Re: [gentoo-user] eth0 not up when init script returns?

2006-05-06 Thread Matthias Bethke
Hi Hani,
on Thursday, 2006-05-04 at 11:19:33, you wrote:
 Have you looked through the '/etc/conf.d/net.example' file?  I'm not too
 familiar with DHCP, but the net.example file has this entry:

As Uwe said, that's not the issue. It's a server box, the one
responsible for dealing out the others' addresses via DHCP. eth0 has a
fixed IP.

cheers!
Matthias

-- 
I prefer encrypted and signed messages. KeyID: FAC37665
Fingerprint: 8C16 3F0A A6FC DF0D 19B0  8DEF 48D9 1700 FAC3 7665


pgpoX8QclaoRH.pgp
Description: PGP signature


Re: [gentoo-user] eth0 not up when init script returns?

2006-05-06 Thread Matthias Bethke
Hi Vladimir,
on Thursday, 2006-05-04 at 17:56:58, you wrote:
 Anyway, I figured out what my problem was. I was starting eth0 twice,
 once through an rc script, and once with ifplugd. When I zapped the rc
 script (rc-update del net.eth0), things started work better.

Still sounds like an ifplugd config problem. As I learnt from this
list a while ago, you *should* start both ifplugd and the net.eth*
scripts---it actually works better like this, e.g. auto-joining wireless
networks via ifplugd never worked until I re-added the init script for
eth1. Here's my /etc/conf.d/ifplugd:

INTERFACES=eth0 eth1
AUTO=yes
BEEP=yes
IGNORE_FAIL=yes
IGNORE_FAIL_POSITIVE=no
IGNORE_RETVAL=yes
POLL_TIME=2
DELAY_UP=0
DELAY_DOWN=0
API_MODE=auto
SHUTDOWN=no
WAIT_ON_FORK=no
MONITOR=no
ARGS=
MONITOR_eth1=yes
DELAY_UP_eth1=5
DELAY_DOWN_eth1=5

cheers!
Matthias
-- 
I prefer encrypted and signed messages. KeyID: FAC37665
Fingerprint: 8C16 3F0A A6FC DF0D 19B0  8DEF 48D9 1700 FAC3 7665


pgpSYjozXPG6G.pgp
Description: PGP signature


Re: [gentoo-user] eth0 not up when init script returns?

2006-05-04 Thread Hani Duwaik
On 5/4/06, Matthias Bethke [EMAIL PROTECTED] wrote:
I just noticed a strange problem on our server that's just been switchedto Gentoo:It's running dhcpd, which init starts right after bringing up thenetwork interface. But dhcpd quits, complaining it couldn't listen on
eth0 because it had address 0.0.0.0. So it seems the interface isn'tfully up yet when the init script returns---probably because theBroadcom Tigon driver seems to be on the slow side when changing
parameters, while the CPU is plenty fast. I've now tried to fix it witha postup() function that simply does a sleep 3, but that's prettyhacky IMHO. Isn't thatere a way to do it properly? I think the script
should ensure its jobs are finished before it returns in any case.Have you looked through the '/etc/conf.d/net.example' file? I'm not too familiar with DHCP, but the net.example file has this entry:


	
	
	
	
	
	

# For passing custom options to dhcpcd
use something like the following.  This
# example reduces the timeout for
retrieving an address from 60 seconds (the
# default) to 10 seconds.
#dhcpcd_eth0=-t 10
=Not sure if it will help, but maybe increasing the wait time might work.-Hani-- If, of the many truths, you select only one and follow it blindly, it will become a falsehood, and you a fanatic.


Re: [gentoo-user] eth0 not up when init script returns?

2006-05-04 Thread Vladimir G. Ivanovic
I have a similar problem to Matthias. /etc/init.d/net.eth0 no longer
returns, so spamd, sendmail, saslauthd, etc. never start up. It used to
work...  In my case, the interface is up, but `/etc/init.d/net.eth0
status' says starting. If I try to start, say, sendmail, it says

WARNING:  sendmail is scheduled to start when net.eth0 has started.

and it never starts. Something has changed in the last couple of weeks,
and I don't know what it is. I'm using hotplug.

