At boot time there are a lot of messages that fly by that I want to read, but its too fast for me to read. Obviously I can look at dmesg at any time, but not everything is there. The same applies if I shift+pageup when I get to a login prompt (init:3).

Syslog/messages omit the same info.

Is there a way to get at these messages? Its not terribly important as my system works ok, but they are errors of some sort and I would like to correct them...

thanks, tobyl

I haven't yet encountered a situation
where I couldn't find my boot-time screens
output in one of the logs ...

Maybe you need to do a
grep -ri "characteristic word" /var/log/*

Cheers,
Tink

Great idea thanks - just need to make an intelligent guess for "characteristic word" as I cant read them. Or, my girlfriend has sharp eyes, I'll see if she can read it !

cheers, tobyl

"dmesg | less" will display one page of dmesg at a time, and "dmesg | grep something" will spit out any reference to something.

You guys don't get what he's saying. I've noticed this too. For example, your system detects a new ethernet card you put in. dmesg might not show it. This was just an example, but you see what he means?

After booting, and as soon as you get the login prompt, hold down the Shift key and press Page-Up to read the boot messages. Yes, dmesg | less will not show everything that went on during boot time. Still, dmesg | less is very helpful.

Thanks for your replies. Artimus has hit the nail on the head. There's stuff flying by that is not recorded elsewhere. I think it is hardware related, and it comes just after a reference to syslog. Then I get about '2 pages' of stuff that I cant read, then it slows down again and carries on with the normal stuff.
I may try recompiling again with extra special care for un-needed hardware.
I tried disabling the rc.hotplug thing to see what happened, but it did not make any difference other than to screw my nvidia driver, so I put that back.

If he were using mandy or "little red riding hood" I'd
understand what you are on about... however:

My Slack doesn't do any "detects a new ethernet card"
type of things since I don't run any kudzus, anacondas
or other nasty things that might modify my box' setup
(remember the words of the windows-victim:"Never run a
changing system!") without my knowledge ... what did you
do to yours?

Cheers,
Tink

You might want to read the following thread:
https://www.redhat.com/archives/redh.../msg00389.html

Good luck,
FreddyH

Thanks FreddyH,
I read that. The main thrust of the thread is that

# Run gettys in standard runlevels
1:12345:respawn:/sbin/mingetty --noclear tty1

can be temporarily added to inittab. This sounds like it would work, only thing is that Slackware uses agetty not mingetty, which does not have the --noclear option.

I'm not feeling confident enough to start messing here.
My particular messages relate to USB & PCI (I think), however it does not seem to adversely affect my system, so I have just been putting up with it.

thanks again, tobyl