On Wed, Jun 04, 2003 at 11:30:00AM +1000, Del wrote:
>
> Hi,
>
> Does anyone have one of these boards working with a recent 2.4
> kernel? Preferably out of the box on Red Hat ...
>
> I can't find any doco on the Intel site w.r.t. this controller
> and 2.4 and I can't see from the kernel sources whether the intel
> supplied driver has been folded into the kernel source yet or
> not.
Synchronicity.
I am currently running up a system with the earlier, uglier
U-21 controller (no longer supported) and have been having
mixed success.
The good news:
- Yup, the family work under linux, though I had to jump
through hoops to get it running.
The RH9 installer kernel does not appear to recognise the
device properly. It loads the i2o_block module, but does
not find a drive. If you can get RH9 installed (I put a
base install on an IDE drive) and run up a kernel with
i2o_pci, i2o_core and i2o_block support, your controller
should work.
- There is a management interface available. The interface
for the U-21 is a web application that required java, was
released for RH 6.2 and is no longer maintained, but it
pretty much works.
- Once the controller is up and running it seems pretty
solid.
The bad news:
- In the kernel source, the controllers are reported to hang
under "certain load patterns". I don't know what they are,
but I _really_ hope it is far rarer than I can imagine.
- For me, interrogating certain parts of the /proc/i2o/
tree caused a kernel panic. This is bad. In
/proc/i20/0x000*/ there are elements that represent the
state of the controller. I got bored seeing which elements
would cause a panic. All the ones I tried did. The odd
thing is that the web management interface seems to access
this same info without the grief. My solution: build the
kernel without i2o_proc
- No logging activity of disk events. I've marked disks as
failed (which triggers activatio of the standby disk I am
using) and I get nil, nix, nada reporting of the event in
the system logs.
- The user-land application that reports on the drive status
sort of works (for me, this is called irview), but some
switches and disk conditions (irview -a with a degraded
array) cause a segfault
HTH and if someone could tell me that I am an oaf and have
missed an obvious solution here.
Seb
>
> Next question would be about an adaptec 2120S but the drivers
> for that appear to be in the kernel.
>
> Thanx,
>
> --
> Del
>
> --
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
