On Mon, 11 Aug 2003 13:01:39 +0200 Rudy Gevaert <[EMAIL PROTECTED]> wrote:
> I noticed that when my server gets rebooted the /etc/init.d/ntpdate > script doesn't get run. > > Even though there is a symlink to it in the rcS.d directory. > > Where can I see that the rcS.d directory gets called? So I am sure > all the scripts in there get run. (I'm actually nearly sure some > don't as ntpd doesn't get run). You can't see that "the rcS.d directory gets called", but you can tell when those scripts are being run by the console messages. For example, "Checking all filesystems ..." tells you S20checkfs.sh is running, and "Mounting local filesystems ..." is the message from S35mountall.sh. "Setting the System Clock using the Hardware Clock as a reference ..." is for hwclock.sh (I changed its number on my system). The very next message should be "Running ntpdate to synchronize clock." (The period at the end indicates the command finished.) It is easy to miss. Looking at the ntpdate script in /etc/init.d, I noted that ntpdate is called with several options. The -s option redirects output to syslog, but syslog isn't started until after all the scripts in rcS.d are run and the system changes to runlevel 2. So I replaced -s with -v, and now I get a message on the console telling me how much the time was adjusted. Kevin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
