Could this have anything to do with the QFT fron the VA71xx diak array limits?

We are running a VA7110 and have the Queue Full Threshold set at 200. We have
13 LUNS configured and from what I understand that the QFT table should be
set to the default of 750 or a minimum of 8*LUNS.

So the 200 setting is well withing the limits issue.
> Joseph Dolliver writes:
>
> > We are running a new N-Class system and are on 7.5 pp1 with reactive patches
> > for a new VA7110. While in glance we see the following frequently during the
> > day and see spikes to 100% CPU usage. There are usually pins 4,5,6,7,8 all
> > running this program and are in the "A" queue. Can you tell me what the
> > system is doing
> >
> > Thanks in advance
> >
> >
> > Process Family for Pin 7 is:
> >
> > Parent is Pin 1 executing program PROGEN.PUB.SYS
> >
> > Child Processes are:
> >
> > No Children for this Process
>
> Progen does not do much of anything until you perform a shutdown at which point
> it
> takes care of signaling processes to terminate and waiting for them to do that.
> The
> rest of the time it is just the "parent" attachment point for system processes.
>
> Pin's 4,5,6,7 & 8 are "port facility processes" or PFP's. These processes handle
> routing of messages to and from ports that for one reason or another cannot send
> a message or receive one. For example, a port's "server" cannot run on the ICS
> (interrupt control stack) but something needs to send it a message from there.
> Certain
> ports also have thresholds of the number of messages they can handle at one
> stretch
> anything over that threshold (which varies port to port) will be routed to its
> assigned
> PFP. None of this is abnormal or a sign of problems, it happens all the time.
>
> While I don't know what might be causing the 100% spikes it is possible that
> seeing
> PFP's running during them is a byproduct rather than the cause.
>
> hth,
>
> Bill
> hp/vCSY
> (to reply: change the 'dot' to a '.' and remove 'nospam')
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *