tephra-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (TEPHRA-266) Identify log messages when multiple instances of Tephra run on a single HBase cluster
Date Thu, 26 Apr 2018 09:18:00 GMT

    [ https://issues.apache.org/jira/browse/TEPHRA-266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16453733#comment-16453733
] 

ASF GitHub Bot commented on TEPHRA-266:
---------------------------------------

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

    https://github.com/apache/incubator-tephra/pull/74#discussion_r184322363
  
    --- Diff: tephra-core/src/main/java/org/apache/tephra/coprocessor/TransactionStateCache.java
---
    @@ -184,4 +185,14 @@ private void refreshState() throws IOException {
       public TransactionVisibilityState getLatestState() {
         return latestState;
       }
    +
    +  protected void setId(@Nullable String id) {
    +    if (id != null) {
    +      this.logPrefix = "[" + id + "] ";
    +    }
    +  }
    +
    +  private String prefixLog(String message) {
    --- End diff --
    
    not sure this is a very good idea. It means you are performing the string operations even
when it is not being logged (for example, for debug messages). Better to add the logPrefix
as an argument to the log message, such as:
    ```
    LOG.debug("[{}] Latest transaction snapshot: {}", logPrefix, latestState.toString()));
    ``` 



> Identify log messages when multiple instances of Tephra run on a single HBase cluster
> -------------------------------------------------------------------------------------
>
>                 Key: TEPHRA-266
>                 URL: https://issues.apache.org/jira/browse/TEPHRA-266
>             Project: Tephra
>          Issue Type: Improvement
>         Environment: 
>            Reporter: Poorna Chandra
>            Assignee: Poorna Chandra
>            Priority: Major
>             Fix For: 0.14.0-incubating
>
>
> When multiple instances of Tehpra are run in a single HBase cluster, the log messages
from the co-processors of all instances (which are identical except for the timestamp) go
into a single region server log file. It becomes very difficult to figure out issues of an
instance due to this.
> it would be good if instance id is logged along with the co-processor log messages so
that we can identity which messages belong to which instance of Tephra.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message