redundant NICs

Hans Fugal hans at fugal.net
Sat Jul 23 08:41:25 MDT 2005


On Fri, 22 Jul 2005 at 13:22 -0600, Lars Rasmussen wrote:
> On 7/22/05, Hans Fugal <hans at fugal.net> wrote:
> > On Thu, 21 Jul 2005 at 22:01 -0600, Lars Rasmussen wrote:
> > > I don't know how to do this with one server and 2 NICs.  I'd be more
> > > worried about fan, power supply, or hard drive failure(all moving
> > > parts)than NIC failure.
> > 
> > All redundant already, and I've seen two NICs fail in important boxes in
> > the past year, tying for first place with power supplies.
> 
> Are you inferring that NICs are as prone to failure as power supplies?

Facetiously. I know NICs are (or at least should be) more reliable than
power supplies. But I've seen two go out in the last year so I'd say
that based on that experience it would be worth the redundancy on a
server where downtime is critical. One wasn't actually a failed NIC,
just a NIC that went to lunch - didn't work until reboot. That was on
this very server and it brought the server offline in spite of the
supposedly redundant NIC.

>  There must be some extenuating circumstances as to what you've
> experienced in the past year.  Saying NICs are as prone to failure as
> power supplies would be hard to back up by sampling MTBF data, even
> with power supply companies grossly inflating their numbers.

*shrug* I only know what I saw. I'm sure power supplies have a lower
mean time to failure when you look at the statistics.

> One could limit sample data to their experience... In your case 100%
> of NICs have failed out of the 2 sampled, in less than 10 years!
> (guessing)

Well I do have more data than that. Each server and IT workstation (the
machines considered in my terribly unscientific flippant "study") has
two power supplies and one or two NICs. Two power supplies failed, two
NICs failed. I present the data and let you reach your own conclusions.
:-)

I don't really care how power supplies compare to NICs, the point is
that NICs fail. Even more important, switches fail, switches lose power,
cables between switches and servers and other switches fail, and dumb
system administrators accidentally unplug cables. I'm quite sure all
those problems combined are more frequent than failing power supplies.


-- 
 .O.  Hans Fugal            | De gustibus non disputandum est.
 ..O  http://hans.fugal.net | Debian, vim, mutt, ruby, text, gpg
 OOO                        | WindowMaker, gaim, UTF-8, RISC, JS Bach
---------------------------------------------------------------------
GnuPG Fingerprint: 6940 87C5 6610 567F 1E95  CB5E FC98 E8CD E0AA D460
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://plug.org/pipermail/plug/attachments/20050723/5b105028/attachment.bin 


More information about the PLUG mailing list