No syncro - NTP

This is a discussion on No syncro - NTP ; Hello I have a two computers, A and B. Version xntp 4.1.1. I have too a server ntp (S). My problem is the following one: Computer A, run xntpd start, connect with server (S), update the hour and It works ...

+ Reply to Thread
Results 1 to 13 of 13

Thread: No syncro

  1. No syncro

    Hello

    I have a two computers, A and B. Version xntp 4.1.1. I have too a
    server ntp (S).

    My problem is the following one:

    Computer A, run xntpd start, connect with server (S), update the hour
    and It works correctly.

    Computer B, run xntpd start, connect with server (S), It updates the
    hour but alone this time.

    Computer A:

    ntpq> pe
    remote refid st t when poll reach delay
    offset jitter
    ================================================== ====================
    emaddns1.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    4000.00
    emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    4000.00

    ntpq> rv
    status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    version="ntpd 4.1.1@1.786 Wed Oct 16 12:55:05 UTC 2002 (1)",
    processor="i686", system="Linux2.4.21-278-default", leap=11,
    stratum=16,
    precision=-16, rootdelay=0.000, rootdispersion=0.375, peer=0,
    refid=0.0.0.0, reftime=00000000.00000000 Thu, Feb 7 2036
    7:28:16.000,
    poll=4, clock=c97e9c6f.89437c56 Thu, Feb 15 2007 9:46:39.536,
    state=0,
    offset=0.000, frequency=0.000, jitter=0.015, stability=0.000


    Computer B:
    ntpq> pe
    remote refid st t when poll reach delay
    offset jitter
    ================================================== ====================
    emaddns1.endesa emball601.endes 2 u 46 64 7 1.807 69662.4
    15236.6
    emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    4000.00

    ntpq> rv
    status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    version="ntpd 4.1.1@1.786 Wed Oct 16 12:55:05 UTC 2002 (1)",
    processor="i686", system="Linux2.6.5-7.244-smp", leap=11, stratum=16,
    precision=-16, rootdelay=0.000, rootdispersion=2.790, peer=0,
    refid=0.0.0.0, reftime=00000000.00000000 Thu, Feb 7 2036
    7:28:16.000,
    poll=4, clock=c97e98d4.1fa14cec Thu, Feb 15 2007 9:31:16.123,
    state=0,
    offset=0.000, frequency=0.000, jitter=0.015, stability=0.000

    Can someone help me?

    Sorry my English.

    Thanks


  2. Re: No syncro

    Juan C. Gonzalez wrote:
    > Hello
    >
    > I have a two computers, A and B. Version xntp 4.1.1. I have too a
    > server ntp (S).
    >
    > My problem is the following one:
    >
    > Computer A, run xntpd start, connect with server (S), update the hour
    > and It works correctly.
    >
    > Computer B, run xntpd start, connect with server (S), It updates the
    > hour but alone this time.
    >
    > Computer A:
    >
    > ntpq> pe
    > remote refid st t when poll reach delay
    > offset jitter
    > ================================================== ====================
    > emaddns1.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    > 4000.00
    > emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    > 4000.00
    >
    > ntpq> rv
    > status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    > version="ntpd 4.1.1@1.786 Wed Oct 16 12:55:05 UTC 2002 (1)",
    > processor="i686", system="Linux2.4.21-278-default", leap=11,
    > stratum=16,
    > precision=-16, rootdelay=0.000, rootdispersion=0.375, peer=0,
    > refid=0.0.0.0, reftime=00000000.00000000 Thu, Feb 7 2036
    > 7:28:16.000,
    > poll=4, clock=c97e9c6f.89437c56 Thu, Feb 15 2007 9:46:39.536,
    > state=0,
    > offset=0.000, frequency=0.000, jitter=0.015, stability=0.000
    >
    >
    > Computer B:
    > ntpq> pe
    > remote refid st t when poll reach delay
    > offset jitter
    > ================================================== ====================
    > emaddns1.endesa emball601.endes 2 u 46 64 7 1.807 69662.4
    > 15236.6
    > emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000
    > 4000.00
    >
    > ntpq> rv
    > status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    > version="ntpd 4.1.1@1.786 Wed Oct 16 12:55:05 UTC 2002 (1)",
    > processor="i686", system="Linux2.6.5-7.244-smp", leap=11, stratum=16,
    > precision=-16, rootdelay=0.000, rootdispersion=2.790, peer=0,
    > refid=0.0.0.0, reftime=00000000.00000000 Thu, Feb 7 2036
    > 7:28:16.000,
    > poll=4, clock=c97e98d4.1fa14cec Thu, Feb 15 2007 9:31:16.123,
    > state=0,
    > offset=0.000, frequency=0.000, jitter=0.015, stability=0.000
    >
    > Can someone help me?
    >
    > Sorry my English.
    >
    > Thanks
    >


