Re: kernel inclusion planned?

On Wed, 28 Jun 2006 18:37:01 +0200 (MET DST)
"Peter T. Breuer" <[email protected]> wrote:

> "Also sprach Stephan von Krawczynski:"
> > I am one of those who think kernel inclusion would be a desirable thing. My
> > major reasons are:
> > 1) enbd needs more "eyes". 
> 
> I agree.
> 
> > Every now and then I set up an enbd-testbed to find out how things are going.
> > The last one was some weeks ago and it was  - again - not quite successful.
> > Setup worked as usual quite ok, but the boxes crashed when I stressed the
> > enbd-raid 1 (with fr)(*). This was not the first time that happened to me,
> > effectively all former tests ended more or less that way. Up to now I have
> 
> Then you want to publish the test so that the problem can be
> reproduced for investigation.  You also want to check out which
> versions of enbd in combination with which version of the kernel
> reproduce the problem, and most importantly, which do not.

The problem at my side is: I do not have spare boxes available for such tests.
Every now and then I get my hands on some Xeon or Opteron boxes with 4-8 GB
ram and raid1 or raid5 disks.
But they do not stay particularly long so I cannot perform real testings. What
I do is install the stuff based on 2.4 smp kernel (last available from
kernel.org at the time), make a enbd-fr-raid1 and simply copy around 10 GByte
of data from a local DVD to the enbd-fr-raid1 
(cp -av /media/dvd/. /my-enbd-drive/ ). None of the installations so far
completed this single task from a single user. This is not a very heavy or
complicated test, it's more or less a can-happen-every-day-in-my-office kind of
a job. All my setups are done with a cross-over cable for exclusive enbd
usage, most of the time with e1000.
My last try ended up with both servers frozen and plain dead.
That is unfortunately all I can tell about the story. If you copy only 1 or 2
GBytes around it looks pretty ok, at least does not freeze.
You may be right with your guess it is something with buffers or with smp
racing, both could have smashed the setup.

-- 
Regards,
Stephan