Re: install on nec mobilepro 780

Hi Aaron!

I'm happy that I'm not the only one trying to get this going!

At this point I have a working install on my 780, though I'm still struggling with some configuration issues.

Where in the boot process are you getting hung up?

Once you got your partitions/labels set up and your sets extracted, did you remember to switch the kernel in your hpc boot utility from the install kernel (netbsd.gz) to the boot kernel (netbsd-GENERIC.gz)? That was one of the steps I didn't catch on to the first time through the install docs as it's not explicitly mentioned.

On Thu, Apr 15, 2021 at 6:56 PM Aaron Peters <[email protected]> wrote:
> 
> Hi everyone, I've been lurking here for some time with the intent of asking
> about this very issue. I'm on a MobilePro 790, essentially the same as the
> 780 except for a bit of persistent RAM (useful for storing the bootloader).
> 
> I'm happy to help troubleshoot/test, I'm in the same situation as Michael.
> sysinst worked well until the step where sets are extracted, where that
> process always errors out. I took the same path of dropping into a shell
> and extracting them manually, I'm stuck trying to get the new filesystem to
> boot. I suspect this is as simple as not having made the right
> slice/partition active, but since my last attempt was using 9.0 I'm happy
> to re-do the whole thing for the newer version.
> 
> Aaron
> 
> 
> On Thu, Apr 15, 2021, 7:20 AM Martin Husemann <[email protected]> wrote:
> 
>> On Wed, Apr 14, 2021 at 04:18:24PM -0400, Michael Grayson wrote:
>>> Update on this. After quite a bit of tinkering I was able to
>>> bootstrap the system by using sysinst to build the filesystem and then
>>> manually installing the sets from the shell. I believe sysinst is
>>> broken for this architecture; specifically, it is dumping core from the
>>> progress utility. Not sure who to talk to about that.
>> I can help with sysinst specific issues, but I don't have that hardware
>> and no idea how to easily test. What install medium is used, and
>> can we run it in gxemul?
>> If basic tools like progress do not work, then something more serious
>> is going on and we need to fix that first.
>> Martin