NTP won't update

[Nathaniel Hall]

Hi all,

I am trying to get ntp running on a couple of Fedora Core 3 machines.  
It seems that ntpd is running, however when I stop the ntpd service and 
try to run ntpdate, it does nothing.

I have configured ntp on multiple Fedora Core 3 systems, but cannot get 
it to run correctly on two of them.  NTP was installed when the system 
was installed and I have even uninstalled and reinstalled the rpm with 
the same result.

Has anybody had the same problem or have any ideas?

-- 
Nathaniel Hall, GSEC
Intrusion Detection and Firewall Technician
Ozarks Technical Community College -- Office of Computer Networking

halln(a)otc.edu
417-447-7535

[Kevin Wang]

On Tue, 30 Nov 2004 13:16:52 -0600, Nathaniel Hall <halln(a)otc.edu&gt; wrote:
>  I am trying to get ntp running on a couple of Fedora Core 3
> machines.
>  It seems that ntpd is running, however when I stop the ntpd service and
>  try to run ntpdate, it does nothing.
>  
>  I have configured ntp on multiple Fedora Core 3 systems, but cannot get
>  it to run correctly on two of them.  NTP was installed when the system
>  was installed and I have even uninstalled and reinstalled the rpm with
>  the same result.
>  
>  Has anybody had the same problem or have any ideas? 
What commands are you executing?

What does your /etc/ntp.conf config file look like?  Note that just
adding servers lines to the default config file doesn't work, you also
need to loosen the security to let that site send you time updates.

and also firewall config, though if 'ntpdate pool.ntp.org' works, then
ntpd should (though you can only run one or the other)  There is a
flag that'll let you bind to a different port and run ntpdate even if
ntpd is running, though.

   - Kevin

[Nathaniel Hall]

I am executing ntpdate /timeserver/.

Here is my ntpq -p:

    ntpq: read: Connection refused

Like I said, I have ntp running correctly on several Fedora Cor 3 
servers, as well as RHAS3 servers without problems.  I have double 
checked our firewalls and do not see a problem.  The firewall on the 
local box is disabled so that shouldn't be a problem.  I did ask the guy 
who actually installed the system if he enabled SELinux and he did.  The 
only thing that still throws me off is that we have other systems with 
SELinux enabled that didn't give us any problems.

Nathaniel Hall, GSEC
Intrusion Detection and Firewall Technician
Ozarks Technical Community College -- Office of Computer Networking

halln(a)otc.edu
417-447-7535



Kevin Wang wrote:

> On Tue, 30 Nov 2004 13:16:52 -0600, Nathaniel Hall
> <halln(a)otc.edu&gt; wrote:
>   
>
> >I am trying to get ntp running on a couple of Fedora Core 3 machines.
> >It seems that ntpd is running, however when I stop the ntpd service and
> >try to run ntpdate, it does nothing.
> >
> >I have configured ntp on multiple Fedora Core 3 systems, but cannot get
> >it to run correctly on two of them.  NTP was installed when the system
> >was installed and I have even uninstalled and reinstalled the rpm with
> >the same result.
> >
> >Has anybody had the same problem or have any ideas?
> >    
> >
>
> What commands are you executing?
>
> What does your /etc/ntp.conf config file look like?  Note that just
> adding servers lines to the default config file doesn't work, you also
> need to loosen the security to let that site send you time updates.
>
> and also firewall config, though if 'ntpdate pool.ntp.org' works, then
> ntpd should (though you can only run one or the other)  There is a
> flag that'll let you bind to a different port and run ntpdate even if
> ntpd is running, though.
>
>    - Kevin
>   
>

[Kevin Wang]

On Tue, 30 Nov 2004 13:35:30 -0600, Nathaniel Hall <halln(a)otc.edu&gt; wrote:
>   I am executing ntpdate timeserver. 
>   
>   Here is my ntpq -p:
>   
>       ntpq: read: Connection refused
>   
>   Like I said, I have ntp running correctly on several Fedora Cor 3 servers,
>  as well as RHAS3 servers without problems.  I have double checked our
>  firewalls and do not see a problem.  The firewall on the local box is
>  disabled so that shouldn't be a problem.  I did ask the guy who actually
>  installed the system if he enabled SELinux and he did.  The only thing that
>  still throws me off is that we have other systems with SELinux enabled that
>  didn't give us any problems. 
ntpq tries to talk to the ntpd running on your machine, so it likely
isn't running.

but again, please post your ntp.conf. that will help immensely.

and the output of "ntpdate pool.ntp.org" would help too.

