> -----Original Message-----
> From: IBM Mainframe Discussion List On Behalf Of Barbara Nitz
>
> [ snip ]
> But: I am curious how many other folk delete health checker
> checks because they're just plain stupid.
> I get the impression that almost all components were pressed
> to write 'health checks' come hell or high water. In my
> experience, they're also not willing to listen and fix things:
>
> CHECK(IBMASM,ASM_LOCAL_SLOT_USAGE) : Once this check trips,
> it will trip all the time until you IPL or manage to
> terminate the address space that was put to that local page
> data set. We had this trip on all of our locals, added
> another local page dataset, and then the check still
> complained until the next IPL.
>
> CHECK(IBMXCF,XCF_SIG_STR_SIZE): Using the CFsizer I defined
> the structures for full signalling connectivity between the
> number of systems that are in that sysplex. The check doesn't
> care about that. It takes the arbitrary number that the xcf
> CDS was formatted with and screams that there isn't full
> connectivity for this number of systems. Never mind that we
> don't have that many systems in the plex.
> Once I format the sysplex cds to have exactly the number of
> systems that actually are in the plex, then it screams that
> there is no room for growth.
>
> But what really galled me with this check was that the parm
> ALLOWAUTOALT(YES) on the signalling structures resulted in
> XCF altering the structure so that full signalling
> connectivity was destroyed. IBM refused to fix the
> destruction of the full connectivity. I turned off
> allowautoalt to prevent another occurance of that.
>
> I could go on....
It appears you view the "health checker" as somewhat a "quack"......
:-)
-jc-
----------------------------------------------------------------------
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
