This post is more than 5 years old
18 Posts
0
657
CSL 9.0.0 and BCSCHECK
Greetings.
I've made a DASD-initializationprocedure that, a.o., tries to get rid of all the VVDS-catalogentries for all the volumes I've just hit with a DFDSS INIT. As their old volser doesn't exist anymore, we do not want any old VVDS-entries floating around. Wherever they might be.
I do this by generating a:
BCSCHECK ACTION=FAST-DELETE,
GDVO,
INCLUDEVVDS,
INCCAT=**,
INCDSN=SYS1.VVDS.Vxxxxxx,
INCDSN=SYS1.VVDS.Vyyyyyy
This cleanup runs to completion without a hitch. That is, almost always. It does happen, however, that this operation suddenly "hangs" no IO, no CPU, nothing. It seems to be waiting for something, but what? We then cancel the job and everything seems to be OK: no garbled catalogs or anythng like that.
Anybody seen anything like this?
Willem Vermeer
ING, the Netherlands
I've made a DASD-initializationprocedure that, a.o., tries to get rid of all the VVDS-catalogentries for all the volumes I've just hit with a DFDSS INIT. As their old volser doesn't exist anymore, we do not want any old VVDS-entries floating around. Wherever they might be.
I do this by generating a:
BCSCHECK ACTION=FAST-DELETE,
GDVO,
INCLUDEVVDS,
INCCAT=**,
INCDSN=SYS1.VVDS.Vxxxxxx,
INCDSN=SYS1.VVDS.Vyyyyyy
This cleanup runs to completion without a hitch. That is, almost always. It does happen, however, that this operation suddenly "hangs" no IO, no CPU, nothing. It seems to be waiting for something, but what? We then cancel the job and everything seems to be OK: no garbled catalogs or anythng like that.
Anybody seen anything like this?
Willem Vermeer
ING, the Netherlands
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
November 3rd, 2009 07:00
Jon Scheffter
Technical Support Engineer 4
GTS Solutions -- PREM Symm Mainframe Software
Message was edited by:
OldCSLguy
Davidy8s
154 Posts
0
October 29th, 2009 08:00
Have you noticed a resource contention while this is going on such as in GRS, or MIM ? Also, what kind of time frame; does it normally run for 2 minutes, 10?
What about collateral activity also? Is there any system wide maintenance going on while this happens?
Dave Yates
EMC TSE3
Benevolten Host S/W & Mainframe Forum Moderator
"Il Moderatore Benevolo"
Anonymous
5 Practitioner
5 Practitioner
•
274.2K Posts
0
October 29th, 2009 09:00
Jon Scheffter
Technical Support Engineer 4
GTS Solutions -- PREM Symm Mainframe Software
Message was edited by:
OldCSLguy
WillemV1
18 Posts
0
November 3rd, 2009 06:00
No concurrent system wide maintenance going on.
I did a D GRS,CONTENTION,ENQ and there does seem to be contention going on between my job and the CATALOG addressspace. At first I thought that my job had found those particular entries in this catalog and was trying to delete them and that that was the reason for the ENQ-conflict.
However, when I look at the CSL-output, CSL already seems to have finished processing this catalog (not having found any of the entries). So, why is it waiting for CATALOG to release the usercat? Why is it going for a repeat-performance? Also, why does CATALOG have an Exclusive on this usercat anyway and not on any of the others? CSL didn't have any problems processing any of the others.
By the way, after I cancelled the job I SCANned the usercat for those entries and didn't find any of them. And, as I said, CSL BCSCHECK didn't find anything either.
I guess I'll be going for a systemdump after I manage to ged rid of Dumpmaster.