[
https://issues.apache.org/jira/browse/CASSANDRA-8228?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14214020#comment-14214020
]
Rajanarayanan Thottuvaikkatumana commented on CASSANDRA-8228:
-------------------------------------------------------------
[~yukim], Thanks for the advice. Now here is the approach I am planning to
take.
In the prepareForRepair method of the
org.apache.cassandra.service.ActiveRepairService class, before the callback
definition, I am going to define an object like this
{code} AtomicReference<String> failedNode = new AtomicReference<String>();
{code}
Then in the OnFailure method of the callback definition, I am going to collect
the address of the node in the above defined failedNodes variable like this
{code} failedNode.set(from.getHostAddress()); {code}
Then in the error message display section, I will see whether there is any
value in the "failedNodes" variable and if yes, I will display that. Does that
make sense? If yes, please let me know and I will provide a revised patch.
Thanks
> Log malfunctioning host on prepareForRepair
> -------------------------------------------
>
> Key: CASSANDRA-8228
> URL: https://issues.apache.org/jira/browse/CASSANDRA-8228
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Juho Mäkinen
> Assignee: Rajanarayanan Thottuvaikkatumana
> Priority: Trivial
> Labels: lhf
> Attachments: cassandra-trunk-8228.txt
>
>
> Repair startup goes thru ActiveRepairService.prepareForRepair() which might
> result with "Repair failed with error Did not get positive replies from all
> endpoints." error, but there's no other logging regarding to this error.
> It seems that it would be trivial to modify the prepareForRepair() to log the
> host address which caused the error, thus ease the debugging effort.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
