Slashdot Mirror

Someone else replied, but let me actually EXPLAIN.

pool.ntp.org is a collection of volunteer NTP servers, served up via DNS. You should not expect to get meaningful results from pointing a Web browser at such a host name, but because it is random, you could end up hitting Amazon.com (assuming they volunteered) or some guy that just set up an Apache server.

http://www.pool.ntp.org/ is what you meant, as a simple google search for "pool ntp" would have told you.

I think part of the problem is that it is http://www.pool.ntp.org. That's a pretty common "mistake" (not really right word) in my mind. I always setup my web sites and DNS to work with or without the www host name (ala ServerAlias or html redirect) -- I'm just simple that way.

BUT, since pool.ntp.org uses round-robin DNS, each server would need to setup the appropriate forwarding to be all friendly like. Or at least turn off httpd. Security Rule "#0" still stands.

I decided to check out the pool.ntp.org website. Comedy. It's the default Apache "you've successfully installed your webserver!" page. Either put something meaningful there or shut off httpd.

Security Rule #0: You shouldn't run services you don't need.

http://www.pool.ntp.org/

These people may disagree. This doesn't change the fact that D-Link makes shitty firmware.

The NTP Pool website makes it look like it is a good idea if every machine on a network syncs to the NTP Pool, instead of setting up internal servers, which is how NTP is really designed to work.

From http://www.pool.ntp.org/use.html, second to last line in "Additional Notes":

If you are synchronising a network to pool.ntp.org, please set up one of your computers as a time server and synchronize the other computers to that one. (you'll have some reading to do - it's not difficult though. And there's always the comp.protocols.time.ntp newsgroup.)

According to the project web page you can expect 10-20Kbit/sec of traffic, which works out to 6 gigabytes per month of traffic. It doesn't say which direction but I suspect NTP would be pretty symetrical so this would triple the inbound volume to my co-lo.

Thats a lot of volume for me, so I don't see how I could contribute a server.

Its a shame that they can't include a dynamic DNS hack into the system. My home system has heaps of volume at a fixed price, but it is on a dynamic IP.

what constitutes a "public" NTP server - the DNS name, or its inclusion on a particular published list?

According to the FAQ on the NTP homepage "NTP uses UTC as reference time"

Further down there is a discussion of how leap seconds are handled. I was curious so I checked my computer clock (which is synced using NTP) against my alarm clock (which uses the radio signal from the MSF service and they are the same. So it seems that NTP must have observed leap seconds contrary to your original post.

Really, all most of us need to know about NTP can be found on Ntp.org.. Particularly useful is their NTP Pool project, which uses DNS aliasing to allow you to sync from random servers, and avoid placing a burden on any one server.

Really, all most of us need to know about NTP can be found on Ntp.org.. Particularly useful is their NTP Pool project, which uses DNS aliasing to allow you to sync from random servers, and avoid placing a burden on any one server.

Dave Mills, author of the NTP RFC and main implementor of the NTP server software since time immemorial (back in the Fuzzball distro for PDP-11 days) has an extensive collection of information about time and how it relates to networks. Here's a specific page about the leap second:

http://www.eecis.udel.edu/~mills/leap.html

For more info, see the NTP site:

http://www.ntp.org/

and the site of the Network Time Synchronization Project:

http://www.eecis.udel.edu/~mills/ntp.html

There is a project to serve NTP round-robin from a number of servers. You can use this pool thusly with ntpd:

server pool.ntp.org

If you live in Canada or the US you can even do:

server north-america.pool.ntp.org

Read more at:

http://www.pool.ntp.org/

Sorry, I could not resist to reply to this.
Running ntpdate from a cronjob (that's what you're talking about) is silly. You would then still have 59 minutes in which the clock can skew as much as it likes.
If you want to do it the right way, download the ntp client from ntp.org. That one constantly adjusts your clock depending on the skew it carefully measures over a longer period of time.

(climbs of soapbox)

AOL cant block as they cant figure out if your using HTTP or NTP

Yeah, but they'd certainly wonder why there's so much NTP (http://www.ntp.org/) traffic... how many times a day do you have to set your clock?

oops, I didn't answer the other part about pool.ntp.org:
http://www.pool.ntp.org/#news
see the "2004-09-07" entry.

