Jan Ceuleers wrote:
Done; I have sent you the resulting file off-list (due to its size).
I believe that the principal symptom in there is the message "Found
interface index -1 for address x.y.z.t".
I have now created a bug report (#631) and have attached the debug log
to it.
Now leaving o
Steve,
Steve Kostecke wrote:
Start ntpd in debug mode (w/ -d or -dd) before you restart your PPP
session and redirect the debugging messages to a file.
Done; I have sent you the resulting file off-list (due to its size).
I believe that the principal symptom in there is the message "Found
in
Steve Kostecke wrote:
Please can you let me know which data I should capture the next time I
see this problem?
Start ntpd in debug mode (w/ -d or -dd) before you restart your PPP
session and redirect the debugging messages to a file.
Will do and will report back when I have something to repor
[EMAIL PROTECTED] wrote:
Hello,
I am trying to compile ntp-4.2.2 on a Solaris 8 sparc server and
keep getting the following error in configure:
checking for perl... /usr/local/bin/perl
configure: error: How do we create an unsigned long constant?
I tried seraching for this error in this g
I am having stability problems with the newly-released 4.2.2 which I am
presently unable to be sufficiently specific about to be able to open a
bug report. I am therefore seeking guidance on how to document this problem.
I have found it necessary to drop and re-establish the PPP session to my
Well, nevermind. I forgot to replace the Sun grep with the GNU grep
(like I did on the other server) and it looked like configure was
trying to do an egrep (which you can't do with Sun's grep) as part of
it's check routine. It's always the little things.
-Chris Noyes
[EMAIL PROTECTED] wro
Hello,
I am trying to compile ntp-4.2.2 on a Solaris 8 sparc server and
keep getting the following error in configure:
checking for perl... /usr/local/bin/perl
configure: error: How do we create an unsigned long constant?
I tried seraching for this error in this group and also throught
Googl
>I would choose to find out why the on-site server "goes a bit askew".
> Machines running Linux or Windows have been known to lose clock
>interrupts when busy. If your stratum two server is always slow when
>it's off and "steps ahead" and is running Linux or Windows that is
>probably the cau
Richard B. Gilbert <[EMAIL PROTECTED]> wrote:
>Free BSD should be okay. Knowing that it's BSD, I'd look elsewhere for
>the problem.
A colleague recently upgraded one of our NTP servers to FreeBSD 6, and
something in the new kernel upset ntpd. It started drifting by several
milliseconds per hour.
James Ray wrote:
On Mon, 12 Jun 2006 07:54:26 -0400, Richard B. Gilbert wrote:
James Ray wrote:
All,
Here is what we are after doing. Our networks team provides us with a
stratum 2 NTP server to use as well as our ISP providing us with 3
stratum 1 NTP servers. Ideally I would like a set of
James Ray <[EMAIL PROTECTED]> wrote:
> would like to prefer the on-site server to the off-site server until it
> steps majorly out of line (say more than 0.8ms). At the moment there is
> constant jumping between the on-site and off-site servers if I have them
> both listed even with less than .8ms
On Mon, 12 Jun 2006 07:54:26 -0400, Richard B. Gilbert wrote:
> James Ray wrote:
>> All,
>> Here is what we are after doing. Our networks team provides us with a
>> stratum 2 NTP server to use as well as our ISP providing us with 3
>> stratum 1 NTP servers. Ideally I would like a set of 4 peers ru
James Ray wrote:
All,
Here is what we are after doing. Our networks team provides us with a
stratum 2 NTP server to use as well as our ISP providing us with 3 stratum
1 NTP servers. Ideally I would like a set of 4 peers run by me syncing
from the stratum 2 server provided on-site (since that gets
All,
Here is what we are after doing. Our networks team provides us with a
stratum 2 NTP server to use as well as our ISP providing us with 3 stratum
1 NTP servers. Ideally I would like a set of 4 peers run by me syncing
from the stratum 2 server provided on-site (since that gets it data from
the s
David,
David Woolley wrote:
> That's the point of SNTP, it doesn't specify how time offsets are handled.
> However, the rules you quote are, in my view, not at all satisfactory,
> and if w32time actually obeyed them it would make it even more unsuitable for
> feeding real NTP clients than it alrea
15 matches
Mail list logo
