On Friday 25 May 2001 22:44, Jeff Lee wrote: > I've run into an interesting problem with ntpdate- we have it on most of > our servers, run once on boot as is the standard. The /etc/init.d/ntpdate > file is configured correctly, and for most of our systems it works fine. > The other day, my manager moved a system to another part of the office, > where it doesn't have network connectivity. When the ntpdate script > executed, rather than timing out after about 5 seconds (1 sec default > timeout on waiting for response from server * 5 loops), it hung the machine > for over an hour before he decided to do a hard reset. The machine wouldn't
I think that should be a grave bug in the package. Is it already reported as such in the BTS? > that, one possibility is that when /etc/init.d/ntpdate is run, and ntpdate > exits with an abnormal code, the system doesn't realize that ntpdate isn't > running, and locks there? Any advice would be appreciated :) Another thing you can do is enable the SAK which easily enables you to kill the offending process (such init script programs tend to have /dev/console open at the time they fail). -- http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/projects.html Projects I am working on http://www.coker.com.au/~russell/ My home page