forcedeth to forcedeth data corruption (CentOS 4.4, 2.6.9-42.0.3.EL and ELsmp)
00:0a.0 Ethernet controller: nVidia Corporation CK804 Ethernet Controller (rev a3)
CentOS 4.4, 2.6.9-42.0.3.EL, 2.6.9-42.0.3.ELsmp, 2.6.9-42.0.2.EL platform i386
It looks like for x86_64 this problem exists too, but I didn't test it hard
There are 2 issues:
messages in dmesg like this: eth0: too many iterations (6) in nv_nic_irq.
fixed by options forcedeth max_interrupt_work=20
During big data transfer (more than 1-1.5 Gb) I have some data corruption, like 300-500 random bytes per 1Gb, error usually group in 2-3 groups.
This problem exists between forcedeth<->forcedeth ethernet controllers only, I've tried forcedeth, tg3, sk98lin, e1000 and some 100Mbit cards like e100.
Please, give me some ideas.
I'll continue tests, try to install 2.6-latest kernel and will submit additional info.
Re: forcedeth to forcedeth data corruption (CentOS 4.4, 2.6.9-42.0.3.EL and ELsmp)
The 'too many iterations' messages are harmless, and you've already implimented the solution.
Which version of forcedeth are you using?
For the data corruption, how can that problem be reproduced & detected?
What kind of motherboard are you using?
How many systems are exhibiting this corruption?
Sorry for disturbing, it's not forcedeth, but nfsd/nfs-utils software bug.
I don't known why this bug appears on new WSs (new is: Athlon64 Dual Core 3800+, all other PC is slower than P4-3000), anyway now I'm sure that it was not hardware problem.
|All times are GMT -5. The time now is 06:37 AM.|
Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©1998 - 2014, nV News.