Barbara, I don't think we set any limit via IEFUSI but the MEMLIMIT in the SMF parms is set to 0 (the default). Perhaps the main reason why we haven't seen bad effects for DB2 V8.
1. Running my simple driver as a job gets SYSTEM COMPLETION CODE=DC2 REASON CODE=4A001620 2. Running my simple driver as a job with MEMLIMIT=100M works fine 3. Running my simple driver as a STC works fine without any MEMLIMIT setting. So the SMF MEMLIMIT doesn't apply to STC. Did you setup SMF to invoke your IEFUSI for STC? Did you know that z/OS R6 and Java 1.4.2 might require also storage above the bar? If needed for testing I can share my simple driver. Such a display is no my Todo list for SHOWMzOS but time is limited. Roland -----Original Message----- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Barbara Nitz Sent: Thursday, October 27, 2005 2:06 PM To: [email protected] Subject: MEMLIMIT and IEFUSI In early September we had another lengthy discussion about this topic. Now I have gotten around to changing our IEFUSI, after having learned that there is shared storage above the bar that doesn't count against MEMLIM and now that we are starting to use DB2 V8. That brings up two question for me: 1. DB2's DBM1 address space doesn't care what IEFUSI sets for a memlim, it just goes and overwrites the USI specified value with 4TB. (Not this clearly, but it is sort of documented in a DB2 apar.) I would be interested to know how they do it: Is there an official interface (that I just haven't found) or do they go and overwrite the value in the RAX? (Hopefully by also documenting that they did this by setting RAXLVAUTH.) Clearly the Extended addressability guide is wrong when it says that "the MEMLIMIT specified in an IEFUSI exit routine overrides all other MEMLIMIT settings" - so should I turn in an RCF? 2. Is there any way (other than TMONMVS, I believe) to detect that an address spaces uses storage above the bar? (I know - RMF can do it when VSTOR for private is active - this I wouldn't want to do) Given that DB2 is authorized, a bug in there could use up all 4TB even if we were in an AUX storage shortage - and our paging configurations will not be able to handle 4TB page space - no matter what the DB2 apar states. (And just how many mod3 would that be, anyway?) ---------------------------------------------------------------------- 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
