View previous topic :: View next topic
|
Author |
Message |
findmadhav
New User
Joined: 02 May 2006 Posts: 7
|
|
|
|
Hi,
I am dealing with an application which consists of several CICS programs. Now each of these programs writes, reads and rewrites files.
Now at a point in the program chain, one of the programs keeps a file locked and the next program is not able to write into that file.
I have used EXEC CICS UNLOCK FILE and also SYNCPOINTS in my programs (for read-update and rewrite). But this condition still persists.
Is there a way to solve this problem. Also, is it possible to know which particular operation in which program is locking this file??? Please help me with this... |
|
Back to top |
|
|
ofer71
Global Moderator
Joined: 27 Dec 2005 Posts: 2358 Location: Israel
|
|
|
|
I assume you are talking about a VSAM files.
Check the STRINGS value for that file. You can do it using:
"CECI IN FILE(file_name)"
The you can set STRINGS to a higher number using CECI, DFHCSDUP or any other resource update method.
O. |
|
Back to top |
|
|
neoconvoy
New User
Joined: 22 Jul 2006 Posts: 9
|
|
|
|
We recently also encountered this problem. We are working in a CICSPLEX environment, and one of our files had been locked by a certain task/job/user which we were not able to locate.
The problem started like this :
- Some online transaction initiated which was updating an online file. This transaction could be under Intertest. I am not too sure though.
- Tried to close the file in the Application Owning Regions (AORs) using CEMT S CLOSE, but managed to close all except one, which gave the status PENDING REQUEST.
- When the another transaction was started that need to update the file, it abended with AFCV.
- We tried to issue an UNLOCK, but the above situations persisted.
Questions:
1. Is there any way to find out who had the lock on the file ? So far, we can only managed to detect the AORs using the file.
2. Is there a way to force the CLOSE ?
Thanks ! |
|
Back to top |
|
|
findmadhav
New User
Joined: 02 May 2006 Posts: 7
|
|
|
|
We tried a workaround for the above problem. Instead of directly doing operations on the files, we introduced online access routines, wherein we used syncpoints and unlock statements. It seemed to work. |
|
Back to top |
|
|
|