synapse-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Charith Dhanushka Wickramarachchi (JIRA)" <j...@apache.org>
Subject [jira] Updated: (SYNAPSE-445) LB endpoints can retry forever.
Date Mon, 09 Mar 2009 05:40:50 GMT

     [ https://issues.apache.org/jira/browse/SYNAPSE-445?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Charith Dhanushka Wickramarachchi updated SYNAPSE-445:
------------------------------------------------------

    Attachment: patch.txt

patch to resolve this issue 

This resolve this issue by making user enable to set the maximum number of consecutive faults
 per endpoint.
if this is set and If a given end point in Load balance algorithm failed consecutively to
that number it will be assumed as a not working endpoint and will be removed from the algorithm.
 

ex:
<loadbalance maxFaults="3">
                        <endpoint>
                            <address uri="http://localhost:9001/services/LBService1">
                                <enableAddressing/>
                                <suspendDurationOnFailure>1</suspendDurationOnFailure>
                            </address>
                        </endpoint>
                        <endpoint>
                            <address uri="http://localhost:9002/services/LBService1">
                                <enableAddressing/>
                                <suspendDurationOnFailure>1</suspendDurationOnFailure>
                            </address>
                        </endpoint>
 </loadbalance> 

> LB endpoints can retry forever.
> -------------------------------
>
>                 Key: SYNAPSE-445
>                 URL: https://issues.apache.org/jira/browse/SYNAPSE-445
>             Project: Synapse
>          Issue Type: Bug
>          Components: Endpoints
>    Affects Versions: 1.1, 1.1.1, 1.2, 1.3
>            Reporter: Jason Walton
>            Priority: Minor
>         Attachments: patch.txt
>
>
> If a message to a SALoadBalanceEndpoint encounters an error, the SALoadBalanceEndpoint.onChildEndpointFail
method will try to resend the message so long as another endpoint is active.  If, however,
the suspendDurationOnFailure for an endpoint is sufficiently short, and all endpoints in the
group are failing (say, because the destination endpoints are down), then Synapse will retry
the message forever.

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


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@synapse.apache.org
For additional commands, e-mail: dev-help@synapse.apache.org


Mime
View raw message