cxf-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Cyrille Le Clerc (JIRA)" <j...@apache.org>
Subject [jira] Updated: (CXF-2141) Expose faults count in ResponseTimeCounter in addition to existing invocations count
Date Sun, 29 Mar 2009 19:00:50 GMT

     [ https://issues.apache.org/jira/browse/CXF-2141?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Cyrille Le Clerc updated CXF-2141:
----------------------------------

    Attachment: CXF-2141.patch

Proposed patch to expose fault counts in ResponseTimeCounter. Description of the patch :

1) ResponseTimeCounter & ResponseTimeCounterMBean : add getNumCheckedApplicationFaults(),
getNumLogicalRuntimeFaults(), getNumRuntimeFaults(), getNumUnCheckedApplicationFaults()

2) org.apache.cxf.management.interceptor.AbstractMessageResponseTimeInterceptor & MessageHandlingTimeRecorder
 : propagate the faultMode property from the Message to MessageHandlingTimeRecorder  via the
Exchange to hold the type of fault if one occured during the exchange.

3) ResponseTimeMessageInInterceptor, ResponseTimeMessageInvokerInterceptor & ResponseTimeMessageOutInterceptor
: implement handleFault(message) method to track faults

5) add unit tests and fix few existing easymock recording that have changed.

> Expose faults count in ResponseTimeCounter in addition to existing invocations count
> ------------------------------------------------------------------------------------
>
>                 Key: CXF-2141
>                 URL: https://issues.apache.org/jira/browse/CXF-2141
>             Project: CXF
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 2.1.4, 2.2
>            Reporter: Cyrille Le Clerc
>         Attachments: CXF-2141.patch
>
>
> Expose faults count in ResponseTimeCounter  in addition to existing invocations count.
Faults can be split by  FaultMode (RUNTIME_FAULT, LOGICAL_RUNTIME_FAULT, CHECKED_APPLICATION_FAULT,
UNCHECKED_APPLICATION_FAULT)  to improve accuracy.
> Monitoring evolution of the faults with the granularity of the type of fault (runtime
fault, checked application fault, unchecked application fault) is key to our operations team.
If the application's SOAP runtime does not expose such indicators, ours operations team looks
at log files parsing or "SOA Appliances" like Data Power.

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


Mime
View raw message