ntpd seems to work but how well?
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1. One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English? # ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
Gonda wrote regarding '[SLE] ntpd seems to work but how well?' on Wed, Sep 01 at 08:11:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English?
# ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
Your ntp isn't working well, I think. root@newwww:/tmp > ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== LOCAL(0) LOCAL(0) 4 - 32 64 377 0.000 0.000 0.000 -ntp-0.gw.uiuc.e truechimer.cso. 2 u 35 1024 377 16.029 -0.519 0.152 #gilbreth.ecn.pu molecule.ecn.pu 3 u 32 1024 377 80.024 -3.171 37.271 #harbor.ecn.purd atom.ecn.purdue 4 u 539 1024 377 57.604 16.357 34.192 -molecule.ecn.pu usno.pa-x.dec.c 2 u 541 1024 377 125.504 24.895 0.000 *caesar.cs.wisc. ben.cs.wisc.edu 2 u 10 1024 377 10.708 -0.307 0.121 +milo.mcs.anl.go clepsydra.dec.c 2 u 931 1024 377 58.452 -0.247 0.252 +mcs.anl.gov clepsydra.dec.c 2 u 176 1024 377 58.688 0.221 0.678 wuarchive.wustl 0.0.0.0 16 u - 1024 0 0.000 0.000 4000.00 -dense.utcc.utor tyme.gw.utoront 2 u 536 1024 377 23.569 -4.097 0.000 The line with the * at the beginning is the best peer right now, and the one ntpd is syncing off of. You should have one of those, even with one server: dsauer@danny-pc:/tmp> /usr/sbin/ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== *newwww.internal 130.126.24.53 3 u 1000 1024 377 0.512 -0.649 0.359 Anyway, the important columns, IMHO, are st - stratum (there are 2 hosts between a stratum3 host and an atomic clock), offset (the difference in your clock and theirs), jitter (the reliability of that server), the delay (the distance between you and them) the refid (that server's reference). Also, the first char is important. It'll take some ntpd docs that I don't have in' front of me to explain that. You may wanna look at /usr/share/docs/xntpd to get more useful information... --Danny, probably not helping enough
Gonda Brouerius van Nidek wrote:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English?
# ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
1st step: Check if you can talk to the server:
laza@nics:~> ntpdate -v -q -u time.windows.com
1 Sep 19:39:49: ntpdate 4.1.1@1.786 Thu Oct 2 20:27:36 UTC 2003 (1)
server 207.46.130.100, stratum 2, offset -0.008529, delay 0.23824
1 Sep 19:39:50: adjust time server 207.46.130.100 offset -0.008529 sec
If you got no answer or errors (no server suitable for synchronization
found), check server's configuration - allowed hosts, networks, etc (only if
it is the case :-)).
To really adjust the date, as root type
ntpdate -b
Marcos Vinicius Lazarini wrote:
Gonda Brouerius van Nidek wrote:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English?
# ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
1st step: Check if you can talk to the server: laza@nics:~> ntpdate -v -q -u time.windows.com 1 Sep 19:39:49: ntpdate 4.1.1@1.786 Thu Oct 2 20:27:36 UTC 2003 (1) server 207.46.130.100, stratum 2, offset -0.008529, delay 0.23824 1 Sep 19:39:50: adjust time server 207.46.130.100 offset -0.008529 sec
If you got no answer or errors (no server suitable for synchronization found), check server's configuration - allowed hosts, networks, etc (only if it is the case :-)).
To really adjust the date, as root type ntpdate -b
or ntpdate -b -u Note: this command only perform one request to the server - it should be called from a startup script at boot time. After you got a correct date/time, the regular ntp daemon should take place and keep the time drift under control. Then you should check ntp.drift, etc, etc, etc
http://www.whirlnet.co.uk/linux/ptktime is the utility I use to check my local time is keeping in step with the server. Just make it executable and place it in /usr/local/bin/. I've been using it for years now, amazingly it still works on SuSE 9.1 both x86 and x86_64. Regards Sid. -- Sid Boyce .... Hamradio G3VBV and keen Flyer =====LINUX ONLY USED HERE=====
On Wednesday 01 September 2004 05:11 am, Gonda Brouerius van Nidek wrote:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English?
# ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
How long did you wait after starting ntpd to check these numbers? It takes a while to sync you know. 5 to 10 minutes. After a bit you will see an asterisk on one of the clock lines. That means its synced with that clock. From then on it will be dead on perfect. (for some definitions of perfect). It will try to sync with a clock in the highest strata (smallest ST number) that has the lowest Delay. But it sometimes will select another clock by some hocus pocus I don't understand. By the way, your machine will never sync if its clock is off by a LOT, so get it close before you start. Your ISP should supply a strata 2 or 3 clock, and that is usually the one to use, but some ISPs dont do it, or let theirs get hopelessly out of sync. Generally its recommend you NOT use one of the strata 1 clocks because if everybody did, it would get overloaded. There are lists of clocks on the internet somewhere. Google will find them, or look here: http://www.ntp.org/ http://www.eecis.udel.edu/~mills/ntp/servers.html http://www.ntp.org/ntpfaq/NTP-a-faq.htm -- _____________________________________ John Andersen
On Thursday 02 September 2004 03:51, John Andersen wrote:
On Wednesday 01 September 2004 05:11 am, Gonda Brouerius van Nidek wrote:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
One of the possibilities to see if it really works should be to use ntpq -p I get the following output with this command but apart from recognizing the four addresses that I have chosen I do not understand the meaning of the output. Somebody with an explanation in plain English?
# ntpq -p remote refid st t when poll reach delay offset jitter ===================================== 202.180.0.71 .INIT. 16 u - 64 0 0.000 0.000 4000.00 my1.doubleukay. 192.6.38.127 2 u 40 64 1 1039.17 -183.26 0.002 ns01.deu.edu.tr .INIT. 16 u 49 64 0 0.000 0.000 4000.00 darkstar.sanet. 212.82.32.15 2 u 58 64 1 908.414 -82.537 0.002
I think the 4000 refer to unreachable servers. I double checked my setup and I forgot to reopen the firewall to ntp packets. Should the firewall not opened to ntp packets by the install script? -- Collector of vintage computers http://www.ncf.ca/~ba600 Looking for: PICMG, Nabu CP/M disks Open Source Weekend http://www.osw.ca
On Thursday 02 September 2004 14:51, John Andersen wrote:
On Wednesday 01 September 2004 05:11 am, Gonda Brouerius van Nidek wrote:
I have read the info on ntpd several times and I think I understand the basics. I hope I have ntpd working on my dialup system with 9.1.
How long did you wait after starting ntpd to check these numbers?
It takes a while to sync you know. 5 to 10 minutes.
After a bit you will see an asterisk on one of the clock lines. That means its synced with that clock. From then on it will be dead on perfect. (for some definitions of perfect).
By the way, your machine will never sync if its clock is off by a LOT, so get it close before you start.
Your ISP should supply a strata 2 or 3 clock, and that is usually the one to use, but some ISPs dont do it, or let theirs get hopelessly out of sync. Generally its recommend you NOT use one of the strata 1 clocks because if everybody did, it would get overloaded. There are lists of clocks on the internet somewhere. Google will find them, or look here: http://www.ntp.org/ http://www.eecis.udel.edu/~mills/ntp/servers.html http://www.ntp.org/ntpfaq/NTP-a-faq.htm I leave my ISP out of the time question ;-). In a country where time is elastic I suspect that my ISP is no good time keeper. Have read the faqs and other info available but the hands on experience is something you do not easy find there.
Thanks for the help. Think that the connection via my dial up is sometimes overloaded. Must have been a big download or a big spam which took all of the connection bandwidth. Right now I get a star for darkstar.sanet and after a ten minutes I also got a plus sign in front of ns01.deu.edu.tr What does the plus mean? Just before I wanted to send this email my1.doubleukay also got a plus in front. 202.180.0.71 .STEP. 16 u - 128 0 0.000 0.000 4000.00 my1.doubleukay. 164.67.62.194 2 u 49 64 107 1022.56 -58.216 1.617 +ns01.deu.edu.tr 195.113.144.201 2 u 59 64 177 915.493 -0.621 13.994 *darkstar.sanet. 212.82.32.15 2 u 56 64 177 840.781 -17.948 40.706
On Thursday 02 September 2004 05:01 am, Constant Brouerius van Nidek wrote:
Right now I get a star for darkstar.sanet and after a ten minutes I also got a plus sign in front of ns01.deu.edu.tr What does the plus mean? Just before I wanted to send this email my1.doubleukay also got a plus in front.
202.180.0.71 .STEP. 16 u - 128 0 0.000 0.000 4000.00 my1.doubleukay. 164.67.62.194 2 u 49 64 107 1022.56 -58.216 1.617 +ns01.deu.edu.tr 195.113.144.201 2 u 59 64 177 915.493 -0.621 13.994 *darkstar.sanet. 212.82.32.15 2 u 56 64 177 840.781 -17.948 40.706
An asterisk * in the first column marks the reference time source which is currently preferred by the NTP daemon, the + character marks high quality candidates for the reference time which could be used if the currently selected reference time source should become unavailable. The column remote displays the IP address or the host name of the reference time source, where LOCAL refers to the local clock. The refid shows the type of the reference clock, where e.g. LOCAL or LCL refers to the local clockagain, .DCFa. refers to a standard DCF77 time source, and .PPS. indicates that the reference clock is disciplined by a hardware pulse-per-second signal. Other identifiers are possible, depending on the type of the reference clock. The column st reflects the stratum number of the reference time source. In the example above, the local clock has stratum 12, the remote time server at 172.16.3.103 has stratum 1 which is the best you can see across the network, and the local radio clock has stratum 0, so the radio clock is currently being preferred. Every time a when count reaches the poll number in the same line, the NTP daemon queries the time from the corresponding time source and resets the when count to 0. The query results of each polling cycle are filtered and used as a measure for the clock's quality and reachability. The column reach shows if a reference time source could be reached at the last polling intervals, i.e. data could be read from the reference time source, and the reference time source was synchronized. The value must be interpreted as an 8 bit shift register whose contents is displayed as octal values. If the NTP daemon has just started, the value is 0. Each time a query was successful a '1' is shifted in from the right, so after the daemon has been started the sequence of reach numbers 0, 1, 3, 7, 17, 37, 77, 177, 377. The maximum value 377 means that the eight last queries were completed successfully. The NTP daemon must have reached a reference time source several times (reach not 0) before it selects a preferred time source and puts an asterisk in the first column. The columns delay, offset and jitter show some timing values which are derived from the query results. In some versions of ntpq the last column is labeled disp (for dispersion) instead of jitter. All values are in in milliseconds. The delay value is derived from the roundtrip time of the queries. The offset value shows the difference between the reference time and the system clock. The jitter value indicates the magnitude of jitter between several time queries. -- _____________________________________ John Andersen
On Thursday 02 September 2004 11:10 am, John Andersen wrote:
On Thursday 02 September 2004 05:01 am, Constant Brouerius van Nidek wrote:
Right now I get a star for darkstar.sanet and after a ten minutes I also got a plus sign in front of ns01.deu.edu.tr What does the plus mean?
An asterisk * in the first column marks the reference time source which is currently preferred by the NTP daemon, the + character marks high quality candidates for the reference time which could be used if the currently selected reference time source should become unavailable.
By the way, forgot to mention, the above was cribbed from http://networking.ringofsaturn.com/Protocols/ntp.php which is a pretty good reference for this stuff. -- _____________________________________ John Andersen
The Wednesday 2004-09-01 at 23:51 -0800, John Andersen wrote:
Your ISP should supply a strata 2 or 3 clock, and that is usually the one to use, but some ISPs dont do it, or let theirs get hopelessly out of sync.
Uau. I have not found any ISP that does in my part of the world... (for a time, I worked for an ISP, and I needed a time reference, internally, for "serious things". They had no idea what I was talking about). -- Cheers, Carlos Robinson
On Saturday 04 September 2004 01:19, Carlos E. R. wrote:
I have not found any ISP that does in my part of the world...
(for a time, I worked for an ISP, and I needed a time reference, internally, for "serious things". They had no idea what I was talking about).
~ at the eastern end of the mediterranean, Forthnet Greece does . . . very advanced :) -- best wishes ____________ sent on Linux ____________
The Saturday 2004-09-04 at 18:21 -0000, pinto wrote:
~ at the eastern end of the mediterranean, Forthnet Greece does . . . very advanced :)
Nice :-) Perhaps we'll have to organize the Olympics here to get some new advances ;-) Mmmm... come to think of it, China would like ipv-6 to really start, I guess. I'd like that. -- Cheers, Carlos Robinson
On Saturday 04 September 2004 03:19, Carlos E. R. wrote:
The Wednesday 2004-09-01 at 23:51 -0800, John Andersen wrote:
Your ISP should supply a strata 2 or 3 clock, and that is usually the one to use, but some ISPs dont do it, or let theirs get hopelessly out of sync.
Uau.
I have not found any ISP that does in my part of the world...
(for a time, I worked for an ISP, and I needed a time reference, internally, for "serious things". They had no idea what I was talking about).
-- Cheers, Carlos Robinson
There's an excellent server at 130.149.17.21 Absolutely spot on each time every time. Cheers, Steve.
The Saturday 2004-09-04 at 20:24 +0200, steve-ss wrote:
I have not found any ISP that does in my part of the world...
(for a time, I worked for an ISP, and I needed a time reference, internally, for "serious things". They had no idea what I was talking about).
There's an excellent server at 130.149.17.21
That's "hora.cs.tu-berlin.de" - the domain is german, but the word "hora" is spanish. Funny. However, that was not the point; I know there are timeservers in Spain. The point is that ISPs here do not provide NTP servers inside their organizations, as a service to their clients, as far as I know. They should be something.terra.es, something.tiscali.es, something.auna.es, something.ono.es, etc. All of them big providers. As far as I know, none has time servers, or they keep silent about it. As a matter of fact, I don't know if we have any "oficial" server. We do have (or had?) an atomic clock somewhere, but to what is it connected? Not even a dedicated radio signal, for example. -- Cheers, Carlos Robinson
participants (10)
-
Carlos E. R. -
Constant Brouerius van Nidek -
Danny Sauer -
Gonda Brouerius van Nidek
-
John Andersen -
Marcos Vinicius Lazarini -
Mike Kenzie -
pinto -
Sid Boyce -
steve-ss