Hi Ed, 


Tha nk you for the heads up on this problem.  Has this problem occurred ONLY on 
EAVs? 



I am in the in the Modify activity of the ServerPac install for 1.12  and I 
will look into the APAR you mentioned. 



U p time in gereral  is a huge thing for us.  I am up now for 18 months 
straight on a production monoplex .  Our law enforcement customers really like 
that.  So, considering the issues you have seen with PDSe would you think it 
warranted to, in ServerPac, change those datasets that can be changed back to 
PDS instead of PDSE?  Monoplex LPAR, no PAV or EAV.  



Thanks, 

Linda 

----- Original Message -----


From: "Edward Jaffe" <[email protected]> 
To: [email protected] 
Sent: Thursday, October 27, 2011 9:41:23 AM 
Subject: Please Help IBM Fix 'Broken' PDSE Problems 

We have been suffering for a long time with large PDSEs on EAVs spontaneously 
'breaking' every couple/few weeks or months. 

In June of last year, I opened PMR 31744,227,000 where I reported S0F4 RC24 
failures in IGWDACND+1AFA at z/OS 1.11 base level. The recommended maintenance 
was to apply the PTF for OA30338. That did not help. 

Since then we have been hoping someone else would discover and fix the real, 
underlying problem. [Wishful thinking.] It continued unchanged in z/OS 1.12. 
When we saw the problem again under z/OS 1.13, I opened PMR 57302,227,000 to 
report S0F4 RC24 failures in IGWBITX1+10F8. 

This time the service folks pointed me to APAR OA37090. The PDSE development 
team opened this APAR as follow up maintenance to OA30338. The newer APAR, like 
OA30338, is an attempt to addresses errors associated with PDSE VDF index page 
processing. The APAR error description reads: 

"DIAGNOSTIC CODE TO TRAP WRITING OUT BAD PDSE DIRECTORY ENTRIES" 

Clearly IBM is attempting to improve PDSE RAS to catch a corruption issue when 
it actually occurs. Without such logic, it's the job that next uses the PDSE 
that gets abended trying to use the broken PDSE. The APAR also resolves a code 
issue that under certain circumstances allowed a page buffer to be discarded 
twice. 

We have not seen more broken PDSEs since applying this service, but it's only 
been two weeks since the IPL that made it active. 

In any case, I'd like to suggest people install this service in the hope that 
it 
will help YOU avoid the hassle of broken PDSEs and/or better enable the z/OS 
community to assist IBM in tracking down some of these nagging PDSE problems. 

PDSEs are required more and more every day. The sooner we get these issues 
fixed, the better off we will all be. 

-- 
Edward E Jaffe 
Phoenix Software International, Inc 
831 Parkview Drive North 
El Segundo, CA 90245 
310-338-0400 x318 
[email protected] 
http://www.phoenixsoftware.com/ 

---------------------------------------------------------------------- 
For IBM-MAIN subscribe / signoff / archive access instructions, 
send email to [email protected] with the message: GET IBM-MAIN INFO 
Search the archives at http://bama.ua.edu/archives/ibm-main.html 

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to