On Wed, 2006-01-11 at 17:28 +0000, James Pattinson wrote:
> Ming Zhang wrote:
>
> >it is one e1000 or dual e1000 port? if 2, have u connect both to
> >network?
> >
> >
> >
> It's just one port.
ok. e1000 has some issue that can not support 2 ports easily in one
network.
if this is not too troublesome, could u change to another NIC and see if
ok?
i saw a lot of retrans, checksum error, and spanning tree packets and
time delay between them are pretty long. so iscsi layer might begin to
do task management or relogin or many ugly stuff. :P
>
> >>Do you think it's the e1000 driver at fault?
> >>
> >>
> >not sure but suspect.
> >
> >u old run-well box is the same physical one, u just update the kernel
> >and iet version then met this issue? or different box?
> >
> >ming
> >
> >
> I have two iSCSI target machines.
>
> 1. calculon (192.168.100.3, in the logs) was working OK at 2.6.12-gentoo
> and 0.4.12 iet, but now is broken with 2.6.14-gentoo and SVN iet
> 2. crocodile - amd64 machine running 64 bit kernel, worked fine at
> 2.6.12 and 2.6.14-r2 with 0.4.12 each time
so this one u did not tried the svn code?
>
> So in answer to your question, yes the box worked well before I updated
> the kernel and IET version
so they are the same box.
>
> Cheers
> James
>
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
|
