Connect Direct

saikat · New User Joined: 23 Jun 2008 Posts: 26 Location: pune

Hi,

I am using one connect direct process to send file from mainframe to unix server.

This process includes both copy steps and run task.

Sometimes the runtask on remote server takes much time to complete causing runtask failure on mainframe whereas the process on remote server is successful.

It's because the time to execute the remote process is exceeding the C:D time limit and causing runtask failure.

Please advise how to avoid this kind of problem.

Smilies disabled

Bill Woodger · Posted: Mon Nov 19, 2012 1:10 pm

If, when you Preview or look at your post, you see a "smilie" you didn't deliberately code as such, then there is an option to "Disable Smilies in this post" so that we can see what you really meant.

saikat · New User Joined: 23 Jun 2008 Posts: 26 Location: pune

Sorry for the smilie

Ed Goodman · Active Member Joined: 08 Jun 2011 Posts: 556 Location: USA

Suggest you split the job into two parts, one for file transfer and one for the execute task.

Better yet, can you start the task, but break the connection, then have the Unix box send a signal back when the task is complete?

saikat · New User Joined: 23 Jun 2008 Posts: 26 Location: pune

Thanks for the suggestion.

Could you please advise how to implement both options i.e is there any connect direct parameter using which we can achieve this?

We have tried maxdelay=unlimited parameter also but still no luck.

dick scherrer · Posted: Tue Nov 20, 2012 7:53 pm

Hello,

What happened when you split the job into 2 parts? If you have not tried this, suggest you do so and post the result(s) of the test(s).

saikat · New User Joined: 23 Jun 2008 Posts: 26 Location: pune

This is not a problem with job.

The problem is with connect direct script.

It's not possible to trigger runtask on remote system without connect direct command.

If there is any way to split connect direct script please let me know.