Inaccessible bubba not acquiring network address

[jonj1]

Hi,

I have a bubba2 server that is/was configured to acquire network addresses via dhcp on both lan and wan. It's connected to a netgear adsl router via the lan port. To keep it short, I ran the command

Code: Select all

/etc/init.d/networking restart

on bubba over an ssh connection. Bad move. Bubba had previously acquired 'the wrong' ip address during a router re-boot and by NFS shares were not working. I thought restarting networking would fix it. Being over an ssh connection, the restart command didn't complete and now bubba doesn't acquire a network address at all. What can be done to recover it?

thanks,

[carl]

Hi,

I have a bubba2 server that is/was configured to acquire network addresses via dhcp on both lan and wan. It's connected to a netgear adsl router via the lan port. To keep it short, I ran the command

Code: Select all

/etc/init.d/networking restart

on bubba over an ssh connection. Bad move. Bubba had previously acquired 'the wrong' ip address during a router re-boot and by NFS shares were not working. I thought restarting networking would fix it. Being over an ssh connection, the restart command didn't complete and now bubba doesn't acquire a network address at all. What can be done to recover it?

thanks,

If you can't now get contact to the bubba at all, I would recommend using a rescue disk to ssh into the bubba and make the necessary changes.

/Carl

[jonj1]

If you can't now get contact to the bubba at all, I would recommend using a rescue disk to ssh into the bubba and make the necessary changes.

/Carl

Thanks, will give it a go.

[jonj1]

Hi again,

I've attached a rescue disk and accessed bubba|two in rescue mode. The /dev/sda1 root partition mounts ok so I'm having a look around but I'm not sure where to go from here. These config files look ok:

Code: Select all

# pwd
/root/sda1/etc
# cat resolv.conf 
nameserver 192.168.0.1
# cat network/interfaces 
auto lo
iface lo inet loopback

iface eth1 inet dhcp

iface eth0 inet dhcp

There's nothing in sda1/var/log/boot. This is the tail end of kern.log:

Code: Select all

Mar 18 21:06:24 bubba kernel: PHY: e0024520:00 - Link is Up - 100/Full
Mar 18 21:06:24 bubba kernel: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
Mar 18 21:06:34 bubba kernel: eth1: no IPv6 routers present
Mar 18 23:17:19 bubba kernel: nfsd: peername failed (err 107)!
Mar 18 23:22:12 bubba kernel: PHY: e0024520:00 - Link is Up - 100/Full
Mar 18 23:22:19 bubba kernel: eth1: no IPv6 routers present
Mar 18 23:27:56 bubba kernel: BUBBATWO: Timeout!<7>BUBBATWO: Invoking usermodehelper<4>nfsd: last server has exited
Mar 18 23:27:56 bubba kernel: nfsd: unexporting all filesystems
Mar 18 23:27:57 bubba kernel: Kernel logging (proc) stopped.
Mar 18 23:27:57 bubba kernel: Kernel log daemon terminating.

Any ideas how to proceed greatly appreciated!

[jonj1]

Looking at /etc/interfaces on the rescue disk I realise there was a problem with mine on bubba, and it now reads

Code: Select all

auto lo
iface lo inet loopback

auto eth1
iface eth1 inet dhcp 

iface eth0 inet dhcp 

This gets bubba a little further, so that it acquires an ip address from the router. SSH and HTTP don't come up though, so still inaccessible...

[jonj1]

Well to complete the thread, bubba is finally back online. Once 'interfaces' was corrected it took a really long time to come back up and when it did, the file system was read only. Back into rescue mode and

Code: Select all

fsck.ext3 /dev/sda1

cleared it up...