[9fans] still down

[9fans] still down

[geoff]

You may have noticed that the Bell Labs Plan 9
servers are still down.  We reloaded the raid array
controller's firmware, which it was demanding.
That appears to have been successful, and it seems
to have retained its configuration in NVRAM
(though why it lost its firmware is puzzling).
We're a little nervous about putting the disks back
into it because these devices have a nasty habit of
reformatting their disks when you change their configurations,
and it's hard to know if this one considers its
configuration to have changed.

Does anyone else have experience with these?  The model
naming is a little obscure but it seems to be Maxtronic
Arena Industrial II.  It takes eight IDE disks and presents
a single SCSI interface to the host.

Re: [9fans] still down

[Brantley Coile]

Hope you get if figured out.  This is why Coraid's
raid formats are public infomation.  We even have a
program calle srcat that will copy the blocks out
of a set of disks removed from our equipment.  I hate
to be held hostage by some raid controller.

If you need some new disks so you can dup the old ones
I can send you some.  Just let me know how big and if they
are PATA or SATA.

  bc

geoff@collyer.net wrote:
> You may have noticed that the Bell Labs Plan 9
> servers are still down.  We reloaded the raid array
> controller's firmware, which it was demanding.
> That appears to have been successful, and it seems
> to have retained its configuration in NVRAM
> (though why it lost its firmware is puzzling).
> We're a little nervous about putting the disks back
> into it because these devices have a nasty habit of
> reformatting their disks when you change their configurations,
> and it's hard to know if this one considers its
> configuration to have changed.
> 
> Does anyone else have experience with these?  The model
> naming is a little obscure but it seems to be Maxtronic
> Arena Industrial II.  It takes eight IDE disks and presents
> a single SCSI interface to the host.

Re: [9fans] still down

[geoff]

We're back up, except for a niggling mysterious problem with our
external httpd.

After making image copies of the disks in our raid array and
consulting with the maker, we crossed our fingers and it all just
worked.

Re: [9fans] still down

[geoff]

Our external httpd and wiki are back up.

wikifs still grows to hundreds of megabytes, which tends to run the
outside machine out of RAM and make it thrash until our scripts notice
and restart wikifs.  Until we solve this problem, the outside web
server and wiki may wedge from time to time.