User Abend u0777

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

I am faing the user abend that 'U0777'

that means " The application program terminated abnormally because a
potential resource was in the deadlock condition".

Could you please help me out how to fix this.

When we rerun the same job that runs successfully.But while on live it keeps falied at particular time.

I am suspecting the same DB may be uesd by morethan one system

Terry Heinze · Posted: Thu Jan 15, 2009 9:30 pm

What program generated the U0777? Was it a COBOL program using DB2? IMS? Other? Except for a range of codes generated by IMS itself, user abend codes are generated by the source code. That's where you need to look. We need more information.

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

Its a cobol program with IMSDB.

Terry Heinze · Posted: Thu Jan 15, 2009 10:10 pm

Google "IMS user abend codes". Look at all of the messages that were generated by your job (I'm assuming batch). Check the IMS log (or DB2 log if your program also contains DB2). Somewhere you should find what resource caused the deadlock. The more information you can supply, the better your chances are of getting some help.

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

This is not related to DB2 only IMS.This is batch program.

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

The application program terminated abnormally because a potential resource was in the deadlock condition.

This abend is used to terminate an MPP and cause it to be rescheduled. This only occurs when the PPUR=Y option is specified in the DFSVSMxx member and a HALDB partition fails authorization due to a database command being in progress. No TYPE67FF log record is snapped to the log.

System action:
The application program terminates abnormally. If it was running in an MPP or IFP region it will automatically be rescheduled.

System Programmer Response: If a BMP failed, rerun the job.

Hi Terry,
Could you explain on this:what is mean by MPP and PPUR=Y,HALDB.I hope if i know about this,I could solve

Terry Heinze · Posted: Thu Jan 15, 2009 11:47 pm

"IEC130I TD561U01 DD STATEMENT MISSING"
This message in the MVS System Messages Manual Vol. IV looks like it could be the problem. I have little IMS experience myself, so I can only try to point you in the right direction. Do you have coworkers at your shop that can assist?

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

Thanks for your help Terry

Anuj Dhawan · Posted: Fri Jan 16, 2009 3:25 pm

Hello,

There are two things to ask:

1. You used BMP or DLI? (assumption: I believe it was BMP)

2. Check in which step of Your JOB abended, per this statement from the error messages you posted

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

I do not think the probelm with the DD statement misssing.The probel is two or more than systems or batch pgm are triying to use the same database.When i checked in ims abend codes for UO777 which clearly says due to deadlock situtation only this abend comes.

dick scherrer · Posted: Sat Jan 17, 2009 12:36 am

Hello,

If this is a production problem, suggest you work with the scheduling people to schedule the conflicting jobs so they do not execute at the same time.

Anuj Dhawan · Posted: Sat Jan 17, 2009 1:53 pm

Hi,

Usually, deadlocks are due to design problems. There is no "magic" to make them go away. It usually takes considerable effort. As asked earlier, if this is a production problem, one way around the design problems is scheduling. Another is to significantly reduce the runtimes of the problem job(s).

For doing schedule changes, You need to identify correctly the jobs involved in the deadlock at a given moment and How do we do that..can be next logical question - You can generate human-readable deadlock reports using DFSERA10 and DFSERA30 utilities, if you want to.

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

Its fixed now.There are two jobs are tried to pick the IMS db at the same time du to which u0777 abend thrown.

Thanks for your help guys.

Anuj Dhawan · Posted: Tue Jan 20, 2009 11:31 am

Hi,

Thanks for keeping us posted - what solution was used by the way, schedule changes ?..just curious..

vijayakumar subramani · New User Joined: 07 Aug 2008 Posts: 13 Location: Chennai

Yes we did the schedule changes for avoiding contention.

Anuj Dhawan · Posted: Wed Jan 21, 2009 1:05 pm

Thanks for posting the solution..

. This may help others who might be looking for solutions of similar problem...

Ad