I recently installed an older (celeron 633MHz) computer with debian. This pc is just for testing purpose. The computer has 2 network cards and is connected to 2 networks. Both cards have been configured with a static addresses. My interfaces file looks now like this: auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet static address 192.168.1.114 netmask 255.255.255.0 network 192.168.1.0 gateway 192.168.1.1 auto eth0 iface eth1 inet static address 192.168.2.114 netmask 255.255.255.0 network 192.168.2.0 Restarting the network with /etc/init.d/networking restart works fine but eth1 doesn't come up. so I figured I need to add 'auto eth1' to the end of the the file. When doing that en restarting the network I get some weird warning message: Reconfiguring network interfaces...if-up.d/mountnfs[eth0]: waiting for interface eth1 before doing NFS mounts (warning). done. I've never seen something like this before. But i guesse there are a lot of stuff I've never seen Like this everything works but it is quite annoying. Any ideas of how I do get rid of it? btw: i don't have any nfs mounts on my system (nor in fstab or in my network i think). Thanks in advance, Carew
solved I solved the problem. First I took a look where the error came from: /etc/network/if-up.d/mountnfs There is a line that refers to another file (in mountnfs I searched for the warning msg) and I came up with this file: /etc/network/run/ifstate No clue what it does. But in the file there are two lines: lo=lo eth0=eth0 And i just added eth1=eth1. Anyone does know what the ifstate file does?
not solved That's not the solution. After reboot the warning just came back. If anyone knows how to get rid of it. Let me know.
seems like there is no gateway configured for eth1? and auto eth1 goes on top of eth1 configurations.
I'll try it right away and post you results. The fact that there is no gw is because there is none. This network is not connected to the internet. But I guess I could give it a default gateway that doesn't exist? I don't think he would complain about that should he? I'll post my new /etc/network/interfaces file as I made the changes. Grtzz carew
oh btw, you can do ifconfig -a to see all of your adapters. sometimes when the NICs dont have an IP they do not show up in ifconfig alone you have to use -a to see them. then you can use; ifconfig eth1 inet 192.168.1.115 netmask 255.255.255.0 mtu 1500 up to give it a static IP.(on the command line) your interfaces file seems to be OK. Also you can make a tunneling (VPN) kind of NIC by tunctl -t tun0 then give it an IP with ifconfig
Well i did make the changes you said. And the gw doesn't change anything. I also moved the auto eth1 before the assingment of the ip and guess what it didn't change anything. /etc/network/interfaces: # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 auto eth0 iface eth0 inet static address 192.168.1.114 netmask 255.255.255.0 network 192.168.1.0 gateway 192.168.1.1 auto eth1 iface eth1 inet static address 192.168.2.114 netmask 255.255.255.0 network 192.168.2.0 gateway 192.168.2.110 Restarting networkinterfaces: $ sudo /etc/init.d/networking restart Reconfiguring network interfaces...if-up.d/mountnfs[eth0]: waiting for interface eth1 before doing NFS mounts (warning). done. $ Ifconfig -a shows all interfaces up and running (and they work just fine) but it is a annoying warning.
Oh, I am just curious why "allow hot-plug" for eth0? and not eth1? I think this is the problem bcs if you man interfaces it says allow hot-plug brings 1 NIC up so use auto eth0 and auto eth1 for both NICs and take out the allow hot-plug. anyway.. The message obviously comes from the script under /if-up.d/ folder called mountnfs (because its taking too long for eth1 to get config-ed) mountnfs: msg="if-up.d/mountnfs[$IFACE]: waiting for interface $i before doing NFS mounts" mountnfs: # Wait until all auto interfaces are up before attemting to mount mountnfs: exit_unless_last_interface (the misspelled word is inside script lol) Good Luck!
I did take out the line with no effect at all I guess. I don't seem to find in the manual pages something about allow hot-plug. Where did you find it? And I don't see any misspelled words in mountnfs neither. Grtzz and thanks for the help.
that is strange because I was almost sure that (hot plug) is the reason for the delay.(I read it some place..)? the man pages come up for me and I am using ubuntu 9.04 jaunty.(+updates). man interfaces brings up a lot of info. Oh and about the misspell... its kind of invisible, it says "before attemting".. and not "attempting"! lol, it dont matter since its not a command but a comment. I wonder if you change the order for the 2 NICs and put eth1 first, it would make any difference? just a shot in the dark really... Good Luck!
Where can I turn around the order for brining up the networkdevices? I have no clue. Thx for the help, Carew
just edit the interfaces file and put these in this order auto lo iface lo inet loopback auto eth1 iface eth1 inet static address 192.168.2.114 netmask 255.255.255.0 network 192.168.2.0 gateway 192.168.2.110 auto eth0 iface eth0 inet static address 192.168.1.114 netmask 255.255.255.0 network 192.168.1.0 gateway 192.168.1.1
well I switched the interfaces as you said (just copy past from you're post). i get the same warning just the way around now: $ sudo /etc/init.d/networking restart Reconfiguring network interfaces...if-up.d/mountnfs[eth1]: waiting for interface eth0 before doing NFS mounts (warning). done. $ weird, maybee I should just try to install nfs? And add some mounts to see if they all work? Thx for the help
wo.. that is really strange?! You got me. I am not really sure why it gives this message? As I said b4 it was in that script but I really thought it was the hot-plug statement that caused that. I also noticed that sometimes there are network configuration errors in the Xen guest machines (the virtual machines) bcs there are 2 sets of config files 1 in the host before the boot up and that is a xen config file which causes the VM to boot up and then there are of course interfaces files inside these guest machines. I wonder if this problem is due to having two different subnets? 192.168.1 and 192.168.2? have you tried to play around with netmask values? like changing them from 255.255.255.0 to 255.255.0.0 ? just a shot in dark really, I wished I knew the answer, it is an interesting problem! You see I have one NIC in my machine and then I created a virtual one as well and I called tun0 and my eth0 address is like 192.168... and my tun0 is 10.1... 2 very different addressing scheme but I have no problem in fact I wrote this (tunctl -t tun0) in a form of script and have it execute at the boot up then configured my interfaces file to give tun0 its address. tun is for tunneling or vpn use, may be you should try using tun0 instead of your NIC just to try it out? sorry for a long post , I am not expert as you know but am trying to learn and been hacking on these boxes here at work for a while so I have seen a lot of errors before but nothing like that...anyway... Good Luck!
You can ignore this message about mount nfs. It's normal and doesnt impact network performance. I'd also suggest putting all auto config lines on one line. For example auto eth0 eth1 eth0:1 eth0:2 etc etc In some cases if you run the network script and have many interfaces configure the interfaces wont go up if you have them on separate lines such as auto eth0 auto eth1 auto eth0:1 etc
I'll just ignore it. Because I'd like to install some nfs-shares. I just don't know yet where or how. I gotta think about that first. But I can confirm (but not proof anymore) that I had a computer with the same setup a few years ago (different mb and processor) and I didn't get the warning. Thanks for the help anyways. Grtzz, Carew @ norman: I did put al the auto's on one line (it doesn't change anything)
