| View previous topic :: View next topic |
| Author |
Message |
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
 Posted: Mon May 30, 2005 5:36 pm Post subject: [SOLVED] NTP Error |
 |
|
Hi,
AMD64 with 64-bit OS.
I installed NTP and it starts fine.
When I type ntpq -p I get the error: ntpq: read: Connection refused
What is wrong? Any idea?
As I recall I did follow the same steps as for other PC.
Thank you,
Spiro
Last edited by SAngeli on Wed Jun 08, 2005 2:38 pm; edited 1 time in total |
|
| Back to top |
|
 |
rutski89
Guru
Joined: 14 Mar 2005
Posts: 468
Location: United States N.Y.
|
| Posted: Mon May 30, 2005 5:43 pm Post subject: |
|
|
This is what got me synced with ntp on a fresh install just yesterday. | Code: | | ntpdate -b -u pool.ntp.org |
I hope this helps.
Peace
_________________
<< ^ | ~ >> |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Mon May 30, 2005 6:19 pm Post subject: |
|
|
I will try it tomorrow and let you know.
Thank you,
Spiro |
|
| Back to top |
|
|
Maedhros
Bodhisattva
Joined: 14 Apr 2004
Posts: 5511
Location: Durham, UK
|
| Posted: Mon May 30, 2005 6:54 pm Post subject: |
|
|
Moved from Installing Gentoo.
_________________
No-one's more important than the earthworm. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Tue May 31, 2005 7:11 pm Post subject: |
|
|
Hi,
no luck. I tried ntpdate -b -u pool.ntp.org which run fine.
Also ntpd loads fine at boot.
But still: ntpq: read: Connection refused
Any idea?
Thank you,
Spiro |
|
| Back to top |
|
|
christsong84
Veteran
Joined: 06 Apr 2003
Posts: 1002
Location: GMT-8 (Spokane)
|
| Posted: Tue May 31, 2005 7:15 pm Post subject: |
|
|
bad conncetion...almost sounds like a firewall issue...does it work if the firewall is down?
_________________
while(true) {self.input(sugar);}  |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Tue May 31, 2005 7:23 pm Post subject: |
|
|
I have two PC on the same network and they both run fine.
Does it have anything to do with IPv6 (although I have no idea of this)?
This is what I fould:
| Quote: | | The easiest command to verify that xntpd is still running is ntpq -p. This command will contact xntpd on the local host, and it will list all configured servers together with some health status. If xntpd is not running, the typical error message is ntpq: read: Connection refused. |
Could it be? But, I cannot locate /usr/sbin/xntpd? Perhaps I do not have it?
Spiro |
|
| Back to top |
|
|
larand54
l33t
Joined: 20 Feb 2004
Posts: 688
Location: Sweden
|
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Fri Jun 03, 2005 12:14 pm Post subject: |
|
|
I had no luck.
I wish to post some data so that you could see better:
from logs:
| Code: | May 29 14:29:19 (none) ntpd[7485]: ntpd 4.2.0@1.1161-r Sat May 28 17:23:49 CEST 2005 (1)
May 29 14:29:19 (none) ntpd[7485]: signal_no_reset: signal 13 had flags 4000000
May 29 14:29:19 (none) ntpd[7485]: precision = 1.000 usec
May 29 14:29:19 (none) ntpd[7485]: kernel time sync status 0040
May 29 14:29:19 (none) ntpd[7485]: configure: keyword "212.204.235.156" unknown, line ignored
May 29 14:29:19 (none) ntpd[7485]: configure: keyword "83.137.103.134" unknown, line ignored
May 29 14:29:19 (none) ntpd[7485]: configure: keyword "213.84.46.114" unknown, line ignored
28 May 16:39:44 ntpd[18167]: frequency error -512 PPM exceeds tolerance 500 PPM
28 May 16:40:47 ntpd[18167]: frequency error -512 PPM exceeds tolerance 500 PPM
28 May 16:42:56 ntpd[18167]: frequency error -504 PPM exceeds tolerance 500 PPM
28 May 17:10:59 ntpd[18167]: time reset +0.559925 s
29 May 14:29:19 ntpd[7485]: cap_set_proc() failed to drop root privileges: Operation not permitted
29 May 14:35:34 ntpd[7609]: cap_set_proc() failed to drop root privileges: Operation not permitted
29 May 14:37:06 ntpd[7605]: cap_set_proc() failed to drop root privileges: Operation not permitted
29 May 14:40:55 ntpd[8263]: cap_set_proc() failed to drop root privileges: Operation not permitted
29 May 22:35:37 ntpd[7601]: cap_set_proc() failed to drop root privileges: Operation not permitted
30 May 11:59:07 ntpd[7564]: cap_set_proc() failed to drop root privileges: Operation not permitted |
/etc/ntp.conf
| Code: | server pool.ntp.org iburst
#server 215 212.204.235.156 iburst
#server 216 83.137.103.134 iburst
#server 245 213.84.46.114 iburst
restrict 127.0.0.1 notrust nomodify
driftfile /var/lib/ntp/ntp.drift
logfile /var/log/ntpd.log |
/etc/conf.d/ntpd
| Code: | # Copyright 1999-2004 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
# $Header: /var/cvsroot/gentoo-x86/net-misc/ntp/files/ntpd.confd,v 1.15 2004/07/15 00:05:46 agriffis Exp $
# Options to pass to the ntpd process
# Most people should leave this line alone ...
# however, if you know what you're doing, feel free to tweak
NTPD_OPTS="-u ntp:ntp" |
/etc/conf.d/ntp-client
| Code: | NTPCLIENT_CMD="ntpdate"
NTPCLIENT_OPTS="-b pool.ntp.org"
NTPCLIENT_TIMEOUT=30
#NTPCLIENT_OPTS="-Q -b -u pool.ntp.org" |
When I turn the PC off, I get the error ntpd[7736] Failed to stop NTPD....
Please help.
Also, in rc.conf would it be better to set CLOCK="UTC" rather than "local" as I have it?
I was reading that by using UTC in conjunction with /etc/localtime it should report the same time as if I would use "local" Corrent (Y/n)?
Thank you,
Spiro |
|
| Back to top |
|
|
plut0
Apprentice
Joined: 21 Dec 2004
Posts: 271
|
| Posted: Fri Jun 03, 2005 12:54 pm Post subject: |
|
|
| The "Connection refused" error usually means your daemon is not running. Anyway, by your comment on the shutdown ntpd failing to stop thats what seems to be happening. |
|
| Back to top |
|
|
plut0
Apprentice
Joined: 21 Dec 2004
Posts: 271
|
| Posted: Fri Jun 03, 2005 12:57 pm Post subject: |
|
|
| I use localtime myself. Do you have the correct timezone set? Does date -u give the correct time? |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Fri Jun 03, 2005 1:10 pm Post subject: |
|
|
hi plut0,
here is what I get for dates:
As of now, I have "local" and this is what I get:
| Code: | # date -u
Fri Jun 3 13:07:05 UTC 2005
# date
Fri Jun 3 15:07:14 CEST 2005
|
My timezone is properly set.
So, I see two different times. What should I do?
My timezone is Italy and now it is 15:07
As for ntpd, at boot it starts properly.
Here is also my rc-status:
| Code: | local [ started ]
netmount [ started ]
domainname [ started ]
net.eth0 [ started ]
hotplug [ started ]
syslog-ng [ started ]
ntp-client [ started ]
ntpd [ started ]
numlock [ started ]
gpm [ started ]
splash [ started ]
sshd [ started ]
|
What should I do?
Please help, if you can.
Thanks,
Spiro
Last edited by SAngeli on Tue Jun 07, 2005 8:43 pm; edited 1 time in total |
|
| Back to top |
|
|
plut0
Apprentice
Joined: 21 Dec 2004
Posts: 271
|
| Posted: Fri Jun 03, 2005 1:34 pm Post subject: |
|
|
You see two different times because one is your time and the other is Universal Time. The UTC should be the same for everybody no matter where you are. Your times are correct.
I realize that ntpd is set to start at default run level but I don't think it is running. See if the process is running `ps aux | grep ntpd | grep -v grep` I suspect the frequency errors is what is causing it to crash. Can you post the output of `cat /var/lib/ntp/ntp.drift`? |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Fri Jun 03, 2005 1:39 pm Post subject: |
|
|
Hi,
Time: So, as all being correct, should I live it to local or to UTC?
As for ntpd, you are perfectly correct:
ps aux | grep ntpd | grep -v grep
did not return anything.
cat /var/lib/ntp/ntp.drift
-58.220
Spiro |
|
| Back to top |
|
|
plut0
Apprentice
Joined: 21 Dec 2004
Posts: 271
|
| Posted: Fri Jun 03, 2005 1:52 pm Post subject: |
|
|
| Code: | # Set CLOCK to "UTC" if your system clock is set to UTC (also known as
# Greenwich Mean Time). If your clock is set to the local time, then set CLOCK
# to "local". This setting is used by the /etc/init.d/clock script. |
Leave it as "local", it is easier to read. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Fri Jun 03, 2005 1:54 pm Post subject: |
|
|
ok for time.
please let me know for ntpd if you are aware of how to solve it.
Thanks,
Spiro |
|
| Back to top |
|
|
plut0
Apprentice
Joined: 21 Dec 2004
Posts: 271
|
| Posted: Fri Jun 03, 2005 1:57 pm Post subject: |
|
|
I wonder if the problem has to due to your latency to the server. What is your latency if you `ping pool.ntp.org`? If that is the problem you might have to search for a server closer to you.
For now, try appending "-x" in NTPD_OPTS in /etc/conf.d/ntpd. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Fri Jun 03, 2005 2:35 pm Post subject: |
|
|
| Code: | root # ping -c 5 pool.ntp.org
PING pool.ntp.org (62.94.26.10) 56(84) bytes of data.
64 bytes from ip-26-10.sn1.eutelia.it (62.94.26.10): icmp_seq=1 ttl=54 time=98.5 ms
64 bytes from ip-26-10.sn1.eutelia.it (62.94.26.10): icmp_seq=2 ttl=54 time=107 ms
64 bytes from ip-26-10.sn1.eutelia.it (62.94.26.10): icmp_seq=3 ttl=54 time=163 ms
64 bytes from ip-26-10.sn1.eutelia.it (62.94.26.10): icmp_seq=4 ttl=54 time=103 ms
64 bytes from ip-26-10.sn1.eutelia.it (62.94.26.10): icmp_seq=5 ttl=54 time=131 ms
|
Spiro |
|
| Back to top |
|
|
F.Ultra
Apprentice
Joined: 17 Mar 2004
Posts: 169
Location: Sweden
|
| Posted: Fri Jun 03, 2005 7:20 pm Post subject: |
|
|
| Code: | | failed to drop root privileges: Operation not permitted |
This means that the ntpd daemon cannot change from beeing root into the ntp:ntp user, and it does therefore quit. And that is why you cannot see it running and why the init.d script cannot kill it (since it is not running).
Perhaps you have no ntp user? Check whats in /etc/passwd there should be a line with "ntp:xx" in it. If everything else fails you can tell ntpd to run as root by changing | Code: | | NTPD_OPTS="-u ntp:ntp" |
into in /etc/conf.d/ntpd atleast to see if it works that way, it is though not recommended to run ntpd as root. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Sat Jun 04, 2005 9:07 pm Post subject: |
|
|
Hi,
finaly a right answer! 
This is what I have in this PC:
| Code: | ntp:x:123:123:added by portage for ntp:/dev/null:/bin/false
|
I tried to make the above changes to /etc/conf.d/ntpd and it works. Here is the output:
| Code: | root # ps aux | grep ntpd | grep -v grep
root 8923 0.0 0.5 12564 5272 ? SLs 10:23 0:00 /usr/bin/ntpd -p /var/run/ntpd.pid
root # ntpq -p
remote refid st t when poll reach delay offset jitter
==============================================================================
*ns3.dns.pciwest 132.163.4.103 2 u 36 64 1 227.995 -72.077 1.973 |
So, this tells me that I have an issue with ntp user. What to do now? Is there anything to do with permissions? Maybe, does it have anything to do with udev?
In stopping ntp services, is the correct sequence ntp-client first and then ntpd?
Thank you,
Spiro |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 27780
Location: 56N 3W
|
| Posted: Tue Jun 07, 2005 6:59 pm Post subject: |
|
|
SAngeli,
I read the thread but can't really add anything - I has problems with ntpd because the accumulater errors were toobe (I switch my PC off at night). I used ntp-date which is a bit of a blunt instument.
You may need to run that at startup before ntpd runs, to fix any gross errors.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Tue Jun 07, 2005 7:50 pm Post subject: |
|
|
Hi NeddySeagoon,
thank you so much for taking your time and reading about my issue.
could you explani better this part:
| Quote: | | because the accumulater errors were toobe |
What is ntp-date?
Thanks,
Spiro |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 27780
Location: 56N 3W
|
| Posted: Tue Jun 07, 2005 8:27 pm Post subject: |
|
|
SAngeli,
ntpd attempts to correct your clock by applying very small corrections frequently.
It maintains a drift rate file to help it do this.
However, it needs a long time to calibrate your clock against time servers and may fail if you switch the PC off.
It will only correct small errors too. (I forget how much, but less than 1 second)
If you switch your PC off, you can use ntpdate once at power on to fix any errors that are too big for ntpd.
Then ntpd does not quit because the errors are bigger than it can manage.
ntpdate gets time from a timeserver and updates the clock, no matter what the error.
There is no calibration or frequent small updates, just get time, update clock and exit.
I have a system that looses over 0.5 sec an hour, so I run nptdate every hor in a cron job.
Be aware that stepping the time backwards can cause problems. Linux will complain if it encounters file timestamps in the future.
Make may go wrong too. Its OK to use this as long as time changes are forward.
Heres the script from /etc/cron.hourly | Code: | #!/usr/bin/perl
{
# sync clock
$ntp = `/usr/sbin/ntpdate -u -t 5 ntp0.zen.co.uk`;
`logger -f /var/log/everything/current -t timesync "clock sync1: $ntp"`;
$ntp = substr $ntp, (index $ntp, "offset");
`logger -f /var/log/everything/current -t timesync "clock sync2: $ntp"`;
if ($ntp eq '')
{
$ntp = "failed3";
}
# update hardware clock
`/sbin/clock --systohc`;
# write to log
`logger -f /var/log/everything/current -t timesync "clock sync4: $ntp"`;
} |
ntp0.zen.co.uk is my ISPs timeserver. It may not be public.
and ntpdate may have moved from /usr/sbin/ntpdate too. so you may need to edit that script before you can use it.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
SAngeli
l33t
Joined: 16 Apr 2004
Posts: 904
Location: Italy
|
| Posted: Tue Jun 07, 2005 8:47 pm Post subject: |
|
|
Hi NeddySeagoon
thank you for the explanation. I assume I am wrong here but should I not solve the ntpq: read: Connection refused first?
One comment above mentioned this:
| F.Ultra wrote: | failed to drop root privileges: Operation not permitted
This means that the ntpd daemon cannot change from beeing root into the ntp:ntp user, and it does therefore quit. And that is why you cannot see it running and why the init.d script cannot kill it (since it is not running). |
I believe perhaps I need to solve this issue first so that when I can succesfully run ntpd -p then I can investigate if my system loose time or not and then apply your suggestion.
See also this and portion of the thread that I believe is important to solve:
What do you think?
Spiro |
|
| Back to top |
|
|
hgerstung
Tux's lil' helper
Joined: 22 Nov 2004
Posts: 123
Location: Northwestern Germany (OWL)
|
| Posted: Wed Jun 08, 2005 12:07 pm Post subject: |
|
|
Hi Spiro,
just check out this thread:
http://forums.gentoo.org/viewtopic-t-162583.html
It basically says the solution is to load a specific module called "capability" and is related to the CONFIG_SECURITY option(s) in your kernel configuration.
Kind regards,
Heiko |
|
| Back to top |
|
|
|
Networking & Security
