We recently upgraded our Test system LPAR from V6R1 to V7R2.
Everything seemed to have worked fine with the exception of one software package. It was failing for security reasons...then we discovered that a generic profile it uses had access to many QSYS objects/commands.
We didn't see anything in the upgrade guide indicating that a RSTAUT was needed nor could we find anything security-related that was missed during the upgrade. We performed the upgrade using the DVD optical drive...disk-by-disk.
I am thinking that if we had done a RSTAUT at the end of the upgrade (but before a SAVSYS) that we might have avoided this problem. OTOH, I'm not sure where the RSTAUT would get its information from if we didn't restore user profiles from a prior backup.
Dummy me...I thought we might be able to do a SAVSECDTA to a *SAVF and then use that to do a RSTAUT...but that does not appear to be a viable method.
Anyway, before we begin to think about upgrading our Production LPAR...I was hoping someone might have some thoughts about where we went wrong -or- what we should have done different.
Thanks for any input
Terry
Note: I also posted the the MidRange mailing list...but didn't see it come thru...not sure why.
Everything seemed to have worked fine with the exception of one software package. It was failing for security reasons...then we discovered that a generic profile it uses had access to many QSYS objects/commands.
We didn't see anything in the upgrade guide indicating that a RSTAUT was needed nor could we find anything security-related that was missed during the upgrade. We performed the upgrade using the DVD optical drive...disk-by-disk.
I am thinking that if we had done a RSTAUT at the end of the upgrade (but before a SAVSYS) that we might have avoided this problem. OTOH, I'm not sure where the RSTAUT would get its information from if we didn't restore user profiles from a prior backup.
Dummy me...I thought we might be able to do a SAVSECDTA to a *SAVF and then use that to do a RSTAUT...but that does not appear to be a viable method.
Anyway, before we begin to think about upgrading our Production LPAR...I was hoping someone might have some thoughts about where we went wrong -or- what we should have done different.
Thanks for any input
Terry
Note: I also posted the the MidRange mailing list...but didn't see it come thru...not sure why.
Comment