View previous topic :: View next topic
|
Author |
Message |
FazerKing
New User
Joined: 25 Jan 2007 Posts: 18 Location: HK
|
|
|
|
enrico-sorichetti wrote: |
Does Your first jobrun long enough to ...
from Your tso session enter the command
"ISRDDN" or "TSO ISRDDN" if the first one fails,
enter "ENQ"
enter the job name
You will get a list of the resources enqued upon by the job
the resources in red will be held in exclusive control
if Your dataset is in red investigate the program why a shared enq becomes
an exclusive one |
Hi there, I have tried what you suggested and found that both datasets are on the list and one of the dataset is highlighted (in exclusive mode).
The first job has enough time to go. Actually, it is a whole day job which will keep running until 2:00am and it will be refresh itself and run again and again.....until the next refresh time. Anyway, will keep you guys updated if I find anything new. |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
Now the only thing left to do is to find out why the dataset is
under an exclusive enq...
good luck! |
|
Back to top |
|
|
FazerKing
New User
Joined: 25 Jan 2007 Posts: 18 Location: HK
|
|
|
|
Just found the reason. For file sharing, it seems like it's only allowed while online function, but not for batch run. |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
Quote: |
ust found the reason. For file sharing, it seems like it's only allowed while online function, but not for batch run. |
the reason is WRONG but if You are satisfied with it it' s ok for me.. |
|
Back to top |
|
|
FazerKing
New User
Joined: 25 Jan 2007 Posts: 18 Location: HK
|
|
|
|
The reason came up from someone who is one of the senior of the ex company. of course, I'm not satisfied with it. But I can't flight back with a better answer. Did you try to simulate the same situation to run a batch job like this? |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
Quote: |
simulate the same situation to run a batch job like this? |
it happened at a customer location...
never had any problem sharing in the way You are trying...
we sometimes had some read inconsistencies due to a programming glitch
of not enqueuing properly ( buffer refresh issue )
( we also changed the option to share 4 which refreshes the buffers for the read )
but a question comes to my mind ..
do You have on that cluster a aix/path define with the upgrade option ??
that could be the problem.. |
|
Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007 Posts: 10873 Location: italy
|
|
|
|
Quote: |
( we also changed the option to share 4 which refreshes the buffers for the read ) |
please delete the quoted statement from my previous append...
never do brain multitasking ... :-) |
|
Back to top |
|
|
FazerKing
New User
Joined: 25 Jan 2007 Posts: 18 Location: HK
|
|
|
|
Yes.....sorry man....did not mention the alternative index. The program file came with the alternative index.....is that the major problem on file sharing? |
|
Back to top |
|
|
FazerKing
New User
Joined: 25 Jan 2007 Posts: 18 Location: HK
|
|
|
|
Never mind. The sharing is now working. Thanks guys. |
|
Back to top |
|
|
|