Hard IPL on partition with SAP instances

Question

Hello,
Over the weekend we had a partition IPL itself (for reasons still trying to be determined - working with IBM).  Below is a list of things I'm compiling for us to walk through should this happen again (we hope it does not).  Volker helped us already by suggesting some things and we appreciate his help.  Does anyone have experience with this or any ideas of other things we should add to the list?  I appreciate any feedback.
~TJ
u2022     Before Starting SAP
o     Refer to SAP note: 71258 and run a RCLSTG *DBXREF for the database cross reference files
o     SAVLIB for all SAP libraries to see if there are any damaged objects
o     Delete SQL packages for each SAP system
o     Reapply SAP kernel to each SAP system
u2022     After Starting SAP
o     Check the following transactions:
     SICK
     ST22
     SM21
     SM50
     DB02
o     Follow this link for how to do iSeries database consistency checks: http://help.sap.com/saphelp_erp2005vp/helpdata/en/c4/3a7811505211d189550000e829fbbd/frameset.htm
u2022     Consistency Check for database: Goto -> Check -> Installation -> ABAP Dictionary <->
u2022     If the consistency check finds inconsistent objects, go to the object in SE14 and activate / recreate / whatever is necessary to fix it.
u2022     In the past, we had a transport fail because the object didnu2019t exist in SAP and nothing appeared in SE14.  The object still existed at the OS level u2013 where we had to first delete it, save a copy in your personal library of all objects before deleting at the OS level.  Then the transport could successfully create the object and SAP and the OS were both consistent
     STMS
u2022     Check TMS connections / setup
u2022     Attached partitions
o     When the hosting partition comes up, the hosted partitions will still be down, but theyu2019ll act like they are still running in: WRKNWSD
o     Login to the HMC and manually stop all hosted partitions.
o     Once down, start the hosted partition via WRKNWSD like normal
u2022     Other
o     Contact IBM and open a ticket to troubleshoot the cause of the outage
o     Find and save the MSD (main storage dump) if there is one, to send IBM
o     Check the LIC log (license internal code) and save to send to IBM

Answer 1

Hi TJ,
I had a few hard IPL due to a faulty UPS before. Luckily it was not the production system, and the only thing I did was deleting the SQL packages - for performance, not for availability.
If you have Volker on board, nothing to worry. Unless IBM says that you have to rebuild the system/database, and you don't have backups.
Best regards,
Victor