, follow these steps to manually apply the new HAWQ configuration. (Use
these steps *instead* of following Step 7 in the above procedure.):
+#### <a id="manual-config-steps"></a>Cluster Expansion Workaround
+If you need to expand your HAWQ cluster without restarting the HAWQ
service, follow this procedure to manually apply the HAWQ configuration changes
required for this operation. (Perform these steps *instead of* following Step
9b in the above procedure.)
-1. Update your configuration to use the new
`default_hash_table_bucket_number` value that you calculated:
- 1. SSH into the HAWQ master host as the `gpadmin` user:
- ```shell
- $ ssh gpadmin@<HAWQ_MASTER_HOST>
- ```
- 2. Source the `greenplum_path.sh` file to update the shell environment:
- ```shell
- $ source /usr/local/hawq/greenplum_path.sh
- ```
- 3. Verify the current value of `default_hash_table_bucket_number`:
- ```shell
- $ hawq config -s default_hash_table_bucket_number
- ```
- 4. Update `default_hash_table_bucket_number` to the new value that you
calculated:
- ```shell
- $ hawq config -c default_hash_table_bucket_number -v <new_value>
- ```
- 5. Reload the configuration without restarting the cluster:
- ```shell
- $ hawq stop cluster -u
- ```
- 6. Verify that the `default_hash_table_bucket_number` value was updated:
- ```shell
- $ hawq config -s default_hash_table_bucket_number
- ```
-2. Edit the `/usr/local/hawq/etc/slaves` file and add the new HAWQ
hostname(s) to the end of the file. Separate multiple hosts with new lines. For
example, after adding host4 and host5 to a cluster already contains hosts 1-3,
the updated file contents would be:
+1. Perform the procedure identified in [Updating HAWQ Configuration
Without Cluster Restart](#manual-cfg-norestart) to update your configuration to
use the new `default_hash_table_bucket_number` value that you calculated and
set via the Ambari UI. You may skip Step 1 of this procedure.
--- End diff --
I think this is going to be a lot for a user to navigate (jump out from one
procedure into the workaround, then jump to another procedure, skipping a step,
return to finish the workaround, then back to the finish the original).
Instead of redirecting them a second time here, let's just repeat the necessary
cli steps right here (hawq config-s, hawq config -c, hawq stop --reload)
focusing just on the default_hash_table_bucket_number setting. It's only a
couple of commands and I think it will be much easier to deal with.
> doc ambari hawq config change procedure that does not require cluster restart
> -----------------------------------------------------------------------------
>
> Key: HAWQ-1372
> URL: https://issues.apache.org/jira/browse/HAWQ-1372
> Project: Apache HAWQ
> Issue Type: Improvement
> Components: Documentation
> Reporter: Lisa Owen
> Assignee: David Yozie
> Priority: Minor
>
> document the workaround for updating hawq configuration via ambari (for
> ambari-managed clusters) in cases where a complete cluster restart cannot be
> tolerated:
> update config via ambari, do not restart
> update config via "hawq config -c xxx -v xxx"
> hawq stop cluster --reload
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