    69662.4 seconds exceeds the panic threshold. ntpd decides that the
    situation is hopeless and gives up!

    Your clock appears to be off by almost 20 hours!! Try starting ntpd
    with the -g switch! That will cause it to set the time unconditionally
    when it starts. If, for any reason, you can't use the -g switch, use
    ntpdate to set the clock. If you can't use ntpdate, set it by hand!




  3. Re: No syncro

    On 2007-02-15, Juan C. Gonzalez wrote:

    > I have a two computers, A and B. Version xntp 4.1.1. I have too a
    > server ntp (S).


    The current version of NTP is 4.2.4 and it hasn't been known as xntp
    since version 3. But that's not germane to your problem.

    > Computer A, run xntpd start, connect with server (S), update the hour
    > and It works correctly.
    >
    >ntpq> pe
    >remote refid st t when poll reach delay offset jitter
    >================================================== ===============
    >emaddns1.endesa 0.0.0.0 16 u - 64 0 0.000 0.000 4000.00
    >emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000 4000.00


    This ntpd has not been able to poll those two remote time servers (note
    the 0 in the reach column).

    It is possible that you did not allow enough time to elapse after
    starting ntpd. If you don't see any change in the reach, delay, offset,
    and jitter columns after ~5 minutes you have something blocking the NTP
    packets between your ntpd and the remote time servers.

    > Computer B, run xntpd start, connect with server (S), It updates the
    > hour but alone this time.
    >
    >ntpq> pe
    >remote refid st t when poll reach delay offset jitter
    >================================================== ===============
    >emaddns1.endesa embal... 2 u 46 64 7 1.807 69662.4 15236.6
    >emaddns3.endesa 0.0.0.0 16 u - 64 0 0.000 0.000 4000.00


    Here, ntpd has polled 'emaddns1' 3 times and has determined that your
    clock off-set is 69662.4 ms, or 69.6 seconds.

    Again, you should allow some time to pass before using 'ntpq -p'.

    Ntpd will abort if it sees an offset greater than 1024 seconds (~ 17
    minutes). If your version of ntpd supports the '-g' option you may want
    to add it to the ntpd command-line. Doing so will allow ntpd to make one
    step (usually the first one) of more than 1024 seconds.

    If you can't use '-g' you will need to use ntpdate to preset your clock
    before ntpd is started.

    --
    Steve Kostecke
    NTP Public Services Project - http://ntp.isc.org/

  4. Re: No syncro

    Hello

    I have update ntpd, now I have version 4.2.4.

