Re: (PM) Rebooting PM3
Florian Lohoff (flo@mini.gt.owl.de)
Tue, 23 Jun 1998 10:16:41 +0200
On Mon, Jun 22, 1998 at 02:39:46PM +0200, Frank van Berkel wrote:
> > At 12:39 PM 12/23/97 +0100, you wrote:
> > >On Sun, 21 Dec 1997, Jim Wells wrote:
> > >
> > >> LE ticket #48378 was opened over a month ago.
> > >> We are still getting phantom reboots of all 3 of our PM3s.
> > >> They seem to be at random, but each one reboots every 36 hours or so.
> > >>
> > >> The latest theory from LE is routing loops, but I have no way to know if
> > >> these are occurring or how to prevent them.
> > >>
> > >> A console log shows why the reboots are occurring as follows:
> > >>
> > >> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> n_newbuf: No available netbufs
> > >> Assertion failed: nbuf_p->data_ptr, file mdp_os.c, line 1268
> > >>
> > >> Livingston Enterprises, Inc. Boot Prom Rev P
> > >>
> > >> Testing Low Memory....
> > >>
> > >> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > >> So there is code in the PM3 that is deciding to do a reboot,
> > >> apparently because of a lack of buffers. What is not known is what
> > >> condition leads to this result.
> > >
> > >Same problem had the Max4000 recently and the USR TC Netserver
> > >do have the same problem currently. All have random reboots. The Maxes
> > >had this because of non-freed per call buffers ... Does Livingston
> > >look at the # of calls the rebooting boxes have ?
> > >
> > >Flo
> > >--
> > >Florian Lohoff mailto:flo@mini.gt.owl.de Phone:+49-5241-470566
> > >Privates Internet Ostwestfalen-Lippe, Guetersloh - http://www.gt.owl.de
The NetServers are meanwhile fixed, it was a buffer problem in the
X.75 code as far as i remember ... BTW: The first reboot bug
in the PM3 has also been resolved and another one appeared
in 3.8b15, running 3.8b13 now ...
BTW: Wouldnt it be good to have the ComOS developers write
a diary what they done and fixed ? So we all could have an imagination
on what is going on ... Just a daily "one sentence per day or something",
like everyone of us has some pages to visit with the first
coffee in the morning :). I would like to know
where R&D is what is fixed or added, or what the just working on.
Like Alan Cox writes a diary in the Web ... (For the one
not knowing Alan Cox -> Amazing linux kernel hacker - now at redhat)
http://roadrunner.swansea.uk.linux.org/alan.shtml
Flo
--
Florian Lohoff flo@mini.gt.owl.de +49-5241-470566
-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-users' in the body of the message.
Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>