Jonathan Adams wrote: >On Thu, Jun 15, 2006 at 04:44:06PM +0800, Darren Reed wrote: > >>David Powell wrote: >> >> >>>On Wed, Jun 14, 2006 at 09:56:09PM +0800, Darren Reed wrote: >>> >>> >>>>So what do I think is wrong here? >>>>1) "svcadm disable ntp" has no affect on anything >>>> >>>> >>>False. It disabled the service. We assume you want your services >>>shut down cleanly, and part of that is waiting for a partially >>>started service to finish starting cleanly. The NTP start method has >>>a timeout of 30 minutes; perhaps that should be shortened. >>> >>> >>Ugh...30 minutes? How many others have timeouts like this? >>I'd say 3 seconds was bordering on too long for ntpdate to return. >> > >What if you have a busy server, and NTP takes an uncorrectable memory fault >and needs to be restarted. If that takes more than 3 seconds, then your >timeout is too short, and it'll end up in maintenance. > >The timeouts are conservative for a reason. 30 minutes may be too long, but >three seconds is way too short. >
Or if the NTP server is rebooting or.... Come on, lets get real - this is ntpdate. If ntpdate fails, it shouldn't be putting NTP into maintanance at all. The idea here is to start up xntpd which is quite capable of handling the task that ntpdate is there to do, in its own time. The reason for using ntpdate in the script is to quickly sync up the system clock at bootup if there is a large discrepency rather than wait for xntpd to bring it into line over time (it can take quite a while.) In other words, whether or not ntpdate succeeds is not really very important in the overall scheme of things NTP related. Darren
