Ubuntu 9.10 caused ntp to hang on lish

http://www.linode.com/forums/viewtopic.php?t=4782

?

Thanks I followed that and I can get in via SSH but I am still getting the error messages about NTP.

Any ideas about these? or are other people getting the same?

It might help if you tell us what the errors are :)

@ollietrex:

I have upgraded to ubuntu 9.10 and everything went ok apart from when I log into lish I get the screen.

• Stopping NTP server ntpd

• Starting NTP server ntpd

• Stopping NTP server ntpd

• Starting NTP server ntpd

This doesn't sound like a problem with ntpd at all, as for some reason under upstart ntpd restarts itself like it's going out of style. I've observed this on my own 9.10 install.

ntpd isn't hanging on your console, it's that you don't have anything listening on your Linode's virtual console. From SSH, try this:

mknod /dev/hvc0 c 229 0
vim /etc/init/hvc0.conf

Add this to hvc0.conf:

# hvc0 - getty for Linode console 

start on stopped rc RUNLEVEL=[2345] 
stop on runlevel [!2345] 

respawn 
exec /sbin/getty -8 38400 hvc0

Back to shell:

:wq
service hvc0 restart

And check LISH again.

Thanks for the update, I managed to sort the lish SSH and I will just ignore what is going on with the ntpd service.

FWIW, I haven't confirmed this at all or anything, but I suspect the ntpd restarts are because of ntpdate:

* Upstart starts ntpd

• Upstart brings up some network interface

• ntpdate ifup script stops ntpd, runs ntpdate, restarts ntpd

I have no proof, but it's the only plausible explanation I've thought of.

I eventually disabled the ifup script (/etc/network/if-up.d/ntpdate) when I was restarting my IPv6 tunnel interface while working on it, and the ntpd restarts were annoying me. The host runs ntpd, so unless something goes wrong, the clock will be accurate enough on boot, and ntpd can handle it from there. I'm not generally recommending it or anything, though.

Edit: Fix BBCode.