[jira] Updated: (OPENJPA-376) Need more trace for transaction demarcation

["Patrick Linskey (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/OPENJPA-376?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Patrick Linskey updated OPENJPA-376:
------------------------------------

    Fix Version/s:     (was: 1.0.2)

Removing from 1.0.2 as there doesn't seem to be much activity on this enhancement.

Regarding the enhancement itself: what events in particular are needed? In my experience, turning on SQL and JDBC logging has provided me with assurances about data getting sent to the database, and this includes transaction and thread ID information in the log.

Regarding the question about what trace channel to use: Runtime sounds reasonable; if it's already pretty noisy, then creating a new Transaction channel sounds reasonable too.

> Need more trace for transaction demarcation
> -------------------------------------------
>
>                 Key: OPENJPA-376
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-376
>             Project: OpenJPA
>          Issue Type: Bug
>          Components: jdbc
>    Affects Versions: 1.0.0
>            Reporter: Kevin Sutter
>            Assignee: Kevin Sutter
>             Fix For: 1.1.0
>
>
> As we are debugging problems in an application server environment (ie. WebSphere), it is becoming apparent that we need more trace output when demarcating transaction begin/commit/rollback.  I have had to debug several problems relating to whether data updates were actually pushed out to the database or not.  Without turning on additional trace mechanisms in other software components (WebSphere, the database itself, the application, etc), it's impossible to tell what OpenJPA is processing as far as the transaction is concerned.  This becomes increasingly important when dealing with multiple clients and multiple transactions and knowing what gets processed in what order.
> Question:  What trace channel should I use?  Should I go with the general openjpa.Runtime channel?  Or, should I create a new channel (openjpa.Transaction)?  Seems like overkill.  But, if people are concerned about the amount of trace data, we could go this route.  Suggestions?
> Kevin

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.