On Tue, Apr 25, 2000 at 10:56:04PM -0500, w trillich wrote
> John Pearson wrote:
> > 
> > On Mon, Apr 24, 2000 at 03:02:42PM +0200, Kovacs Istvan wrote
> > > Hello!
> > >
> > > I've just installed potato, re-compiled the kernel, and now (some of)
> > > my modules will not auto-load. I can load them with modprobe and/or
> > > insmod. I have to load the driver for my Initio SCSI card, SB AWE32
> > > sound card and RTL8019-based NIC by hand. The question is: why? And of
> > > course: how do I correct the situation?
> 
> [snip]
> 
> > After editing files in /etc/modutils always run update-modules,
> > so that /etc/conf.modules gets updated.
> 
> conf.modules? or modules.conf?
> 

Well, that apparently depends on which release you're tracking,
and when you last updated it.  I'm a slink man myself.

> my syslog was complaining of something quite similar, so maybe
> you already helped my problem a bit. from syslog...
> Apr 25 21:29:48 server modprobe: Note: /etc/modules.conf is more
> recent than /lib/modules/2.0.36/modules.dep
> Apr 25 21:29:48 server insmod: Note: /etc/modules.conf is more recent
> than /lib/modules/2.0.36/modules.dep
> 
> but i probably need more help yet...
> 

Others have reported similar problems after upgrading to recent
modutils, and have been told to run
# depmod -a
to update modules.dep.  Maybe that will help.

> i recently did the apt-get dist-upgrade and hadn't rebooted since...
> until today. now we've got similar troubles to those Kovacs wrote 
> about, above...
> 
> most important to us is getting connected to the 'net. what a pain!
> need all kinds of handwaving before it gets past "SIOCADDRT" errors
> and lets us ping out thru eth1 to the 'net.
> 
> by handwaving, i mean stabbing in the dark with tries like
>       init 1
>       ifup  # SIOCADDRT: Network is unreachable
>       /etc/init.d/networking start # SIOCADDRT error
>       ipmasq
>       ping some.host.org # also breaks
>       init 2
> repeating and stabbing until a ping actually works.
> then everything is fine even tho it seems sketchy...
> 
> right now, i tried
>       # ifup
>       SIOCADDRT: Network is unreachable
> eh?!
> 
> yet i can ping anywhere, POP and SMTP email, DNS is working
> like a champ. very odd...
> 

Can't really comment on that, as it's a potato thing with which
I'm not familiar.  

Did your apt-get dist-upgrade complete OK?  Do you see any
alarming messages if you run apt-get check?  It could be you're
only halfway through the upgrade.

> in case of accidental relevancy here are some log slices--
> 
> from /var/log/messages:
> Apr 25 22:03:05 server syslogd 1.3-3#33: restart.
> Apr 25 22:03:05 server kernel: klogd 1.3-3#33, log source = /proc/kmsg 
> started.
> Apr 25 22:03:05 server kernel: Cannot find map file.
> Apr 25 22:03:05 server kernel: Loaded 28 symbols from 6 modules.
> 
> from /var/log/syslog:
> Apr 25 22:03:05 server named[1896]: listening on [127.0.0.1].53 (lo)
> Apr 25 22:03:05 server named[1896]: listening on [192.168.1.1].53 (eth0)
> Apr 25 22:03:05 server named[1896]: listening on [208.33.90.85].53 (eth1)
> Apr 25 22:03:05 server named[1896]: Forwarding source address is 
> [0.0.0.0].1130
> Apr 25 22:03:05 server named[1897]: Ready to answer queries.
> Apr 25 22:03:05 server /sbin/rpc.statd[1901]: unable to register
> (SM_PROG, SM_VERS, udp).
> Apr 25 22:03:05 server ypbind[1915]: Unable to register (YPBINDPROG,
> YPBINDVERS, udp).
> Apr 25 22:03:06 server /usr/sbin/gpm[1924]: Skipping a data packet (?)
> Apr 25 22:03:08 server rwhod[2002]: sending on interface eth1
> Apr 25 22:03:08 server rwhod[2002]: sending on interface eth0
> Apr 25 22:03:09 server ntpdate[2005]: step time server 128.46.154.76
> offset -0.833622 sec
> Apr 25 22:03:09 server afpd[2018]: main: atp_open: Invalid argument
> Apr 25 22:03:09 server afpd[2018]: ASIP started on 208.33.90.85:548(0) 
> (1.4b2+asun2.1.3)
> Apr 25 22:03:09 server papd[2020]: restart (1.4b2+asun2.1.3)
> Apr 25 22:03:09 server proftpd[2026]: server - ProFTPD 1.2.0pre9
> standalone mode STARTUP
> Apr 25 22:03:09 server /usr/sbin/cron[2033]: (CRON) INFO (pidfile fd = 3)
> Apr 25 22:03:09 server /usr/sbin/cron[2034]: (CRON) STARTUP (fork ok)
> Apr 25 22:07:12 server modprobe: modprobe: Can't locate module net-pf-18
> Apr 25 22:07:14 server last message repeated 3 times
> 
> unable to register? invalid argument? missing module?
> i'm guessing "net-pf-18" is significant... how do i get that?
> 

The "net-pf-18" message is the kernel trying to load the module
for network protocol family 18, which linux/include/net/socket.h
lists as "Ash", with which I'm not familiar.  If it represents a
protocol family that you don't intend to support, or you just
want to get rid of the messages, add
alias net-pf-18 off
to /etc/modutils/aliases and then run update-modules.  I'd guess
it's something to do with AppleTalk, and I'd avoid getting
alarmed; the AppleTalk suite includes several protocols, only a
few of which you are likely to need to talk to Macs on your
local net.

"Unable to register" sounds like a problem with the portmapper
(rpc.portmap), which manages ypbind, rpc.statd and other rpc
services ; is it running?  Don't worry too much about this if
other things need fixing (unless you rely on yp to get things
working) - come back to it later, if it's still a problem once
things have stabilised.


John P.
-- 
[EMAIL PROTECTED]
[EMAIL PROTECTED]
http://www.mdt.net.au/~john Debian Linux admin & support:technical services

Reply via email to