[ previous ] [ next ] [ threads ]
 From:  "Steve Thomas" <sthomas at consultant dot com>
 To:  "Lonnie Abelbeck" <lists at lonnie dot abelbeck dot com>, "m0n0wall List" <m0n0wall at lists dot m0n0 dot ch>
 Subject:  Re: [m0n0wall] Hardware suggestion
 Date:  Thu, 08 Feb 2007 15:58:27 -0500
Good thought, I don't really know. AFAIK, turning on polling generally
improves performanceover having it turned off. I'm not sure it could
offset anythroughput limitation imposed by the hardware though. I think
if the bug is fixed now, the 4801 may likely haveperformance very close
to the WRAP (as Manuel statedin the article). I think this started as a
comparison of the 4801 vs. WRAP.If you already have the 4801, I don't
think it makes senseto scrap it for a WRAP, it surely works fine. There
is alarge installed base of them apparently running well. In the context
of picking and buying a new one, they arevery close to being the same
thing. I selected the WRAPbecause it costs less than the 4801, the IRQ
thing was aminor bonus. Hopefully both makers will have some newer,
fasterproducts soon at close to the same price points.

  ----- Original Message -----
  From: "Lonnie Abelbeck"
  To: "m0n0wall List"
  Subject: Re: [m0n0wall] Hardware suggestion
  Date: Thu, 8 Feb 2007 07:29:22 -0600

  Steve, et al.
  > "the issue that the 4801 shares interrupts
  > on the NICS. ... Separate interrupts are better.

  Does enabling 'polling' on the net4801's NIC's mitigate the shared
  IRQ issue you raise?


  On Jan 27, 2007, at 2:31 PM, Steve Thomas wrote:

  > Well a driver bug in combination with the issue that the 4801
  > interrupts
  > on the NICS. The shared interrupt part of that can't be fixed, and
  > can
  > re-appear sometimes. Separate interrupts are better.
  > My WRAP has been extremely reliable. No lockups ever and good
  > performance.
  > ----- Original Message -----
  > From: "Max Cristin"
  > To: m0n0wall at lists dot m0n0 dot ch
  > Subject: Re: [m0n0wall] Hardware suggestion
  > Date: Sat, 27 Jan 2007 12:56:14 -0500
  > Beside the price difference, doesn't the same article you quoted
  > say that the performance issue was caused by a bug in an old
  > version and that it is now solved? Can anybody with a 4801 running
  > Mono 1.22 or 1.3b2 confirm or deny that? Thanks.
  > Max
  > To unsubscribe, e-mail: m0n0wall dash unsubscribe at lists dot m0n0 dot ch
  > For additional commands, e-mail: m0n0wall dash help at lists dot m0n0 dot ch