[Info-vax] FREWSLX Alpha 7.3-2

[DJ]

On Apr 25, 9:21 pm, s... at obanion.us wrote:
> On Wednesday, April 25, 2012 6:52:57 PM UTC-7, Richard wrote:
> > Folks,
>
> > Hoping someone out there still looks at this group :-)
> > Any thoughts on whether and what tuning we need to avoid a repeat of
> > this crash?
>
> Google found a thread for Alpha 8.2, concerning a similar crash when the SYSTEM account was running Monitor:http://unix.derkeiler.com/Newsgroups/comp.os.vms/2006-08/msg02017.html
>
> And Volker Halle suggesting:
>
> "This code in the pagefault handler is trying to free a working set list
> entry, to re-use it to bring in the page about to be faulted into
> memory. It cannot find any WSLE (working set list entry), which it can
> free up - after trying hard. Most of the WSLEs are probably somehow
> locked in the workingset.
>
> Your system may be a little bit low on memory. You may want to try to
> increase the working set quota for user SYSTEM, who was running
> MONITOR.
>
> You can verify my analysis with
>
> SDA> SHOW PROC/WORKING
>
> and see, if you can find a WSL entry, which could have been freed ;-)"
>
> When I last worked on Alpha 4100s, we usually ran it with 2 to 4 GB of memory, so the 1 GB you have might be a little low...
>
> Sean

Please read url: http://www.s-and-b.su/syshlp/vaxsys03_072.release_notes

Problems addressed in VAXSYS02_072 kit

If a process working set is reduced, such that the working  set
         only  contains  locked  pages, and the process then
pagefaults,
         the system could bugcheck in  MMG$PAGEFAULT  with  a
"FREWSLX,
         Free  working  set  list  index,  resource  wait"  error.
The
         bugcheck occurs because no free working set list entries can
be
         found.

         Images Affected:

          -  [SYS$LDR]PAGE_MANAGEMENT.EXE

I sounds like the WSQuota is not large enough. I would double the size
of the WSQuota for that process and see if prevents the problem.

Just a guess.

Regards,
Daryl Jones