> > If your specify CONSOLE=(SYSC) with no devices, the default console
> >01F is automatically added. That seems undesirable (and I
> >have already received one customer complaint about it), but I
> >am not at liberty to incompatibly change the behavior, even
> >if it doesn't make sense.
> Yes, I saw the mnote in the assembly. The job ended cc=0, anyway,
> and I decided that that's just default behaviour. I found it strnage,
though.
> Why are others at liberty for incompatible changes?!?
When I say "I am not at libertv", what I really mean is
that I discussed it with Peter Relson (because he is one of
the most careful about compatibility people that I know),
and he recommended against making the change, and I accepted
his recommendation. I could still make a compatible change
by allowing some specification like CONSOLE=(SYSC,NODEVICES).
That require a bit more code in the AMDSADMP macro (in
HLASM macro language, which is not my favorite
programming language). So that is somewhere on my list
of things to to if/when I get around to working on SADMP stuff
again. In the meantime, I recommend that you specify a
device number which does not exist, or, if you are using
all 65,536 device numbers, pick one which is unlikely
to present an interrupt before you can click SEND on the
HMC after you IPL SADMP.
>
> > This will dump all paged out storage:
> >
> >DUMP=('DSP OF ASID(ALL) ALSO PAGETABLES OF DATASPACES X
> >ALSO SP(ALL) IN ASID(ALL) X
> >ALSO HIGH VIRTUAL IN ASID(ALL)'), X
> For performance/timing reasons I had held off sadumping all paged
> out storage. Anyone out there have numbers how long it takes for a -
> say - 8GB real lpar with aux usage at 40% (on 12 locals of 49500
> tracks each) in comparison to only dumping paged-in pages?
I calculate that you have about 2784MB of aux, and
40% of that is about 1114MB. The rate at which SADMP can
read this from the page data sets depends on the I/O
configuration (channel type and speed, generation of DASD
technology, etc). I would recommend that you run
VERBX SADMPMSG 'STATS' under IPCS against an existing dump
from the system of interest. Look for the section of output
titled "Paged Out Virtual Dump Statistics". Under that,
find the line labelled "Average output data rate".
Use this rate to extrapolate how long it would take to
dump 1114MB. It will probably be considerably longer
that you want to wait. That's why we make it an option.
The 4K blocksize used for page data sets is quite
suboptimal for Ficon channels. Use of zHPF by SADMP
for reading from page data sets might improve this, but
that is not an item in our plan at this time. As always,
feel free to submit requirements to IBM.
A general note about the output of
VERBX SADMPMSG 'STATS' :
I originally created this for my own use. It contains some
easily understood data which could be used by anyone,
and a lot of internal information. The information whose
meaning is clear to you is intended for both of us. The
rest is intended for me.
Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