   - Kevin

[Nathaniel Hall]

Here is the configuration of my ntp.conf (comments removed)

    server /Our internal time server/
    server ntp-cup.external.hp.com
    server time.nist.gov
    server 127.127.1.0
    fudge 127.127.1.0 stratum 10

    restrict 127.0.0.1
    restrict 0.0.0.0 mask 255.255.255.255 nomodify notrap noquery notrust

    driftfile /etc/ntp/drift
    multicastclient                 # listen on default 224.0.1.1
    broadcastdelay  0.008

    authenticate no

ntpdate seems to have started working now, however, ntpq -p does not.  
After syncing with the time servers using ntpdate, I start the service 
and run ntpq -p and get this response:

     remote           refid      st t when poll reach   delay   offset  
jitter
==============================================================================
 emh1.otc.edu    .RSTR.          16 u    -   64    0    0.000    0.000 
4000.00
 ntp-cup.externa .RSTR.          16 u    -   64    0    0.000    0.000 
4000.00
 time.nist.gov   .RSTR.          16 u    -   64    0    0.000    0.000 
4000.00
 LOCAL(0)        LOCAL(0)        10 l   62   64    7    0.000    0.000   
0.004

The ntpdates synced without a problem.  I have other servers using the 
same ntp.conf file without a problem.  Any ideas?

Nathaniel Hall, GSEC
Intrusion Detection and Firewall Technician
Ozarks Technical Community College -- Office of Computer Networking

halln(a)otc.edu
417-447-7535



Kevin Wang wrote:

> On Tue, 30 Nov 2004 13:35:30 -0600, Nathaniel Hall
> <halln(a)otc.edu&gt; wrote:
>   
>
> > I am executing ntpdate timeserver. 
> > 
> > Here is my ntpq -p:
> > 
> >     ntpq: read: Connection refused
> > 
> > Like I said, I have ntp running correctly on several Fedora Cor 3 servers,
> >as well as RHAS3 servers without problems.  I have double checked our
> >firewalls and do not see a problem.  The firewall on the local box is
> >disabled so that shouldn't be a problem.  I did ask the guy who actually
> >installed the system if he enabled SELinux and he did.  The only thing that
> >still throws me off is that we have other systems with SELinux enabled that
> >didn't give us any problems.
> >    
> >
>
> ntpq tries to talk to the ntpd running on your machine, so it likely
> isn't running.
>
> but again, please post your ntp.conf. that will help immensely.
>
> and the output of "ntpdate pool.ntp.org" would help too.
>
>    - Kevin
>
>   
>

[Thomas Zehetbauer]

The meaning of 'notrust' was radically changed, don't use it anymore.

Tom

-- 
  T h o m a s   Z e h e t b a u e r   ( TZ251 )
  PGP encrypted mail preferred - KeyID 96FFCB89
      finger thomasz(a)hostmaster.org for key

The horizon of many people is a circle with a radius of zero.
They call this their point of view.

[Harry Hoffman]

Nathaniel,

Are you running ntpdate with a server argument?

i.e.-> ntpdate louie.udel.edu

also when you run: ntpq -p

what does the output show?

--Harry


Nathaniel Hall wrote:
>  Hi all,
>  
>  I am trying to get ntp running on a couple of Fedora Core 3 machines.  
>  It seems that ntpd is running, however when I stop the ntpd service and 
>  try to run ntpdate, it does nothing.
>  
>  I have configured ntp on multiple Fedora Core 3 systems, but cannot get 
>  it to run correctly on two of them.  NTP was installed when the system 
>  was installed and I have even uninstalled and reinstalled the rpm with 
>  the same result.
>  
>  Has anybody had the same problem or have any ideas?
>

[Rodolfo J. Paiz]

On Tue, 2004-11-30 at 13:16 -0600, Nathaniel Hall wrote:
>  I have configured ntp on multiple Fedora Core 3 systems, but cannot
> get 
>  it to run correctly on two of them.  NTP was installed when the system 
>  was installed and I have even uninstalled and reinstalled the rpm with 
>  the same result.
>   
Please post your ntp.conf.

>  Has anybody had the same problem or have any ideas?
>   
Are you using the "notrust" keyword? The meaning of it changed radically
from 4.1 to 4.2 (which IMHO is asinine... never redefine a keyword or
parameter which drastically changes functionality), so now in most
settings you should not use this keyword.

What do the logs show when the ntpd service is restarted?

Cheers,

-- 
Rodolfo J. Paiz <rpaiz(a)simpaticus.com&gt;