Yep, that was it.

VM crashed (I played around too much) so the problem is solved.  This
machine is the center of a PNET star of the other 8 VSE machines. 
Manually ftping the jobs around got old, quick.

I had a count of 5.

When looking at the console logs, I see that Power ran out of 24 bit
space.  No big deal except that I had PNET transfers from 3 other VSE
machines active at that time.  I'm guessing that the PNET transfers were
the active I/O load when the machine was forced off.

The Operator didn't correct the situation, 
didn't try to shutdown the machine normally,
just forced it off.

Just taking any partition down, would have freed up the 24 bit area and
allowed things to continue.

At least that is my guess.

Tom Duerbusch
THD Consulting

>>> [EMAIL PROTECTED] 12/19/2005 1:31 PM >>>
Yours is probably like mine - a non-zero deferred work count in the
VMDBK, but no deferred work to be found anywhere in the system. You
can't tell what did not decrement the count when it finished. I have had
a few of these this past year in a 4.4.0 system.

 -----Original Message-----
From:   VM/ESA and z/VM Discussions [mailto:[EMAIL PROTECTED] 
On Behalf Of Tom Duerbusch
Sent:   Monday, December 19, 2005 11:23 AM
To:     [email protected] 
Subject:        Re: forcing a user

Great plug.

Doesn't show anything as there isn't anything to show<G>.

TRACK 1 USER : NEWESA4        ****   DEV   ****   z64 V4.2   13:22:31 

12/19/05
================================================================================
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
                                                                      

        
(Chan= 0   Ctls= 0    Devs= 0   
)==============================================
CR=Upd CLEAR=Subset  ==>                                              

        

Looks like an IPL tonight.


Tom Duerbusch
THD Consulting

>>> [EMAIL PROTECTED] 12/19/2005 12:15 PM >>>
> My guess is, yes.
> But how can you tell?
>
> Tom Duerbusch

<ShamelessPlug=Perpetual>
Have you given TRACK a try?  TRACK will show you pending I/O's on any
of
the virtual machine's devices, along with many other nifty things.

You can download it here: http://vm.marist.edu/track/ 

_______________________________________
James Vincent
Systems Engineering Consultant
Nationwide Services Co., Technology Infrastructure Engineering
Mainframe, z/VM and z/Linux Support
One Nationwide Plaza  3-25-02
Columbus OH 43215-2220   U.S.A
Voice: (614) 249-5547    Fax: (614) 677-7681
mailto:[EMAIL PROTECTED]

Reply via email to