NTP stratum problem - VMS

This is a discussion on NTP stratum problem - VMS ; I have set up two Alphas as timeservers for our other 20 Alphas. They both use the the same copy of TCPIP$NTP.CONF: server ntp1.sth.netnod.se server ntp2.sth.netnod.se server 2.se.pool.ntp.org server 2.europe.pool.ntp.org server 3.europe.pool.ntp.org server 127.127.1.0 fudge 127.127.1.0 stratum 8 I also ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: NTP stratum problem

  1. NTP stratum problem

    I have set up two Alphas as timeservers for our other 20 Alphas.
    They both use the the same copy of TCPIP$NTP.CONF:

    server ntp1.sth.netnod.se
    server ntp2.sth.netnod.se
    server 2.se.pool.ntp.org
    server 2.europe.pool.ntp.org
    server 3.europe.pool.ntp.org

    server 127.127.1.0
    fudge 127.127.1.0 stratum 8

    I also added that they use each other ass peers

    HUVUD is running TCP/IP Services Version V5.6 - ECO 2
    GLOBAL is running TCP/IP Services Version V5.4 - ECO 2

    Node HUVUD which was the first NTP server started looks fine:

    HUVUD> ntpdc -p
    remote local st poll reach delay offset disp
    ================================================== =====================
    *ntp1.sth.netnod 194.14.159.55 1 1024 377 0.02243 0.005199
    0.01578
    =ntp2.sth.netnod 194.14.159.55 1 1024 377 0.00876 0.006664
    0.01581
    =irc-cobra.unizh 194.14.159.55 4 1024 377 0.11130 -0.031395
    0.01579
    =zit-net1.uni-pa 194.14.159.55 1 1024 377 0.03905 -0.005733
    0.01581
    =LOCAL(0) 127.0.0.1 8 64 377 0.00000 0.000000
    0.00287
    =212.247.117.169 194.14.159.55 2 1024 377 0.04271 0.015330
    0.01579
    +global.huv.daga 194.14.159.55 3 64 376 0.00098 -1.054923
    0.00308

    Node GLOBAL assign the stratum 1 servers a stratum 16 value:
    GLOBAL_GLOB> ntpdc -p
    remote local st poll reach delay offset disp
    ================================================== =====================
    =LOCAL(0) 127.0.0.1 8 64 377 0.00000 0.000000
    0.00191
    =ntp1.sth.netnod 194.14.159.42 16 128 0 0.00000 0.000000
    0.00000
    =ntp2.sth.netnod 194.14.159.42 16 128 0 0.00000 0.000000
    0.00000
    *HUVUD 194.14.159.42 2 64 377 0.00670 1.039268
    0.00302
    =voodoo.banbook. 194.14.159.42 16 128 0 0.00000 0.000000
    0.00000
    =public-timehost 194.14.159.42 16 128 0 0.00000 0.000000
    0.00000
    =ran.r3blog.nl 194.14.159.42 16 128 0 0.00000 0.000000
    0.00000

    Tried to set up another node with TCP/IP Version V5.5 - ECO 2
    but it gave the same result as GLOBAL.

    All nodes are on the same LAN and the same network.

    Why is node GLOBAL not accessing the stratum 1 servers?

    GLOBAL_GLOB> ntpdate -q ntp1.sth.netnod.se
    Server ntp1.sth.netnod.se (192.36.144.22), stratum 0
    offset +0.0000000, delay +0.0000000
    No server suitable for synchronization found from those provided.

  2. Re: NTP stratum problem

    jan.andersson@axfood.se wrote:

    > Why is node GLOBAL not accessing the stratum 1 servers?



    Not sure. However:

    After a power failure, my Mac will refuse to sync with my alpha for some
    time. Eventually it does sync. My take on this is that the alpha take
    some time before it decides to connect to the remote servers and get not
    only its time but also its stratum, and until it happens, I suspect the
    alpha remains are stratum 16 and thus unusable as a server.

    In your config files, do you really need the server 127.127.1.0 and the
    fudge for it ? This is what may be confusing it.

    The ntp software acts as a server in any case. There is no need to
    declare it as a server. It will get its stratum from whatever server it
    has connected to, and then your clients on the lan will get their time
    from your alpha.

  3. Re: NTP stratum problem

    On Oct 30, 8:13*am, jan.anders...@axfood.se wrote:
    > I have set up two Alphas as timeservers for our other 20 Alphas.
    > They both use the the same copy of TCPIP$NTP.CONF:
    >
    > server ntp1.sth.netnod.se
    > server ntp2.sth.netnod.se
    > server 2.se.pool.ntp.org
    > server 2.europe.pool.ntp.org
    > server 3.europe.pool.ntp.org
    >
    > server 127.127.1.0
    > fudge 127.127.1.0 stratum 8
    >
    > I also added that they use each other ass peers
    >
    > HUVUD is running *TCP/IP Services Version V5.6 - ECO 2
    > GLOBAL is running TCP/IP Services Version V5.4 - ECO 2
    >
    > Node HUVUD which was the first NTP server started looks fine:
    >
    > HUVUD> ntpdc -p
    > * * *remote * * * * * local * * *st poll reach *delay * offset * *disp
    > ================================================== =====================
    > *ntp1.sth.netnod 194.14.159.55 * *1 1024 *377 0.02243 *0.005199
    > 0.01578
    > =ntp2.sth.netnod 194.14.159.55 * *1 1024 *377 0.00876 *0.006664
    > 0.01581
    > =irc-cobra.unizh 194.14.159.55 * *4 1024 *377 0.11130 -0.031395
    > 0.01579
    > =zit-net1.uni-pa 194.14.159.55 * *1 1024 *377 0.03905 -0.005733
    > 0.01581
    > =LOCAL(0) * * * *127.0.0.1 * * * *8 * 64 *377 0.00000 *0.000000
    > 0.00287
    > =212.247.117.169 194.14.159.55 * *2 1024 *377 0.04271 *0.015330
    > 0.01579
    > +global.huv.daga 194.14.159.55 * *3 * 64 *376 0.00098 -1.054923
    > 0.00308
    >
    > Node GLOBAL assign the stratum 1 servers a stratum 16 value:
    > GLOBAL_GLOB> ntpdc -p
    > * * *remote * * * * * local * * *st poll reach *delay * offset * *disp
    > ================================================== =====================
    > =LOCAL(0) * * * *127.0.0.1 * * * *8 * 64 *377 0.00000 *0.000000
    > 0.00191
    > =ntp1.sth.netnod 194.14.159.42 * 16 *128 * *0 0.00000 *0.000000
    > 0.00000
    > =ntp2.sth.netnod 194.14.159.42 * 16 *128 * *0 0.00000 *0.000000
    > 0.00000
    > *HUVUD * * * * * 194.14.159.42 * *2 * 64 *377 0.00670 *1.039268
    > 0.00302
    > =voodoo.banbook. 194.14.159.42 * 16 *128 * *0 0.00000 *0.000000
    > 0.00000
    > =public-timehost 194.14.159.42 * 16 *128 * *0 0.00000 *0.000000
    > 0.00000
    > =ran.r3blog.nl * 194.14.159.42 * 16 *128 * *0 0.00000 *0.000000
    > 0.00000
    >
    > Tried to set up another node with TCP/IP Version V5.5 - ECO 2
    > but it gave the same result *as GLOBAL.
    >
    > All nodes are on the same LAN and the same network.
    >
    > Why is node GLOBAL not accessing the stratum 1 servers?
    >
    > GLOBAL_GLOB> ntpdate -q ntp1.sth.netnod.se
    > Server ntp1.sth.netnod.se (192.36.144.22), stratum 0
    > offset +0.0000000, delay +0.0000000
    > No server suitable for synchronization found from those provided.


    Legal strata values are in the range of 1 to 15. I'm not familiar with
    HP's TCPIP implementation of this but it would appear that the strata
    1 servers are not reachable by node GLOBAL. The reach value is an 8
    bit bitmask indicating the success of the past 8 polls. Your reach is
    0 for each of these servers so you've not been successful in polling
    them (numerous other values are also 0 indicating the failure of the
    NTP polls. Since the strata is stated to be 16 I would take this as an
    indication that the system has never been able to successfully poll
    the servers so as to learn their strata. I'd look for an intervening
    firewall or some other blockage or denial of service...

