Is LISTSERV staying in MUTEX state or going in and out? In and out is 
normal VMS processing, but rarely does anything stay in MUTEX state 
for more than a blip. 

If it's staying in MUTEX state there are a few possibilities:

1. Ripple effect from some resource that LISTSERV is relying on being 
unavailable, e.g. a disk is hung or some controller is hung. Are you 
seeing any other problems with any disks?

2. Exceeding process quotas. Some resources will cause a MUTEX state 
when quotas get very low or exceeded. So check the process quotas.

3. Some bug in LISTSERV where it is locking but not releasing some 
resource. We run LISTSERV on VMS here and have never encountered 
this. VMS tends to stay up for months at a time (after all, that's 
one of the things that VMS is good for), and we have a lot of traffic 
going through it. So it would have to be an unusual situation or we 
would have noticed it, and from the logs you sent to support, it 
doesn't seem there was anything unusual going on. (You might want to 
go to TABLELESS RUNMODE though, as suggested by the logs).

Regardless of which of these it is, a reboot might break the lock and 
return LISTSERV to service (and note: I DO cringe as I write this 
because the idea of rebooting VMS to fix a problem is usually silly 
but a process inexplicably stuck in MUTEX is one of those rare cases 
where it may actually help). But of course if the problem is not 
fixed it could recur. You need to check the first two, and I'll pass 
this on to development to see if it might be the third, or if there 
are other things it could be.

Francoise 


On 27 Oct 2006 at 10:43, Alan Winston - SSRL Central C wrote:

> VMS 7.3-2
> LISTSERV 14.5
> 
> Current situation:
> 
> 2180023F LISTSERV        MUTEX    5     3013   0 00:00:01.04      1157    838
> 2180025D LSV_SMTP_1      HIB      3     1189   0 00:00:00.36       156    190 
> 
> 
> The LISTSERV process is suspended.  When PMDF tries to deliver to the LISTSERV
> channel, that process hangs.  (And clogged my MAIL$BATCH queue so no mail was
> getting through at all; I've temporarily increased the number of jobs the queue
> may process simultaneously, so that part of my crisis is resolved.)
> No messages are getting delivered; LCMD hangs.
> 
> 
> Background:
> 
> Last night I did an 
> 
> $ LCMD QUIET SET * NOREVIEW FOR *
> 
> which gave me output for a while but not for all lists and stopped.  After that
> LCMD hung and LISTSERV proved to be in a SUSPENDED state.  [Not that I've
> looked at it when things are working properly - I've only started using
> LISTSERV seriously recently].  It was late at night so I was able to reboot the
> server and things seemed to be okay again, but now this is all back.
> 
> What do I do?
> 
> -- Alan
> 
> 
> -- 
> ===============================================================================
>  Alan Winston --- [log in to unmask]
>  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056
>  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025
> ===============================================================================

-- 
Francoise Becker

There is only one LISTSERV(R) -- the product that launched the 
email list communication industry in 1986. To discover the 
story behind LISTSERV, visit:
http://www.lsoft.com/corporate/20anniversary.asp