[SAC] [OSGeo] #724: time on lists.osgeo.org in future

#724: time on lists.osgeo.org in future
---------------------------+------------------------------------------------
Reporter: neteler | Owner: sac@…
     Type: task | Status: new
Priority: critical | Milestone:
Component: Systems Admin | Keywords:
---------------------------+------------------------------------------------
I got notified that the time is in future on the list server.

Current real time:
date -u
Thu May 19 10:48:19 UTC 2011

lists.osgeo.org:
root@peer1_1:neteler # date -u
Fri May 20 07:42:26 UTC 2011

I have restarted ntpd but I am not sure that this helps.

--
Ticket URL: <http://trac.osgeo.org/osgeo/ticket/724&gt;
OSGeo <http://www.osgeo.org/&gt;
OSGeo committee and general foundation issue tracker.

#724: time on lists.osgeo.org in future
---------------------------+------------------------------------------------
Reporter: neteler | Owner: sac@…
     Type: task | Status: new
Priority: critical | Milestone:
Component: Systems Admin | Keywords:
---------------------------+------------------------------------------------

Comment(by warmerdam):

I have moved it forward three hours to be within a minute or two of
correct. I'm not sure why it is/was out of whack.

Perhaps we should keep an eye on it.

Note lists.osgeo.org is osgeo1, the same system used for Drupal and LDAP.

--
Ticket URL: <http://trac.osgeo.org/osgeo/ticket/724#comment:1&gt;
OSGeo <http://www.osgeo.org/&gt;
OSGeo committee and general foundation issue tracker.

On Thu, May 19, 2011 at 01:13:43PM -0000, OSGeo wrote:

I have moved it forward three hours to be within a minute or two of
correct. I'm not sure why it is/was out of whack.

root@peer1_1:~ # date -u
Fri May 20 13:52:58 UTC 2011
^^^^^^^^^^ !!!
root@peer1_1:~ # ntp
root@peer1_1:~ # /etc/init.d/ntpd stop
Shutting down ntpd: [ OK ]
root@peer1_1:~ # ntpdate time.peer1.com
19 May 09:52:03 ntpdate[6291]: step time server 66.132.181.236 offset -86473.375311 sec
root@peer1_1:~ # date -u
Thu May 19 13:52:06 UTC 2011
root@peer1_1:~ # /etc/init.d/ntpd start
Starting ntpd: [ OK ]

Cheers,
  Martin.
--
Unix _IS_ user friendly - it's just selective about who its friends are !
--------------------------------------------------------------------------

#724: time on lists.osgeo.org in future
---------------------------+------------------------------------------------
Reporter: neteler | Owner: sac@…
     Type: task | Status: closed
Priority: critical | Milestone:
Component: Systems Admin | Resolution: fixed
Keywords: |
---------------------------+------------------------------------------------
Changes (by martin):

  * status: new => closed
  * resolution: => fixed

Comment:

Fixed by synchronizing with NTP servers.

--
Ticket URL: <http://trac.osgeo.org/osgeo/ticket/724#comment:2&gt;
OSGeo <http://www.osgeo.org/&gt;
OSGeo committee and general foundation issue tracker.

On Thu, May 19, 2011 at 3:53 PM, Martin Spott <Martin.Spott@mgras.net> wrote:

On Thu, May 19, 2011 at 01:13:43PM -0000, OSGeo wrote:

I have moved it forward three hours to be within a minute or two of
correct. I'm not sure why it is/was out of whack.

root@peer1_1:~ # date -u
Fri May 20 13:52:58 UTC 2011
^^^^^^^^^^ !!!

Yes, 1 day.

root@peer1_1:~ # ntp
root@peer1_1:~ # /etc/init.d/ntpd stop
Shutting down ntpd: [ OK ]
root@peer1_1:~ # ntpdate time.peer1.com
19 May 09:52:03 ntpdate[6291]: step time server 66.132.181.236 offset -86473.375311 sec
root@peer1_1:~ # date -u
Thu May 19 13:52:06 UTC 2011
root@peer1_1:~ # /etc/init.d/ntpd start
Starting ntpd: [ OK ]

For the record: I tried the same but did "restart" of the daemon
which didn't have any effect. Stop/starts seems to be better,

Markus

On Thu, May 19, 2011 at 06:52:16PM +0200, Markus Neteler wrote:

For the record: I tried the same but did "restart" of the daemon
which didn't have any effect. Stop/starts seems to be better,

Yup. BTW, the kernel-PLL appears to be totally out of tune. I've
already tried to apply a few countermeasures (regarding the NTP drift
file and '/etc/adjtime'), but it might take a few days until it's back
into reasonable shape.

Cheers,
  Martin.
--
Unix _IS_ user friendly - it's just selective about who its friends are !
--------------------------------------------------------------------------

please double check `ntpq -p` as well after a while and ensure it
is talking to the time servers you think it is.

(at least Ubuntu 10.04 LTS pushed in a ntp change in the last
versions that silently ignores servers in ntp.conf if the DHCP
host has supplied (or begins to supply) some.. argh!)

Hamish

On Thu, May 19, 2011 at 05:01:50PM -0700, Hamish wrote:

please double check `ntpq -p` as well after a while and ensure it
is talking to the time servers you think it is.

  # ~> while true; do ntpdc -p; sleep 3; done

  :-)

(at least Ubuntu 10.04 LTS pushed in a ntp change in the last
versions that silently ignores servers in ntp.conf if the DHCP
host has supplied (or begins to supply) some.. argh!)

In general I'd welcome if distributions honour the DHCP "ntp-servers"
option, as far as I can tell, very, very few are doing this, if any.

The state on 'osgeo1' is steadily getting better, still not synced to
any of the NTP servers, but the growth of the offset has slowed down
significantly.

Cheers,
  Martin.
--
Unix _IS_ user friendly - it's just selective about who its friends are !
--------------------------------------------------------------------------