🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Thread View: comp.sys.sgi.admin
3 messages
3 total messages Started by Scott Cheadle Wed, 29 Dec 1999 00:00
SGI Misinterpretting the FDDI Standard
#246
Author: Scott Cheadle
Date: Wed, 29 Dec 1999 00:00
18 lines
1861 bytes
Hi,

I have had this ongoing problem with SGI now for over a year, and the
problem is still not fixed. I was wondering if anyone else uses SGI
Challenge L's, or O2's to connect through an FDDI ring to a DIGITAL
Concentrator. Basically, the problem is, that IRIX sends a Break Signal
to the redundant A port once every 50 seconds. This in turn triggers the
SNMP collector to send a trap to HP Openview, triggering an alarm. We
have DU, HP-UX, AIX, and IRIX, and IRIX is the only offending system,
however all SGI seem to be interested in is in blaming Digital's
Concentrators, and not in solving the problem. It is apparent that they
realise this is their problem, but they do not seem particuarly keen to
investigate, (Indeed all they have done, is to give us a Test O2
machine, and ask us to do the testing and debugging for them.). Does
anyone else use Digital Concentrators with SGI IRIX, and if so, have you
also experienced problems?.

Scott Cheadle.
Re: SGI Misinterpretting the FDDI Standard
#247
Author: Mike O'Connor
Date: Wed, 29 Dec 1999 00:00
31 lines
2407 bytes
Adding the FDDI newsgroup to the Newsgroups: line...

In article <386A165D.7ABA9E9@NOSPAM.hotmail.com>,
Scott Cheadle  <schead@NOSPAM.hotmail.com> wrote:
:I have had this ongoing problem with SGI now for over a year, and the
:problem is still not fixed. I was wondering if anyone else uses SGI
:Challenge L's, or O2's to connect through an FDDI ring to a DIGITAL
:Concentrator. Basically, the problem is, that IRIX sends a Break Signal
:to the redundant A port once every 50 seconds. This in turn triggers the
:SNMP collector to send a trap to HP Openview, triggering an alarm. We
:have DU, HP-UX, AIX, and IRIX, and IRIX is the only offending system,
:however all SGI seem to be interested in is in blaming Digital's
:Concentrators, and not in solving the problem. It is apparent that they
:realise this is their problem, but they do not seem particuarly keen to
:investigate, (Indeed all they have done, is to give us a Test O2
:machine, and ask us to do the testing and debugging for them.). Does
:anyone else use Digital Concentrators with SGI IRIX, and if so, have you
:also experienced problems?.

If you're seeing what I think you're seeing, I believe that FDDI is
_supposed_ to do LCT every 50 seconds to check for the DAS link state.
Check out both the URLs below for a good writeup of the situation:

http://www.iol.unh.edu/training/fddi/htmls/lct.html 
http://www.deja.com/=dnc/getdoc.xp?AN=358717781 

Shouldn't OpenView take this into account?

--
 Michael J. O'Connor | WWW: http://dojo.mi.org/~mjo/ | Email: mjo@dojo.mi.org
 InterNIC WHOIS: MJO | (has my PGP & Geek Code info) | Phone: +1 248-848-4481
Re: SGI Misinterpretting the FDDI Standard
#248
Author: Scott Cheadle
Date: Thu, 30 Dec 1999 00:00
53 lines
3919 bytes
It is supposed to check the A port, by going to connecting state, but not send a
Break Signal. I tried sending smtd SIGUSR1 3 Times to increase the level of
logging to Info, and it did send a WH_sig every 50 seconds, but this did not
neccassarily correspond with the break, which did not get sent to syslog. I also
found that upon  increasing smt_max_connects, the frequency between breaks
increased, although this is only when providing a value of 9999. After talking to
SGI, for quite a while, they finally said that he has seen a possible problem
with the source code, but still would not admit that SGI's interpretation is
incorrect, since we have only tested this with Digital Concentrators. However,
the fact is that only IRIX has this 'Feature', and we have set openview to give a
Homer Simpson 'DOH' upon error, so given that we have many SGI machines, each one
giving a 'DOH' every 50 seconds, it gets rather irratating :(. I would like to
know of anyone else using SGI Irix with Digital FDDI Concentrators, since they
claim this is the first time they have seen the problem.

Regards,

Scott Cheadle.


Mike O'Connor wrote:

> Adding the FDDI newsgroup to the Newsgroups: line...
>
> In article <386A165D.7ABA9E9@NOSPAM.hotmail.com>,
> Scott Cheadle  <schead@NOSPAM.hotmail.com> wrote:
> :I have had this ongoing problem with SGI now for over a year, and the
> :problem is still not fixed. I was wondering if anyone else uses SGI
> :Challenge L's, or O2's to connect through an FDDI ring to a DIGITAL
> :Concentrator. Basically, the problem is, that IRIX sends a Break Signal
> :to the redundant A port once every 50 seconds. This in turn triggers the
> :SNMP collector to send a trap to HP Openview, triggering an alarm. We
> :have DU, HP-UX, AIX, and IRIX, and IRIX is the only offending system,
> :however all SGI seem to be interested in is in blaming Digital's
> :Concentrators, and not in solving the problem. It is apparent that they
> :realise this is their problem, but they do not seem particuarly keen to
> :investigate, (Indeed all they have done, is to give us a Test O2
> :machine, and ask us to do the testing and debugging for them.). Does
> :anyone else use Digital Concentrators with SGI IRIX, and if so, have you
> :also experienced problems?.
>
> If you're seeing what I think you're seeing, I believe that FDDI is
> _supposed_ to do LCT every 50 seconds to check for the DAS link state.
> Check out both the URLs below for a good writeup of the situation:
>
> http://www.iol.unh.edu/training/fddi/htmls/lct.html
> http://www.deja.com/=dnc/getdoc.xp?AN=358717781
>
> Shouldn't OpenView take this into account?
>
> --
>  Michael J. O'Connor | WWW: http://dojo.mi.org/~mjo/ | Email: mjo@dojo.mi.org
>  InterNIC WHOIS: MJO | (has my PGP & Geek Code info) | Phone: +1 248-848-4481
Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads