Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: (ASCEND) Pipeline 50 V 6.0.2 resetting frequently
On Wed, 6 May 1998, Jason Nealis wrote:
>
> Looks like the Pipeline is having some sort of memory leak, or it
> can't free up resources. You should file a trouble-ticket with
> TAC, Most likely they will refer you to another Patch release to
> see if it fixes your problem.
I thought something like that! I'm running 6.0.2 on a P50, and according
to my syslog, it resets hourly, but only if the connection isn't up. If
the connection is up, it'll run for days. In every other respect the box
works fine.
The real killer is that if it's close to dying, a telnet or connection
attempt kills it, then by the time it comes back, somebody's email attempt
timed out, and then they tell me the net doesn't work. (Of course it works
when I look at it after the P50 resets).
Get this: If I turn off NTP on my Linux box, it stays up *much* longer.
Now my filters say that passing NTP is OK, but it's not OK to call to do
NTP. The idea is that I only sync up the time when the connection is up
for some other reason. My guess is those unforwarded packets are filling
up memory until it finally runs out and dies. Sniffing reveals that the
P50 doesn't send ICMP back for packets that don't make the call filter if
the port is down.
Ascend technical support, which I finally finished a week's telephone tag
with (I'm only around during tech support hours on certain days), wants me
to send my unit in: the tech claimed that Fatal Error 2 means I've got
a memory defect.
I'm not convinced. Anybody know of a diagnostic that tests memory
definitively? I sure don't want to sink repair costs only to find out that
the hardware wasn't the problem.
I'm going to try 5.1Ap9 as suggested. My box didn't do this before.
I'll also report back.
Kevin
P.S. Here's my very similar fatal-history
FATAL ERROR: Index: 2 Load: b.p50 Revision: 6.0.2
Location: 0019b908 0012c048 001ccc2e 001ccdda 001cd2a0 001d0842
SYSTEM IS UP: Index: 100 Load: b.p50 Revision: 6.0.2
>
> Jason Nealis
> Director Internet Operations / Network Access
> Erols Internet
>
>
> On Wed, 6 May 1998, Florian Sauer wrote:
>
> > Jason,
> >
> > ----------
> > > Von: Jason Nealis <nealis@babylon.erols.com>
> > > An: Florian Sauer <florian.sauer@musaion.de>
> > > Cc: ascend-users@bungi.com
> > > Betreff: Re: (ASCEND) Pipeline 50 V 6.0.2 resetting frequently
> > > Datum: Mittwoch, 6. Mai 1998 03:09
> > >
> > >
> > > Drop into debug mode and type FAT this will show you the Fatal History
> > > log please post output.
> > >
> > > Jason Nealis
> >
> > Here is the complete output of the FATAL log:
> >
> > ****************************************************************************
> > **************
> > > fat
> > FATAL ERROR: Index: 2 Load: b.p50 Revision: 6.0.2
> > Location: 0019b908 0017f650 0010dcda 0010dfb8 0010edf2 001137ee
> >
> > SYSTEM IS UP: Index: 100 Load: b.p50 Revision: 6.0.2
> > ****************************************************************************
> > **************
> >
> > Any further ideas?
> >
> > Thanks for helping us!
> >
> > Regards,
> >
> >
> > Florian
> >
> > Florian Sauer M.A.
> > musaion media production
> > Lindpaintnerstr. 70
> > D-70195 Stuttgart
> > Tel +49-711-69795-40 Fax +49-711-69795-45
> >
> > eMail: florian.sauer@musaion.de
> > Web: http://www.musaion.de
> >
> >
>
> ++ Ascend Users Mailing List ++
> To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
> To get FAQ'd: <http://www.nealis.net/ascend/faq>
>
Kevin A. Mitchell, kam@kamit.com
Personal: http://www.kamit.com/
GIFConverter: gifconverter@kamit.com or kam@kagi.com
http://www.kamit.com/gifconverter/
++ Ascend Users Mailing List ++
To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd: <http://www.nealis.net/ascend/faq>
References: