Pierre Fortin wrote:
> On Tue, 7 Jan 2003 23:33:51 -0600 "David Rankin" > <[EMAIL PROTECTED]> wrote: > > > > > I have been struggling with a weird message lately that hasn't appeared > > over the past 2 years until the past few weeks. I don't have any clue as > > to what this is actually telling me, but I would suspect that it may be > > the initial gasps of a dying LNE100TX network card (tulip driver). It > > may only show up once or twice every week. However, with my very limited > > kernel error message understanding, I'm at a loss. So experts -- Chime > > in! > > > > Setup: > > > > LM 7.2 Odyssey (2.2.19 kernel) > > Abit Kt7 mb > > AMD T-bird 800 > > Linksys LNE100TX > > (No hardware changes since Jan, 2001) > > Does the 'no changes' comment also apply to all the connected systems in > the office? Is anyone using a new app on the non-Linux boxes? Was there > a storm just before this problem started? (One of my LinkSys ports was > zapped by lightning induction -- the connected box was not affected) > Well, this is interesting because there was one hell of a lightning storm several days before this problem appeared. The only reason I don't think this correlates is because the errors didn't appear until a few days after. What about moisture around the CAT5 cables. Our office is an older structure that sits on pier and beam with the CAT5 run underneath. After the storm, it is possible the CAT5 was in a very moist environment. The CAT5 insulation should be good on all the cables. Has anyone heard of moisture around cables causing this type of problem? > > > Network: > > > > server -> 8 port C-net hub [7 out to LAN 1 to] -> 4 port Linksys > > cable/dsl router [1 LAN connection] -> (the rest of the world/internet) > > > > > > No problems for over 580 days. Since 12/26/02 I have received the > > following message with varying frequency. It really hung on the 12/26, > > but after that I only see the message once maybe twice a day or week? > > > > Jan 3 11:28:22 Nemesis kernel: eth0: Tx hung, 1191883 vs. 1191882. > > Jan 3 11:28:22 Nemesis kernel: eth0: PNIC2 transmit timed out, status > > e4000000, CSR6/7 01000000 / effffbff > > CSR12 40a1d0cc, resetting... > > > > OK, this is Jan 7 and I haven't had the message since the 3rd, so: > > > > What in the heck does this error mean? Google says that it may be a > > tulip driver error, but that is no explaination as to why I would go > > 500+ days without the error and then have it magically (or a a result of > > witchcraft) appear? Does it smell like a hardware problem? When it > > occurs, I have noticed that my C-net 8 port hub acts flakey. After I > > disconnect all the clients, reset the modem, reset the router, reset the > > hub -- everything seems OK. So what gives? (I generally do a > > ..../init.d/network restart just to be on the safe side) Which resets > > without any errors. > > > > My real question is what in the hell is an: eth0: Tx hung message mean? > > What does the eth0: PNIC2 message mean? Docos and info is sparse on this > > message. So, has anyone else got a clue as to what I'm dealing with? > > This is gonna require the 7.2 sources (which I no longer have) and a read > of the code -- that was the only source of docs that helped when I had > eepro10 problems on 7.2... > > > As always, any wisdom would be greatly appreciated. > > Running ethereal won't show such a low level problem; but it might provide > a clue if a bug is triggered by something on the LAN condisdering you are > using a hub... > > HTH, > Pierre > > > -- > > David C. Rankin, J.D., P.E. > > Rankin * Bertin, PLLC > > 510 Ochiltree Street > > Nacogdoches, Texas 75961 > > (936) 715-9333 > > > > > > > > ------------------------------------------------------------------------ > Want to buy your Pack or Services from MandrakeSoft? > Go to http://www.mandrakestore.com -- David C. Rankin, J.D., P.E. RANKIN * BERTIN, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 (936) 715-9333 (936) 715-9339 fax
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