  4. Re: NTP stratum problem

    On Oct 30, 1:50 pm, Jim wrote:
    > On Oct 30, 8:13 am, jan.anders...@axfood.se wrote:
    >
    >
    >
    > > I have set up two Alphas as timeservers for our other 20 Alphas.
    > > They both use the the same copy of TCPIP$NTP.CONF:

    >
    > > server ntp1.sth.netnod.se
    > > server ntp2.sth.netnod.se
    > > server 2.se.pool.ntp.org
    > > server 2.europe.pool.ntp.org
    > > server 3.europe.pool.ntp.org

    >
    > > server 127.127.1.0
    > > fudge 127.127.1.0 stratum 8

    >
    > > I also added that they use each other ass peers

    >
    > > HUVUD is running TCP/IP Services Version V5.6 - ECO 2
    > > GLOBAL is running TCP/IP Services Version V5.4 - ECO 2

    >
    > > Node HUVUD which was the first NTP server started looks fine:

    >
    > > HUVUD> ntpdc -p
    > > remote local st poll reach delay offset disp
    > > ================================================== =====================
    > > *ntp1.sth.netnod 194.14.159.55 1 1024 377 0.02243 0.005199
    > > 0.01578
    > > =ntp2.sth.netnod 194.14.159.55 1 1024 377 0.00876 0.006664
    > > 0.01581
    > > =irc-cobra.unizh 194.14.159.55 4 1024 377 0.11130 -0.031395
    > > 0.01579
    > > =zit-net1.uni-pa 194.14.159.55 1 1024 377 0.03905 -0.005733
    > > 0.01581
    > > =LOCAL(0) 127.0.0.1 8 64 377 0.00000 0.000000
    > > 0.00287
    > > =212.247.117.169 194.14.159.55 2 1024 377 0.04271 0.015330
    > > 0.01579
    > > +global.huv.daga 194.14.159.55 3 64 376 0.00098 -1.054923
    > > 0.00308

    >
    > > Node GLOBAL assign the stratum 1 servers a stratum 16 value:
    > > GLOBAL_GLOB> ntpdc -p
    > > remote local st poll reach delay offset disp
    > > ================================================== =====================
    > > =LOCAL(0) 127.0.0.1 8 64 377 0.00000 0.000000
    > > 0.00191
    > > =ntp1.sth.netnod 194.14.159.42 16 128 0 0.00000 0.000000
    > > 0.00000
    > > =ntp2.sth.netnod 194.14.159.42 16 128 0 0.00000 0.000000
    > > 0.00000
    > > *HUVUD 194.14.159.42 2 64 377 0.00670 1.039268
    > > 0.00302
    > > =voodoo.banbook. 194.14.159.42 16 128 0 0.00000 0.000000
    > > 0.00000
    > > =public-timehost 194.14.159.42 16 128 0 0.00000 0.000000
    > > 0.00000
    > > =ran.r3blog.nl 194.14.159.42 16 128 0 0.00000 0.000000
    > > 0.00000

    >
    > > Tried to set up another node with TCP/IP Version V5.5 - ECO 2
    > > but it gave the same result as GLOBAL.

    >
    > > All nodes are on the same LAN and the same network.

    >
    > > Why is node GLOBAL not accessing the stratum 1 servers?

    >
    > > GLOBAL_GLOB> ntpdate -q ntp1.sth.netnod.se
    > > Server ntp1.sth.netnod.se (192.36.144.22), stratum 0
    > > offset +0.0000000, delay +0.0000000
    > > No server suitable for synchronization found from those provided.

    >
    > Legal strata values are in the range of 1 to 15. I'm not familiar with
    > HP's TCPIP implementation of this but it would appear that the strata
    > 1 servers are not reachable by node GLOBAL. The reach value is an 8
    > bit bitmask indicating the success of the past 8 polls. Your reach is
    > 0 for each of these servers so you've not been successful in polling
    > them (numerous other values are also 0 indicating the failure of the
    > NTP polls. Since the strata is stated to be 16 I would take this as an
    > indication that the system has never been able to successfully poll
    > the servers so as to learn their strata. I'd look for an intervening
    > firewall or some other blockage or denial of service...


    Explanation: Port 123 was blocked in the firewall.
    It's working now.
    Thanks all.

+ Reply to Thread