    I continue with the same problem:

    I put more information:

    #date
    Fri Feb 16 09:17:00 CET 2007

    #ntpdate emaddns1.endesa.es
    16 Feb 09:18:51 ntpdate[7929]: step time server 172.29.8.4 offset
    92.118018 sec

    # /usr/local/bin/ntpd -g

    # ps fax|grep ntpd
    7932 ? Ss 0:00 /usr/local/bin/ntpd -g

    # ntptrace
    localhost: stratum 16, offset -0.000074, synch distance 0.00046
    73.78.73.84: *Timeout*

    # ntpq -p
    remote refid st t when poll reach delay
    offset jitter
    ================================================== ============================
    endesa.endesa.e emball601.endes 2 u 51 64 1 1.011
    11914.6 0.002
    emaddns3.endesa .INIT. 16 u - 64 0 0.000
    0.000 0.000

    ntpq> rv
    status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    version="ntpd 4.2.4@1.1437 Wed Feb 14 08:29:41 UTC 2007 (1)",
    processor="i686", system="Linux/2.6.5-7.244-smp", leap=11, stratum=16,
    precision=-19, rootdelay=0.000, rootdispersion=1.110, peer=0,
    refid=INIT, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
    poll=6, clock=c97fe7c5.1e61672d Fri, Feb 16 2007 9:20:21.118,
    state=1,
    offset=0.000, frequency=0.000, jitter=0.002, noise=0.002,
    stability=0.000

    ntpq> as
    ind assID status conf reach auth condition last_event cnt
    ================================================== =========
    1 7034 9014 yes yes none reject reachable 1
    2 7035 8000 yes yes none reject

    # date
    Fri Feb 16 09:21:33 CET 2007
    (date real: 9:23)

    Thanks


  5. Re: No syncro

    On 2007-02-16, Juan C. Gonzalez wrote:

    > I have update ntpd, now I have version 4.2.4.




    > #ntpdate emaddns1.endesa.es
    > 16 Feb 09:18:51 ntpdate[7929]: step time server 172.29.8.4 offset
    > 92.118018 sec


    This means that the NTP port (123/UDP) is open between your ntpd and
    that remote time-server.

    > # /usr/local/bin/ntpd -g
    >
    > # ps fax|grep ntpd
    > 7932 ? Ss 0:00 /usr/local/bin/ntpd -g
    >
    > # ntptrace
    > localhost: stratum 16, offset -0.000074, synch distance 0.00046
    > 73.78.73.84: *Timeout*


    If you ran ntptrace immediately after starting ntpd you did it much too
    soon. Please allow your ntpd to run for about 10 or 15 minutes before
    posting data about it.

    > # ntpq -p
    > remote refid st t when poll reach delay
    > offset jitter
    >================================================== ============================
    > endesa.endesa.e emball601.endes 2 u 51 64 1 1.011
    > 11914.6 0.002
    > emaddns3.endesa .INIT. 16 u - 64 0 0.000
    > 0.000 0.000


    It looks like your ntpd is still within the first poll period. So you
    didn't wait long enough. Please allow your ntpd to run for about 10 or
    15 minutes before posting data about it.

    You should see the number in the reach column at the end of each poll
    period. And, after enough sucessful polls of your remote time servers,
    ntpd will select a sys_peer and mark it the the "*" tally code in left
    column.

    If you want to speed up ntpd's initial sync you must append 'iburst' to
    your server lines.

    > ntpq> rv
    > status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    > version="ntpd 4.2.4@1.1437 Wed Feb 14 08:29:41 UTC 2007 (1)",
    > processor="i686", system="Linux/2.6.5-7.244-smp", leap=11, stratum=16,
    > precision=-19, rootdelay=0.000, rootdispersion=1.110, peer=0,
    > refid=INIT, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
    > poll=6, clock=c97fe7c5.1e61672d Fri, Feb 16 2007 9:20:21.118,
    > state=1,
    > offset=0.000, frequency=0.000, jitter=0.002, noise=0.002,
    > stability=0.000


    You didn't wait long enough. Please allow your ntpd to run for about 10
    or 15 minutes before posting data about it.

    Remember, it takes time to determine the correct time.

