| Author |
Message
|
| vinbud117 |
 Posted: Thu Oct 13, 2005 5:23 am Post subject: connection loss to sql table |
 |
|
Acolyte
Joined: 22 Jul 2005
Posts: 61
|
The database is on another system. At times, the message flow cannot establish the connection and hence sends the message to the error queue, which is totally understandable.
If I pick this message and put it again it is processed successfully. What this means is that the problem of not getting the connection is temporory.
What can be done to avoid the failure???? is there any way where in i can have the message reprocessed automatically after sometime, say a few minutes.
regards,
Vinay. |
|
| Back to top |
|
 |
| jefflowrey |
| Posted: Thu Oct 13, 2005 5:32 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
You can't have it reprocessed after some time, without writing code of some sort.
If you set up a backout queue and a backout retry count, you can have it reprocessed immediately, which might be enough to get it to succeed.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| vinbud117 |
| Posted: Thu Oct 13, 2005 6:05 am Post subject: |
|
|
Acolyte
Joined: 22 Jul 2005
Posts: 61
|
| will it go to the backout queue even when the Error terminals are connected to the SUB_ERROR_HANDLER. In my scenario, all failure terminals are connected to the error handling subflow. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Thu Oct 13, 2005 6:15 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
No, you'd probably have to change your error handling somewhat.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| fschofer |
| Posted: Thu Oct 13, 2005 3:14 pm Post subject: |
|
|
Knight
Joined: 02 Jul 2001
Posts: 524
Location: Mainz, Germany
|
|
| Back to top |
|
|
| wschutz |
| Posted: Thu Oct 13, 2005 3:22 pm Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
re Timeout nodes:
I'm playing with a programming pattern where-by you connect the "catch" node to a TimeoutControl node, which allows you to specify a period of time you want to wait. After that time expires, a TimeoutNotification node can propagate the original message again to the original flow.
Of course, its more complicated that what I just describe. You need to build a timeoutcontrol message to specify the wait interval, you probably want a filter node to do this for only certain exceptions (like db unavilable) and to prevent "poison" messages, and some small shuffling of the original message, but it should work nicely.
_________________
-wayne |
|
| Back to top |
|
|
|
|
