We had a similar issue here: occasionally a mountpoint directory and/or associated symbolics went missing after an OS upgrade and a new root file system was moved into production / development systems.
One of the system programmers had written a job that fired off after IPL; that job called a REXX that scanned the current BPXPRM's and issued a console message if expected filesystems were not mounted. I extended that by writing a REXX that was called from /etc/rc that went a step further. If an expected filesystem was not mounted, it consulted another parm file that listed the necessary mountpoints/symbolics for the filesystems and attempted to create them and then remount the missing filesystem. If the remount failed an email was issued. That works quite well for us, provided the parm file is updated whenever BPXPRM's are changed... I put a big comment box at the top of BPXPRM's as a reminder. Ant. Northern Territory Govt., Australia -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of SAURABH KHANDELWAL Sent: Wednesday, 13 April 2011 4:20 AM To: [email protected] Subject: Copy OMVS file system after service to production system Hello, I have followed the suggested way to make copy root file system before applying service. But we have couple of z/OS running with same version of OS level (example z/OS 1.8). In this we have test system, where we apply RSU and test it. We also have couple of production system with same OS level. In production system there might be changes in sub directories and Symbolic link. So the problem is, If I just copy the root file system from test system to production system after applying service on test root file system, there might be a chances to loose some symbolic link and sub directories. The solution which I can think of is, manually checking each of the production system and note all the sub directories and symbolic detail. Is there any other way, which can reduce my lots of manual effort and make process easy. Regards Saurabh Khandelwal On 4/12/2011 10:57 PM, Shmuel Metz (Seymour J.) wrote: > In<[email protected]>, on 04/11/2011 > at 11:18 PM, SAURABH KHANDELWAL<[email protected]> > said: > >> Can you please suggest me , what approach I should follow >> before applying RSU to make sure, everything will work in root file >> system after that. > How was your system originally installed? How was the last service > done? In paricular, is your root the original target or a copy? > > As others have pointed out, you should never install service on the > running system. You should have a deployment plan for either rotating > among a set of target libraries or for cloning target libraries, > including Unix file systems. > ---------------------------------------------------------------------- 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