    --
    Steve Kostecke
    NTP Public Services Project - http://ntp.isc.org/

  6. Re: No syncro

    Juan C. Gonzalez wrote:
    > Hello
    >
    > I have update ntpd, now I have version 4.2.4.
    >
    > I continue with the same problem:
    >
    > I put more information:
    >
    > #date
    > Fri Feb 16 09:17:00 CET 2007
    >
    > #ntpdate emaddns1.endesa.es
    > 16 Feb 09:18:51 ntpdate[7929]: step time server 172.29.8.4 offset
    > 92.118018 sec
    >
    > # /usr/local/bin/ntpd -g
    >
    > # ps fax|grep ntpd
    > 7932 ? Ss 0:00 /usr/local/bin/ntpd -g
    >
    > # ntptrace
    > localhost: stratum 16, offset -0.000074, synch distance 0.00046
    > 73.78.73.84: *Timeout*
    >
    > # ntpq -p
    > remote refid st t when poll reach delay
    > offset jitter
    > ================================================== ============================
    > endesa.endesa.e emball601.endes 2 u 51 64 1 1.011
    > 11914.6 0.002
    > emaddns3.endesa .INIT. 16 u - 64 0 0.000
    > 0.000 0.000



    You should wait longer after starting ntpd to display the ntpq -p
    banner! Try waiting at least ten minutes. Thirty minutes would be
    better! You have ONE response from endesa.endesa.e..... and NO
    responses from emaddns3.endsa..... Ntpd needs a minimum of five
    responses before it can start setting the clock.

    Your second server is initializing and is not ready to serve time.

    Please note that two servers is the worst possible configuration!


  7. Re: No syncro

    In article <1171614276.639806.269320@h3g2000cwc.googlegroups.c om>,
    Juan C. Gonzalez wrote:

    > #ntpdate emaddns1.endesa.es
    > 16 Feb 09:18:51 ntpdate[7929]: step time server 172.29.8.4 offset
    > 92.118018 sec


    > endesa.endesa.e emball601.endes 2 u 51 64 1 1.011
    > 11914.6 0.002


    > poll=6, clock=c97fe7c5.1e61672d Fri, Feb 16 2007 9:20:21.118,


    You seem to have lost almost 12 seconds (not, of course 12 thousand - you
    should completely ignore the response that suggested that these values
    are seconds) in only a few minutes. As well as making sure that allow
    ntpd to stabilise before trying to diagnose, you should fix this problem
    before running ntpd. ntpd cannot cope with more than somewhat under 30ms
    per minute. 12 seconds in about that time is unsupportable.

    (Alternatively you have time servers that don't agree with each other,
    which means one or all of them are broken. The normal way of achieving this
    is to make them all agree with UTC.)

  8. Re: No syncro

    Here you have more data.

    # date
    Mon Feb 19 14:04:01 CET 2007

    ntpdate emaddns1.endesa.es
    19 Feb 16:08:09 ntpdate[12750]: step time server 172.29.8.4 offset
    7431.142532 sec

    /etc/init.d/xntp start
    Try to get initial date and time via NTP from emaddns1.endesa.es
    emaddns3.endesa.es done
    Starting network time protocol daemon
    (NTPD)
    done

    date
    Mon Feb 19 16:14:57 CET 2007 (time real 16:20)

    ntptrace
    localhost: stratum 16, offset 0.000000, synch distance 0.006060

    ntpq -p
    remote refid st t when poll reach delay
    offset jitter
    ================================================== ===========
    emaddns1.endesa 172.31.147.70 2 u 25 64 177 1.036 298656.
    195271.
    emaddns3.endesa .INIT. 16 u - 64 0 0.000
    0.000 0.000

    ntpq> rv
    assID=0 status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    version="ntpd 4.2.4@1.1437 Wed Feb 14 08:29:41 UTC 2007 (1)",
    processor="i686", system="Linux/2.6.5-7.244-smp", leap=11, stratum=16,
    precision=-19, rootdelay=0.000, rootdispersion=6.420, peer=0,
    refid=INIT, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
    poll=6, clock=c9843da1.45e404a4 Mon, Feb 19 2007 16:15:45.273,
    state=1,
    offset=0.000, frequency=0.000, jitter=0.002, noise=0.002,
    stability=0.000

    ntpq> peers
    remote refid st t when poll reach delay
    offset jitter
    ================================================== ============================
    emaddns1.endesa 172.31.147.70 2 u 37 64 377 1.010 399316.
    226652.
    emaddns3.endesa .INIT. 16 u - 64 0 0.000
    0.000 0.000

    Why the computer does A work correctly and the computer B not?

