1) Check SYS1.PARMLIB(CSFPRM*). It may be pointing to the wrong place. 2) Check the LRECL of the PKDS. 1.11 changed the LRECL of the PKDS. See http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/E0Z2M17A/8.2. 6?DT=20090616151803
HTH. <snip> We are z/os 1.11. We almost never IPL. The last time we IPL'd, we received the following: 11.47.55 STC00014 CSFM450E UNEXPECTED ERROR PROCESSING PKDS, RETURN CODE = 0000000C, REASON CODE = 00001780. 11.47.55 STC00014 CSFM401I CRYPTOGRAPHY - SERVICES ARE NO LONGER AVAILABLE. 11.47.55 STC00014 IEF352I ADDRESS SPACE UNAVAILABLE 11.47.55 STC00014 $HASP395 CSF ENDED We don't use PKI and have no current plans to do so. However, the CSF is critical. With a little experimentation in the testplex, I am currently of the opinion that it is some sort of file sharing issue. When I allocate a fresh PKDS, CSF on LparA comes up just fine. However, CSF on LparB sometimes fails with the above message. The FM seems to say that the PKDS is not completely initialized until the first key is stowed. Not sure how to do that. I'm thinking a PMR. But a user error is usually more likely. Right now my workaround is to point each LPAR to its own PKDS. Of course, I'm a bit nervous as I don't want to accidently break CSF. That would be equivalent to a full outage. What I'd really like to do is to completely shut off PKDS. I've tried starting with no PKDS specified, but CSF refuses to start. </snip> ---------------------------------------------------------------------- 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