Thanks for any help.

--- Vladimir

Gentoo 2.6.16-r6, fully updated.

On Thu, 2006-05-04 at 11:19 -0600, Hani Duwaik wrote:
 
 
 On 5/4/06, Matthias Bethke [EMAIL PROTECTED] wrote:
 I just noticed a strange problem on our server that's just
 been switched
 to Gentoo:
 It's running dhcpd, which init starts right after bringing up
 the
 network interface. But dhcpd quits, complaining it couldn't
 listen on 
 eth0 because it had address 0.0.0.0. So it seems the interface
 isn't
 fully up yet when the init script returns---probably because
 the
 Broadcom Tigon driver seems to be on the slow side when
 changing 
 parameters, while the CPU is plenty fast. I've now tried to
 fix it with
 a postup() function that simply does a sleep 3, but that's
 pretty
 hacky IMHO. Isn't thatere a way to do it properly? I think the
 script 
 should ensure its jobs are finished before it returns in any
 case.
 
 Have you looked through the '/etc/conf.d/net.example' file?  I'm not
 too familiar with DHCP, but the net.example file has this entry: 
 
  
 
 # For passing custom options to dhcpcd use something like the
 following. This
 
 # example reduces the timeout for retrieving an address from 60
 seconds (the
 
 # default) to 10 seconds.
 
 #dhcpcd_eth0=-t 10
 
 =
 
 Not sure if it will help, but maybe increasing the wait time might
 work.
 
 -Hani
 -- 
 If, of the many truths, you select only one and follow it blindly, it
 will become a falsehood, and you a fanatic.

Vladimir G. Ivanovic
Palo Alto, CA 94306
+1 650 678 8014

-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] eth0 not up when init script returns?

2006-05-04 Thread Uwe Thiem
On 04 May 2006 18:19, Hani Duwaik wrote:
 On 5/4/06, Matthias Bethke [EMAIL PROTECTED] wrote:
  I just noticed a strange problem on our server that's just been switched
  to Gentoo:
  It's running dhcpd, which init starts right after bringing up the
  network interface. But dhcpd quits, complaining it couldn't listen on
  eth0 because it had address 0.0.0.0. So it seems the interface isn't
  fully up yet when the init script returns---probably because the
  Broadcom Tigon driver seems to be on the slow side when changing
  parameters, while the CPU is plenty fast. I've now tried to fix it with
  a postup() function that simply does a sleep 3, but that's pretty
  hacky IMHO. Isn't thatere a way to do it properly? I think the script
  should ensure its jobs are finished before it returns in any case.

 Have you looked through the '/etc/conf.d/net.example' file?  I'm not too
 familiar with DHCP, but the net.example file has this entry:

 

 # For passing custom options to dhcpcd use something like the following.
 This

 # example reduces the timeout for retrieving an address from 60 seconds
 (the

 # default) to 10 seconds.

 #dhcpcd_eth0=-t 10
 =

dhcpcd != dhcpd

Uwe

-- 
Why do consumers keep buying products they will live to curse?
-- 
gentoo-user@gentoo.org mailing list



Re: [gentoo-user] eth0 not up when init script returns?

2006-05-04 Thread Vladimir G. Ivanovic
On Thu, 2006-05-04 at 10:56 -0700, Vladimir G. Ivanovic wrote:
 I have a similar problem to Matthias. /etc/init.d/net.eth0 no longer
 returns, so spamd, sendmail, saslauthd, etc. never start up. It used to
 work...  In my case, the interface is up, but `/etc/init.d/net.eth0
 status' says starting. If I try to start, say, sendmail, it says
 
 WARNING:  sendmail is scheduled to start when net.eth0 has started.
 
 and it never starts. Something has changed in the last couple of weeks,
 and I don't know what it is. I'm using hotplug.
 
Of course I meant ifplugd.

Anyway, I figured out what my problem was. I was starting eth0 twice,
once through an rc script, and once with ifplugd. When I zapped the rc
script (rc-update del net.eth0), things started work better.

--- Vladimir

Vladimir G. Ivanovic
Palo Alto, CA 94306
+1 650 678 8014

-- 
gentoo-user@gentoo.org mailing list