    Thanks


  9. Re: No syncro

    On 19 feb, 17:52, "Richard B. gilbert" wrote:
    > Juan C. Gonzalez wrote:
    > > Here you have more data.

    >
    > > # date
    > > Mon Feb 19 14:04:01 CET 2007

    >
    > > ntpdate emaddns1.endesa.es
    > > 19 Feb 16:08:09 ntpdate[12750]: step time server 172.29.8.4 offset
    > > 7431.142532 sec

    >
    > > /etc/init.d/xntp start
    > > Try to get initial date and time via NTP from emaddns1.endesa.es
    > > emaddns3.endesa.es done
    > > Starting network time protocol daemon
    > > (NTPD)
    > > done

    >
    > > date
    > > Mon Feb 19 16:14:57 CET 2007 (time real 16:20)

    >
    > > ntptrace
    > > localhost: stratum 16, offset 0.000000, synch distance 0.006060

    >
    > > ntpq -p
    > > remote refid st t when poll reach delay
    > > offset jitter
    > > ================================================== ===========
    > > emaddns1.endesa 172.31.147.70 2 u 25 64 177 1.036 298656.
    > > 195271.
    > > emaddns3.endesa .INIT. 16 u - 64 0 0.000
    > > 0.000 0.000

    >
    > > ntpq> rv
    > > assID=0 status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    > > version="ntpd 4...@1.1437 Wed Feb 14 08:29:41 UTC 2007 (1)",
    > > processor="i686", system="Linux/2.6.5-7.244-smp", leap=11, stratum=16,
    > > precision=-19, rootdelay=0.000, rootdispersion=6.420, peer=0,
    > > refid=INIT, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
    > > poll=6, clock=c9843da1.45e404a4 Mon, Feb 19 2007 16:15:45.273,
    > > state=1,
    > > offset=0.000, frequency=0.000, jitter=0.002, noise=0.002,
    > > stability=0.000

    >
    >
    >
    > Your time is not close enough to being correct that NTP can correct it.
    > I'm drawing a blank at the moment on whether that figure of 298656. is
    > milliseconds or seconds. In either case it is HUGE!!!!
    >
    > SET YOUR CLOCK TO THE CORRECT TIME BEFORE STARTING NTPD!!!!!!!!!!!!!!!
    > !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!
    > Alternatively you can use the -g option to set the clock at startup but
    > GET THAT CLOCK SET!!!!- Ocultar texto de la cita -
    >
    > - Mostrar texto de la cita -


    When I run ntpd the clock it puts in hour correctly. My question is:
    why then does not it continue supporting the hour?


  10. Re: No syncro

    Juan,

    If you are successfully running ntpdate and that command is setting your
    clock and if you immediately run ntpd and the first 'ntpq -p' shows that
    your clock is already 3 seconds (or whatever) off, and a minute or so later
    it is 4 seconds (or whatever) off, you have a Problem on that machine.

    Please see:

    http://ntp.isc.org/Support/DesigningYourNTPNetwork

    http://ntp.isc.org/Support/StartingNTP4

    http://ntp.isc.org/Support/TroubleshootingNTP

    H

  11. Re: No syncro

    Juan C. Gonzalez wrote:
    > On 19 feb, 17:52, "Richard B. gilbert" wrote:
    >
    >>Juan C. Gonzalez wrote:
    >>
    >>>Here you have more data.

