but have hit some snags where certain commands are not supported in
busybox (service start/stop/reset, chkconfig). At boot ntpd looks to
read from somewhere:
Running ntpdate: 16 Dec 17:39:43 ntpdate[411]: step time server
155.101.3.114 offset 1260988767.511027 sec
done
but issuing an ntpdate command returns:
16 Dec 18:22:01 ntpdate[444]: no servers can be used, exiting
what can I do to test that this is working? 'pgrep ntpd' returns only
one process number where the above link makes it sound I should be
seeing several.
I've posted at AVRFreaks, but they've been slow to respond. I was
wondering whether this would be a more appropriate place to post.
Thanks.
The ntp.conf file typically contains from four to seven server
statements which look like:
server 192.168.1.23
server 10.1.1.13
etc.
You may use the names instead of the numeric IP addresses.
> Running ntpdate: 16 Dec 17:39:43 ntpdate[411]: step time server
> 155.101.3.114 offset 1260988767.511027 sec
> done
>
> but issuing an ntpdate command returns:
>
> 16 Dec 18:22:01 ntpdate[444]: no servers can be used, exiting
>
An ntpdate command requires the name or IP address of a server.
> what can I do to test that this is working? 'pgrep ntpd' returns only
> one process number where the above link makes it sound I should be
> seeing several.
No! There should be one and only one ntpd process!
<snip>
busybox is a statically linked multipurpose single binary executable.
Swiss army knife for small/embedded systems.
behaves dependent on as what it was called, i.e. what is in *argv[0]
uwe
> but have hit some snags where certain commands are not supported in
> busybox (service start/stop/reset, chkconfig).
You don't need those commands. Just run the init scripts directly.
> At boot ntpd looks to
> read from somewhere:
>
> Running ntpdate: 16 Dec 17:39:43 ntpdate[411]: step time server
> 155.101.3.114 offset 1260988767.511027 sec
> done
ntpd != ndpdate
> but issuing an ntpdate command returns:
>
> 16 Dec 18:22:01 ntpdate[444]: no servers can be used, exiting
ntpdate needs at least one time server name on the command-line.
> what can I do to test that this is working? 'pgrep ntpd' returns only
> one process number where the above link makes it sound I should be
> seeing several.
1) ps ax | grep ntpd may show something like this:
3458 pts/2 R+ 0:00 grep ntpd
26773 ? Ss 0:02 /usr/sbin/ntpd -p /var/run/ntpd.pid -u 103:104 -g -N
2) try 'ntpq -p' and 'ntpq -crv'
--
Steve Kostecke <kost...@ntp.org>
NTP Public Services Project - http://support.ntp.org/
service is just a script/program to run one of the start/stop
scritps. It may be a redhat/fedora/linux thing. google will
find the man page.
chkconfig is a utility to setup the links for starting
and stopping services. Again, there is a man page.
If you aren't familiar with how servers get started/stopped,
look in /etc/rc.d and friends.
--
These are my opinions, not necessarily my employer's. I hate spam.
> service is just a script/program to run one of the start/stop
> scritps. It may be a redhat/fedora/linux thing. google will
> find the man page.
> chkconfig is a utility to setup the links for starting
> and stopping services. Again, there is a man page.
> If you aren't familiar with how servers get started/stopped,
> look in /etc/rc.d and friends.
Would this then be a case of it being better to at least occasionally
build one's mnemonic memory circuits from stone knives and bear
skins?-) Or put another way, an example of the perils of speed-dial...
rick jones
--
I don't interest myself in "why." I think more often in terms of
"when," sometimes "where;" always "how much." - Joubert
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...