From: Mike Broderick (broderic@MIT.EDU)
Date: Tue Jul 30 2002 - 12:45:33 EDT
Thanks to J Bacher, Werner Rost, and Colin Bull who all suggested network speed
mismatches. Synching the right speed fixed it. The details...
The system port and the switch port both claim to support 100MBPS, so I first
hard set the speed at the SRM (set ewb0_mode) to 100MBPS (even though the
network port lights on both ends already indicated it was using that speed
already). But that made things worse. (A large file copy took forever and
racked up many of the same errors.) I noticed that our other 3 DS10s (all
going to the same Cisco 5500 switch) all autosync'ed to 10MBPS. So I hardcoded
10MBPS ("twisted-pair") at the SRM and all errors went away as did all user
complaints.
_Mike
Mike Broderick wrote:
> Our users been experiencing dropped [database] connections to one of our
> Alpha servers (4.0g, no patches). We notice lots of errors (see
> below). We've tried swapping the cable and to the other ethernet port
> (DS10 w/ dual onboard ports). Does this sound like a h/w problem?
> Maybe at the switch port?
>
> _Mike
>
>
> tu1 Ethernet counters at Wed Jul 24 15:31:49 2002
>
> 61223 seconds since last zeroed
> 88449480 bytes received
> 185968374 bytes sent
> 597017 data blocks received
> 523581 data blocks sent
> 13156391 multicast bytes received
> 110440 multicast blocks received
> 12383 multicast bytes sent
> 209 multicast blocks sent
> 0 blocks sent, initially deferred
> 0 blocks sent, single collision
> 0 blocks sent, multiple collisions
> 0 send failures
> 0 collision detect check failure
> 572 receive failures, reasons include:
> Block check error
> Framing Error
> 0 unrecognized frame destination
> 0 data overruns
> 0 system buffer unavailable
> 0 user buffer unavailable
>
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:47 EDT