SUMMARY: More on Token Ring Errors

Fred Homan (FHoman@mail.dot.state.nc.us)
Thu, 24 Apr 1997 08:20:14 -0400

Thanks to the responses I received and a ring crash, I believe that the
problems were being caused by my ring and not the machine. The ring
crashed just two days ago, because of a hub problem (ironically, the hub
which was only being used by my and a colleague's Pcs). Point on with
an explanation is the following message from Toby Potts
<tobypotts@pmsc.com>:

"These listings in your syslog are not problems with your machine, they
more than likely indicate issues with the ring to which you are connected.
We have two Enterprise 5000s at my site, each with token ring and
ethernet network connections. I have opened a call with SUN
on this very topic with the exact messages you received. They assured
me that it was not a problem with the SUN machine. The explanation for
the messages was that the SUN token ring drivers are more sensitive to
events occurring on the ring itself, and that when these messages
appear it is and indication of other ring problems. If you do ever get
beacon problems, you may want to check the configuration of your token
ring hub(s). They are probably the source of the errors."

Thanks also to:
Glenn Satchell - Uniq Professional Services
<Glenn.Satchell@Uniq.com.au>
Charles Homan [ext 422] <choman@applix.com> [Special thanks for
brotherly support for being ashamed of my posting a TR question! ;-)]

And, finally, the original message from me:

Dear colleagues,

I have a similar question to the one asked by Shirley Chee Mei Kin about
token ring error messages (and hope to have a similarly easy solution). I
don't get beacons, though, I get the following in my syslog:

Apr 20 20:11:12 unix: NOTICE: tr0: Soft Error
Apr 20 20:11:12 unix: NOTICE: tr0: Ring Recovery
Apr 20 20:11:22 unix: NOTICE: tr0: Ring Recovery
Apr 20 20:11:22 unix: NOTICE: tr0: Soft Error
Apr 21 06:48:34 unix: NOTICE: tr0: Ring Recovery

I've never noticed any connectivity problems (or heard any complaints),
but the machine is underutilized as it is basically just a web server and
not often visited, so it may just be chance. I wouldn't worry except for
the last two entries in my above example which indicate a possible 10-1/2
hour loss of connectivity?

Machine itself is a SpacStation 20 running Solaris 2.5.1 with all the
Sun-recommended patches applied. Summary will be posted, o' course.

Fred Op
Part-time Solaris Admin

-----------------------------------------------------------------------------------------
| Fred Homan, | "The dead have risen from the |
| Metro Information Services | grave and and they're voting |
| On Contract With: | Republican!" |
| North Carolina Dept of Trans., | -- Bart Simpson, |
| MIS, Client Services | "The Simpsons" |
-----------------------------------------------------------------------------------------