Barbara Based upon past posts, I've gotten (and still have) huge respect for your knowlege and expertise on this platform.
My only guess is that your (negative) experience with Health Checker support may have something to do with how IBM support works outside of the US. In my experience, IBM support has been responsive to specific problems and issues I have brought to their attention. I have certainly heard of differences described by those in AP and EMEA with regards to how IBM support works. I can only say that my experience with IBM US support is usually very good. Another vehicle we all also have is the outstanding relationship that exists today between many IBM z/OS development organizations and the US SHARE User Group. I believe you care about this platform. Please, don't give up. Please help us all improve z/OS for the future. Try to attend SHARE. Or, if it is simply not feasable to attend SHARE in the US, please participate with us in some fashion via electronic means. How about this idea? Can we start new threads to discuss specific checks which are not helpful or are incomplete or deficient in some way? One check per discussion thread? For example, lets say you (or anyone else) has something to say about check USS_AUTOMOUNT_DELAY, just to pick one as an example. We could use a discussion thread convention like this: Subject: Health Check USS_AUTOMOUNT_DELAY problems And then, explain the concern with the check, and let others observe and comment. Then, those of us who have occasion to directly discuss with IBM Health Check issues can have specific suggestions for improvement to pass along. Someone might be inspired to collapse/codify the suggestion into a Requirement. IBM might even monitor this discussion directly, but we certainly don't use IBM-MAIN as an official channel to IBM. Again, please help us make Health Checker better. I think it is in all our interests that z/OS be successful in all installations, and that Health Checker can be an important vehicle to help installations keep their z/OS healthy. Brian On Wed, 12 Sep 2007 08:42:51 +0200, Barbara Nitz wrote: >Kees, > >"About your answer above: why do you check a single PGDS utilization? I >think it hardly hirts when one PGDS is over some limit if the total >configuration is within limit? ... Besides that, the question is what >we can do about it, as Barbara already mentioned." > >thanks for basically asking the same questions I have asked in the ETR I had opened. I was given to understand (not as clearly, of course) that I have no idea what I am talking about. And why do I even question IBMs "best practises"? As this has happened for *every* HC ETR I have ever opened (and also for quite a few emails that I had exchanged in hopes of improving the product) I have resolved to not bother anymore. If I cannot make a check fit, I just delete it. We have a line in every checklist that says 'get HC to shut up'. And we installed the downloadable version before we migrated to 1.6, so we've been putting up with HC for a long time. > >The only reason we still start the STC is that *very few* checks actually do make sense, the RACF_sensitive_resources being among them. Another is the RSM MAXCADS check, as this is the only way to actually see how many CADS are in use short of taking a dump. (It may be that showmvs also reports on this.) > >To me it appears that IBM is promoting the health checker as a way to prevent customers from using the variety of options that z/OS supports (just to make life for the support groups easier). To that effect, every component gets beaten to write a 'health check'. Hence some duplicate checks, some extremely poor documentation, some checks that I consider plain stupid, and a lot that appear to be written hastily (we call that "unloved" in German- "lieblos") and not thought out. If you don't distort your installation to follow those so-called 'best practises', IBM will basically tell you that you're on your own, and if you don't do it, it's your own fault if you have problems. > >I still consider the *idea* of looking at best practises very valid and very good, but not how HC implements it, and not what I consider IBMs closedmindedness about the product and its checks. > >My 2 cents, and I'll stop now. > >Regards, Barbara ---------------------------------------------------------------------- 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
