I installed 3.0 from the working qube by unpacking, not from restore cd. 
I have random crashes when http server is accessed by more than 1 client 
at the same time. Kernel is custom compiled to remove the things that I 
don't use.

I am thinking of going back to 1.6.2 which was a very stable version at 
least for me. 2.1 was even less stable than 3.0 and I do not see any 
benefits as far as performance goes. I will probably wait until 3.0 
matures to move my qube to it again.

Regards,
Alex


Claudio Leiva S wrote:
> Hi:
> 
> Any issue installing 3.0 on a Qube 2, I have the process with the CD 
> done but when it restart it get stuck on "Starting Up".
> 
> Any ideas, the specs of my box are.
> 
> 256 mb
> 120 Gb hard drive
> 
> Unfortunatelly I dont have a serial connection with the box..
> 
> 
> 
> Claudio Leiva S.
> Las Vegas, Nevada.
> USA
> http://cleiva.no-ip.org
> (Powered by NetBSD for Cobalt)
> 
> 
> ----- Original Message ----- From: "Daniel Ouellet" <daniel@presscom.net>
> To: <port-cobalt@netbsd.org>
> Sent: Friday, January 06, 2006 12:06 AM
> Subject: Compile PF in kernel on Cobalt is no go on 3.0
> 
> 
>> I tried a few times to enable PF and so far, I can't even compile the 
>> kernel with it enable.
>>
>> As far as I can see, only these needs to be commented out:
>>
>> pseudo-device   pf                      # PF packet filter
>> pseudo-device   pflog                   # PF log if
>>
>> as by default it is not in the GENERIC kernel.
>>
>> I have no problem making any changes to the kernel and get it to 
>> compile, but this two line to turn it up as far as I can see, result 
>> in no go.
>>
>> I wish I had the error to put here, but as it takes a long time to 
>> compile, I did it in the background and sing off, so, I didn't see the 
>> errors the last time, but no matter what variation I do, still no go.
>>
>> Anyone able to do so yet? Am I forgetting to do something else as well.
>>
>> Also, in default there isn't a PF device in the /dev, so I use makedev 
>> to create it, not sure that it is the proper way for this one however.
>>
>> Any clue to offer?
>>
> 
> 
> 
> 
>