Print

[AMDinside]

Hopefully some people still check here who might be able to lend me a hand.

I am an intern at my college working in one of the labs.  All the clients run Linux.  I've picked up a lot about Linux since being here.  However I am having a hard time finding information for a problem we are having.

We found out that if we don't have an internet connection that ntpd hangs on boot, for over 10 minutes.  I am trying to find out if there is a way to change the ntpd boot timeout so if it can't get a connection  within say 30 seconds it aborts and will try again after the system has finished booting.

If anyone has any insight into this please let me know.

[GenTsoChiken]

yea, that went straight over my head.....

I hope pop sees this.

[GenTsoChiken]

http://kerneltrap.org/mailarchive/openbsd-misc/2008/9/6/3214194/thread

idk if you saw that or if it helps.

[Popsikle]

you should set a dummy host in your pool list to the local host.  This way if everything else is down your local server would respond. You could just do this on one box then point all the rest at it (best on the dhcpd server imho).  There is also a config variable "timeout milliseconds" but that might be a compile option.  Since your install doesnt have a dummy host sounds like your using debian/ubuntu breed?

and Psi, of course I sees it. I see all!

[GenTsoChiken]

uh, read again, I'm not psi.

[Popsikle]

see thats what happens when you seees everything but dont read shit lol.

sorry gtc <3

[AMDinside]

you should set a dummy host in your pool list to the local host.  This way if everything else is down your local server would respond. You could just do this on one box then point all the rest at it (best on the dhcpd server imho).  There is also a config variable "timeout milliseconds" but that might be a compile option.  Since your install doesnt have a dummy host sounds like your using debian/ubuntu breed?

and Psi, of course I sees it. I see all!

The machine is running CentOS 5.5, a RH/FC variant.

What is the default timeout?  I saw the ntpd hosts in the conf file, all pointing to the same location.  So if I changed the last one to localhost it should then progress if it can't reach the other hosts, correct?