I hand-built serial cables using plugs I got from this guy (Elsewhere on that site there's links to folks all over the world selling the same plugs for a range of different GPS receivers). Apparently even the tiny little Garmin Geko 201 and Geko 301 (but not the 101 model) also speak serial - and they're tiny cute little things they are!!!

My little eTrex has a menu with a whole bunch of different 'languages' that it will speak (and/or receive) via the serial port. According to the manual (warning: pdf) (page 45) it speaks NMEA 0183, a bunch of proprietary Garmin stuff and a couple of flavours suitable for differential work. I know from fiddling with mine that it also speaks a 'plain text' (they're all plain text, but this one is more so) format that is quite human readable and probably quite easily parseable with some perl.

Another imporant point about GPS and Linux (*nix in general is time). GPS requires incredibly accurate time to operate, so by implication GPS receivers make excellent clocks. Last time I checked xntp had support for NMEA (GPS) as a time source.

A quick freshmeat (if 'google' is a verb, then surely 'freshmeat' can be one too!) will tell you that GPS on *nix is nothing new!!! (Not all of those returns are gps nav related, but there's a lot of stuff to parse gps sentences, moving maps, program receivers, all kinds of goodies!

For that, you should use NTP.

There are many NTP servers that are free to access out there. Please keep them that way by observing a simple netiquette.

Read about DNS round robin sharing of voluntary ntp servers:
http://www.pool.ntp.org

MM
--

So why don't you just use NTPd or OpenNTPd? And why aren't you using pool.ntp.org instead of picking on the poor, overloaded NIST servers?

So why don't you just use NTPd or OpenNTPd? And why aren't you using pool.ntp.org instead of picking on the poor, overloaded NIST servers?

I've always imagined that the proliferation of wireless communications would eventually replace the need for having any sort of portable timing devices... I mean, my computer updates its clock from some atomic NTP server. A wifi clock could do the same.

Why carry an atomic clock, when you can talk to an even more accurate atomic clock, through the air? Although I guess the few ms of lag between the request and response might introduce too much error for some applications?

On a network, the key is to have precise time. That is, the exact same not-neccessarily-correct time on every single computer on the network.

NTP seeks to find coordinated universal time (UTC), that is, the "one true time" as a basis for every local time on the planet. NTP is composed of several stratums of time servers which try to determine UTC using a complex series of algorithms to measure "drift" and deviation from UTC between servers and stratums.

If you need to set your wristwatch, you look at a wall clock for reference, right? If you have a network of wall clocks across 24 time zones, you choose one as a reference and set the rest accordingly. That reference clock might be an atomic clock or a swatch watch; it doesn't particularly matter where the timekeeper is located, only that your relative distance and time differential is measured precisely. In UTC, this is UT1, the 0-median (like Greenwich Mean Time - the time at 0 longitude).

ntpdate and ntpq -pn will give you an idea of the drag between your clock and the timeserver you are connected to - theoretically, when one hour has elapsed on one clock, all clocks should should strike the hour at the same time. Since this is not possible to an infinite degree of precision, the "drag" is the amount of time it takes your clock to "catch up" and strike the hour, whether it's early or late for the period, typically the hour. Extend that out to a year, or more appropriately, several years, and you'll find even clusters of atomic clocks that strike midnight, January 1st in a leap year several nanoseconds ahead of the astronomical date change. Every year, several "leap seconds" are added to account for the differential.

Anyway, the idea is to get your server to acquire, yes, a "not-necessarily-correct time" from a variety of sources and determine the most likely time for your geographic location, usually within a few nanoseconds, and then broadcast that time for every machine on your network to syndicate. The result is not absolute perfection, but a logical use of network resources to acquire a mostly-correct time.

You'd want to set up an NTP server (maybe several) that poll stratum-2 or -3 servers for the time, and ntpd to syndicate the time for the rest of your network. Win XP, Mac OS X, etc. machines can grab the correct time every reboot, or every network logon, or whatever you prefer. The result will invariably result in a slight adjustment every time a client "re-ups" for the right time, but it should be more than sufficient for the accurate synchornization of network-wide tasks.

If you're interested in best-practice scenerios with NTP, you should really check out www.ntp.org.

