[ 
https://issues.apache.org/jira/browse/TEPHRA-240?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16163826#comment-16163826
 ] 

ASF GitHub Bot commented on TEPHRA-240:
---------------------------------------

Github user anew commented on a diff in the pull request:

    https://github.com/apache/incubator-tephra/pull/47#discussion_r138490048
  
    --- Diff: 
tephra-core/src/main/java/org/apache/tephra/TransactionManager.java ---
    @@ -206,12 +210,19 @@ public TransactionManager(Configuration conf, 
@Nonnull TransactionStateStorage p
         // TODO: REMOVE WITH txnBackwardsCompatCheck()
         longTimeoutTolerance = conf.getLong("data.tx.long.timeout.tolerance", 
10000);
     
    -    //
    +    ClientIdRetention retention = ClientIdRetention.valueOf(
    --- End diff --
    
    Hmmm. We don't do that for the other configurations... if any of the 
numeric properties cannot be parsed as a number, it also fails. I think it is a 
good idea to fail on invalid configuration, because if there is a configuration 
that is present, then that is most likely with the intention to change the 
default. If there is a typo or some other error, and we only log a warning, 
that warning is likely to go unnoticed and the system will behave in a way that 
was intended, and that will go undetected until it causes failures. 


> TransactionConflictException should contain the conflicting key and client id
> -----------------------------------------------------------------------------
>
>                 Key: TEPHRA-240
>                 URL: https://issues.apache.org/jira/browse/TEPHRA-240
>             Project: Tephra
>          Issue Type: Bug
>            Reporter: Andreas Neumann
>            Assignee: Andreas Neumann
>             Fix For: 0.13.0-incubating
>
>
> Often transaction conflicts are hard to explain. Having the conflicting key, 
> or even the name of the client that performed the concurrent update would 
> greatly help debug.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Reply via email to