[ 
https://issues.apache.org/jira/browse/HBASE-14261?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14725752#comment-14725752
 ] 

Dima Spivak commented on HBASE-14261:
-------------------------------------

Sorry for being so late to the party here, but some comments about the latest 
patch (which looks awesome, [~srikanth235]):

- I agree with Enis, the HDFS user should default to hdfs instead of hadoop.
- Any reason that DistributedHBaseCluster (and HBaseCluster) now has 
waitForZkNodeToStart and waitForDataNodeToStart? I see that there were 
previously just wait... methods for stopping, so I'm wondering if the wait 
shouldn't just be implicit for starts (i.e. do we ever envision wanting to not 
wait for the service to be started before doing things?).
- Nit: "LOG.info("Waiting for service:" + service + " to stop: " + 
serverName.getServerName());" could use a space after the colon in a couple of 
places. There are a few other instances of this where we should probably have a 
space between colons and service names for the sake of readability in the logs 
later. 
- Super nit: a few comments in your new 
ServerAndDependenciesKillingMonkeyFactory.java are missing trailing periods.
- Just as a general best practices question, in HBaseClusterManager, you throw 
an IOException if hbase.it.clustermanager.{zookeeper,hadoop}.home isn't set. 
Does not passing a correct config qualify as an IOException, or should this 
just be a more general Exception?
- Do you mind adding a TODO to RestApiClusterManager to add this same ZK/HDFS 
functionality? I can get to it eventually. :)
- In HBaseMinCluster, the message you log at waitForDataNodeToStop doesn't 
match the method.

Great work, Awesome. ;)

> Enhance Chaos Monkey framework by adding zookeeper and datanode fault 
> injections.
> ---------------------------------------------------------------------------------
>
>                 Key: HBASE-14261
>                 URL: https://issues.apache.org/jira/browse/HBASE-14261
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Srikanth Srungarapu
>            Assignee: Srikanth Srungarapu
>         Attachments: HBASE-14261-branch-1.patch, 
> HBASE-14261-branch-1_v3.patch, HBASE-14261.branch-1_v2.patch
>
>
> One of the shortcomings of existing ChaosMonkey framework is lack of fault 
> injections for hbase dependencies like zookeeper, hdfs etc. This patch 
> attempts to solve this problem partially by adding datanode and zk node fault 
> injections.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to