    >>
    >>># date
    >>>Mon Feb 19 14:04:01 CET 2007

    >>
    >>>ntpdate emaddns1.endesa.es
    >>>19 Feb 16:08:09 ntpdate[12750]: step time server 172.29.8.4 offset
    >>>7431.142532 sec

    >>
    >>>/etc/init.d/xntp start
    >>>Try to get initial date and time via NTP from emaddns1.endesa.es
    >>>emaddns3.endesa.es done
    >>>Starting network time protocol daemon
    >>>(NTPD)
    >>>done

    >>
    >>>date
    >>>Mon Feb 19 16:14:57 CET 2007 (time real 16:20)

    >>
    >>>ntptrace
    >>>localhost: stratum 16, offset 0.000000, synch distance 0.006060

    >>
    >>>ntpq -p
    >>> remote refid st t when poll reach delay
    >>>offset jitter
    >>>================================================== ===========
    >>> emaddns1.endesa 172.31.147.70 2 u 25 64 177 1.036 298656.
    >>>195271.
    >>> emaddns3.endesa .INIT. 16 u - 64 0 0.000
    >>>0.000 0.000

    >>
    >>>ntpq> rv
    >>>assID=0 status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
    >>>version="ntpd 4...@1.1437 Wed Feb 14 08:29:41 UTC 2007 (1)",
    >>>processor="i686", system="Linux/2.6.5-7.244-smp", leap=11, stratum=16,
    >>>precision=-19, rootdelay=0.000, rootdispersion=6.420, peer=0,
    >>>refid=INIT, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
    >>>poll=6, clock=c9843da1.45e404a4 Mon, Feb 19 2007 16:15:45.273,
    >>>state=1,
    >>>offset=0.000, frequency=0.000, jitter=0.002, noise=0.002,
    >>>stability=0.000

    >>
    >>
    >>
    >>Your time is not close enough to being correct that NTP can correct it.
    >>I'm drawing a blank at the moment on whether that figure of 298656. is
    >>milliseconds or seconds. In either case it is HUGE!!!!
    >>
    >>SET YOUR CLOCK TO THE CORRECT TIME BEFORE STARTING NTPD!!!!!!!!!!!!!!!
    >>!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!
    >>Alternatively you can use the -g option to set the clock at startup but
    >>GET THAT CLOCK SET!!!!- Ocultar texto de la cita -
    >>
    >>- Mostrar texto de la cita -

    >
    >
    > When I run ntpd the clock it puts in hour correctly. My question is:
    > why then does not it continue supporting the hour?
    >


    Your ntpq -p banner shows that ntpd has been running for a very short
    time, less than ten minutes! If the clock was initially set correctly,
    it is now off by at least 69 seconds (assuming that the offset figure is
    in milliseconds). The following possibilities occur to me:
    a. The clock was not set correctly on startup!
    b. The local clock is drifting at a rate sufficient to produce that
    offset in less than ten minutes. If this is the case, you probably have
    a hardware problem!


  12. Re: No syncro

    OK, this time I read the 'ntpq -p' output as being 298 seconds off.

    H

  13. Re: No syncro

    I havent been able to get my ntp server to function successfully for days, and it always displayed:
    reftime=00000000.00000000 Thu, Feb 7 2036

    I followed Harlan's advice and checked out the NTP website:
    http://support.ntp.org/bin/view/Support/ConfiguringNTP

    As mentioned on their website, it's best practice to include at least 4 ntp servers in the ntp.conf file:

    ***
    This is the minimum configuration for an NTP Client which just needs "good enough" time.

    driftfile /var/lib/ntp/ntp.drift
    server 0.pool.ntp.org iburst
    server 1.pool.ntp.org iburst
    server 2.pool.ntp.org iburst
    server 3.pool.ntp.org iburst
    ***

    After updating the ntp.conf on my ntp server as shown above, everything started working as expected.

    Thanks
    K

+ Reply to Thread