Date: Sat, 13 May 2006 14:51:03 -0500 From: Kevin Kinsey <kdk@daleco.biz> To: Barnaby Scott <bds@waywood.co.uk> Cc: freebsd-questions@freebsd.org Subject: Re: very slow boot (newbie) Message-ID: <446638A7.50108@daleco.biz> In-Reply-To: <44662536.4080700@waywood.co.uk> References: <44639855.90102@waywood.co.uk> <4463C5E4.50109@daleco.biz> <4463D2EC.1020100@waywood.co.uk> <20060512125648.GG5531@catflap.slightlystrange.org> <4464A7E5.7060000@waywood.co.uk> <20060512175626.GD34035@catflap.slightlystrange.org> <44662536.4080700@waywood.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
Barnaby Scott wrote:
> Many thanks to all who have helped me on this one - I won't post a
> message in response to every suggestion, but they have all helped -
> thank you!
>
> It turns out it was sendmail causing the delay, so now my /etc/rc.comf
> reads:
>
> hostname="frankbruno"
> ifconfig_re0="DHCP"
> keymap="uk.iso"
> linux_enable="YES"
> moused_enable="YES"
> saver="logo"
> usbd_enable="YES"
> sendmail_enable="NONE"
>
> (Someone asked what was in this file). The last line that I have just
> added does cure the problem. But...
>
> ...it does not cure the problem for me if I decide that I do want
> sendmail! I could cross that bridge when I come to it, but I would
> prefer to gain some insight here if anyone can bear any more on this topic.
>
> Adding
> 127.0.0.1 frankbruno
> to /etc/hosts did not cure the problem. Could that be because the lookup
> that causes the delay is a reverse one? If so, it would be trying to
> find a name for 192.168.0.4 (I think that's the one I have been getting
> recently) which is still not in hosts.
>
> I would rather not mess with the IP allocation if possible - having it
> automatic is much more useful and means I cannot create condradictory
> records in different places.
>
> Looking in /var/run/dmesg.boot turned up nothing obvious to me at least,
> booting in Safe Mode made no difference, and verbose logging turned up
> nothing.
>
> However, I did discover a tip posted a couple of years ago, and that was
> to press ^T when the boot stalled. God knows what this does, but it
> turned up the following response:
>
> load: 0.85 cmd:sendmail 454 [kqread] 0.00u 0.01s 0% 1912k
>
> I thinks that answers all the things that were suggested - can anyone
> see a way of reinstating sendmail without the stalled boot process, and
> without having to reserve specific IP addresses for each computer?
>
> Barnaby Scott
>
Prequel: I know nothing about dhclient/DHCP interfaces other
than if I want one, I can put "ifconfig foo1="DHCP" in rc.conf.
If you can figure out what your address is post-booting, you
should be able to create a simple wrapper script and run it from
cron via the "@reboot" tag.
Pseudo-code:
sleep(5 min);
write "my.ip.ad.ress" >> /etc/hosts
sh /etc/rc.d/sendmail start
But, I'd strongly advise you to use a static IP if your
host needs to be an MTA. Otherwise, configure your MUA's
to use a real MTA (which have static IP's, DNS, etc., etc.).
Kevin Kinsey
--
I didn't get sophisticated -- I just got tired.
But maybe that's what sophisticated is -- being tired.
-- Rita Gain
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?446638A7.50108>