[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#793786: initramfs-tools: fix the broken netconsole feature in initramfs-tools



On Wed, 2015-07-29 at 18:50 +0900, Roger Shimizu wrote:
> Dear Ian,
> 
> Thanks for your response!
> 
> I'd like to ask whether you could offer the "Reviewed-by" for my 
> first
> patch of Bug#793786?
> (0001-advance-the-timing-of-insmod-netconsole.patch)
> You must be a netconsole user that understand the reasoning why to
> load netconsole module with param should be earlier than calling
> "load_modules" routine.

I'm afraid I'm not, the pinnacle of my understanding of netconsole was
back when I wrote that blog post and basically everything I knew (which
wasn't much) was written there. I haven't used it much since that
debugging incident (although I keep thinking I should deploy it
properly).

Is the 1st patch basically because if netconsole is (or might be)
listed in the /etc/initramfs-tools/modules then it is loaded with
whatever options are given there and so when you come to do the load in
/sbin/init it is already loaded and present?

In which case that does seem to make sense, yes.

Does netconsole.netconsole=FOO (which is really
<module>.<param>=<value>, the module and param just happen to have the
same name) on the command line not get honoured by the load inside
load_modules too?

Even if it is then netconsole= is a nice convenience shortcut, so I
think that your first patch probably makes sense, at least so far as I
understand the implications.

Ian.


Reply to: