Reexecute processes from failed node

49 views
Skip to first unread message

maced...@gmail.com

unread,
Jun 7, 2013, 4:08:28 AM6/7/13
to sarasvati...@googlegroups.com
Hi Paul,

I'm using Sarasvati from some days by now, I think that would be useful a functionality that, if I'm not wrong, is not present in the current release.
If I'm wrong please correct me.

Suppose you have a graph whose execution process may fail in some node.
In some cases can be useful to reexecute the same process again starting from the node that failed(invoking engine.cancelProcess(token.getProcess()); ).

I'll try to extend the HibEngine class for introduce such a functionality, for my own purposes.
Maybe a more general approach would be a good enhancement for a future release.

Let me know.
Thank you,
   Pasquale.

Paul Lorenz

unread,
Jun 10, 2013, 9:06:29 AM6/10/13
to sarasvati...@googlegroups.com
Ideally if a node execution fails, you should be able to try to fix whatever condition caused the error and then just re-invoke the node token complete.

If you've got another scheme in mind, happy to look at design and/or patches.

Cheers,
Paul



--
You received this message because you are subscribed to the Google Groups "sarasvati-wf-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sarasvati-wf-us...@googlegroups.com.
To post to this group, send email to sarasvati...@googlegroups.com.
Visit this group at http://groups.google.com/group/sarasvati-wf-users?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

maced...@gmail.com

unread,
Jun 17, 2013, 4:20:51 AM6/17/13
to sarasvati...@googlegroups.com
Hi Paul,

I was thinking at a node implementation whose purpose is integrating with a third party component.
In such situation, if this integration fails, I think could be useful to reexecute the process starting from that node.
For identify a failed node I was thinking for searching records in the wf_node_token table whose complete_date is null.

Thank you,
   Pasquale.

Il giorno lunedì 10 giugno 2013 15:06:29 UTC+2, Paul Lorenz ha scritto:
Ideally if a node execution fails, you should be able to try to fix whatever condition caused the error and then just re-invoke the node token complete.

If you've got another scheme in mind, happy to look at design and/or patches.

Cheers,
Paul

On Fri, Jun 7, 2013 at 4:08 AM, <maced...@gmail.com> wrote:
Hi Paul,

I'm using Sarasvati from some days by now, I think that would be useful a functionality that, if I'm not wrong, is not present in the current release.
If I'm wrong please correct me.

Suppose you have a graph whose execution process may fail in some node.
In some cases can be useful to reexecute the same process again starting from the node that failed(invoking engine.cancelProcess(token.getProcess()); ).

I'll try to extend the HibEngine class for introduce such a functionality, for my own purposes.
Maybe a more general approach would be a good enhancement for a future release.

Let me know.
Thank you,
   Pasquale.

--
You received this message because you are subscribed to the Google Groups "sarasvati-wf-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sarasvati-wf-users+unsub...@googlegroups.com.

Paul Lorenz

unread,
Jun 17, 2013, 11:11:59 PM6/17/13
to sarasvati...@googlegroups.com
What I've done in the past is explicitly tracking failed operations.

So in your node execute method, you so something like

try
  // call external service
  engine.complete(token, Arc.DEFAULT_ARC); // won't get called if exception is thrown
}
catch(Exception e) // or whatever exception is appropriate
{
  notifyOfFailure(token);
}

The notifyOfFailure tracks the token and send a notification to the user, who can then see what happened and retry if appropriate.

Cheers,
Paul



To unsubscribe from this group and stop receiving emails from it, send an email to sarasvati-wf-us...@googlegroups.com.

To post to this group, send email to sarasvati...@googlegroups.com.

khalil...@gmail.com

unread,
Nov 19, 2015, 7:45:56 AM11/19/15
to sarasvati-wf-users
Hi Paul,

How the user will do the retry from the failing node in a new process execution?
To unsubscribe from this group and stop receiving emails from it, send an email to sarasvati-wf-users+unsubscribe@googlegroups.com.
To post to this group, send email to sarasvati...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "sarasvati-wf-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sarasvati-wf-users+unsub...@googlegroups.com.
To post to this group, send email to sarasvati...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages