Re: HEADS UP: panic behaviour changed

[Joerg Sonnenberger <joerg%britannica.bec.de@localhost>]

On Sun, Feb 01, 2009 at 01:19:19AM +0700, Robert Elz wrote:
> But if your issue is that the default should be for ddb.onpanic == 1
> then the right way to achieve that is to put
> 
>       options DDB_ONPANIC=1
> 
> into conf/GENERIC (and other files as appropriate), rather than by making
> the option automatically set itself if the config file doesn't sent it.

Agreed.

> So, which is the default as distributed really applies (or should apply)
> to those regular users who don't build their own kernels, and just keep
> running GENERIC (personally, I think that should amount to no-one at all,
> so I don't think this is a very important issue.)

Yes. My problem now is that most users won't know what to do with
coredumps. They might not even have enough swap or space in /var/crash
for them to be created, even if the kernel successfully dumped core.
Effectively this change means that a PR will now contain "Something
failed and the machine took 15min to reboot".

If the problem Andrew tries to fix is the cryptic ddb prompt, I would
suggest to change the panic message to include something like:
        Please report this issue via send-pr(1). Include at least the
        output of "backtrace". "reboot" will try to create a coredump,
        if the swap partition is large enough.

Joerg