Exactly why this pool makes sense. You want to make it easy for everyone to have at least decent accuracy but also to take the load off of the tier1 servers. Once more machines join in, it will make it much easier to tune to your requirements. Your normally could find more info here, but apparently there exists a discontinuity in the time dimension.

You will definitely want to use either the continent or country code versions of pool.ntp.org, otherwise you might end up sychronizing with a very distant timeserver, resulting in more NTP jitter and less accurate time.

Not only that, but as the good folks at the pool point out:

If your Internet provider has a timeserver, or if you know of a good timeserver near you, you should use that and not this list - you'll probably get better time and you'll use fewer network resources. If you know only one timeserver near you, you can of course use that and two from pool.ntp.org or so.

So check with your ISP and see if you could just hit the pool as a potential backup resource. Of course, if you're on Comcast like me, you might be out of luck looking for an internal NTP server.

Yes, only a domain controller can use this command. (see ntp.org or relavent MS documentation ). As far as I know, this has no effect on a regular windows PC.

Why use a knockoff? Use *real* time synchronization software.

Anyway, there are probably enough machines running in sync so the fireworks should be impressive at 1609Z.

Is the time controlling hardware on a PC board or PDA as accurate as an actual hard crafted caesium clock?

It's called Network Time Protocol Get the time here

Hear hear! Dave Mills was head of all sorts of stuff relating to the early 'net, in addition to inventing NTP. I can't believe they left him out! He was totally robbed.

Its interesting to note that the certification effort was made for the more proprietary (and costlier) Red Hat Advanced Server and not the basic Red Hat distribution

However if you want support for it, it will cost you about $1200 per machine per year. This is cheaper than most other OS's.

Personally, I think you would be better served developing in house resourcs for the support, but that's just me.

I'm also not necessarily happy with RH's choices on some packages to include in AS. The one that jumps out at me is choosing to use a beta version of an ntp4 release as opposed to simply using whatever was the stable version at the time.

And yes, I work somewhere that is probably going to implement hundreds of copies of RH AS, and pay for the support.

www.ntp.org the official NTP site. Links to code, hardware, documentation, you name it.

Also check out NIST's list of Manufacturers of Time and Frequency Receivers.

ntp.org. Did you even look? There are plenty of gps recievers capable of providing a pps signal to ntpd.

Where can I find a good one? I've just set up a stratum 2 time server & it would be nice to have one in case someone actually goes to the URL (and I could put a link to www.ntp.org as well).

For a few of the european hosts, we use GPS time receivers, primarily the Motorolla Oncore UT+ kits. You can get eval units of these, google around. They're nearly as easy to use, but do require a kernel config change.

It's really kind of addictive playing with time. :-) And you get spoiled by never having any clock weirdness on any of your machines...

Go to http://www.ntp.org to get all your time-synchronisation questions answered.

Also for in- or near-Germany living people: http://www.dcf77.de. Wish I knew it was a german-specific service before I came to .au and found out that my DCF77 receiver didn't work here...

> Real synching of clocks can be done with NTP
> [ntp.org], the network time protocol. And yes, it
> can sync one computer's time to GPS...

This in combination with some external software to drive your internal NTP server should keep all of your nodes on the correct time.

Windows net time /set is not synching the time, it's just setting one machine's idea of time to some more-or-less accurate value... without making sure that the time on these machines will no drift apart later....

Real synching of clocks can be done with NTP, the network time protocol. And yes, it can sync one computer's time to GPS...

Read the FAQ for more information... and the newsgroup comp.protocols.time.ntp is read by many helpful people! .

Your points about the obligation to quote all values, self-closing BR tags, disapearing B and I tags ... have all been carefully considered by W3C when they decided to formalize these things.

Advantages of formalism in markup languages:

1. No need to guess where a value ends, just parse and look for the quotation marks within a tag.

2. No need to check for UPPER and lower case versions of the same tags and attributes; everything is in lowercase.

3. Removal of presentation control allows the same source document to be used for a variety of output devices.

   If you are authoring for any medium or major -sized content provider, chances are that making the content outputable in several of the following formats is a must:

   • HTML
   • WML
   • Postscript (includes PDF)
   • MAN page
   • Newsgroup post

   By using XML as the source format (or, in this case, XHTML Basic), all of these can be covered easily using the appropriate Style Sheet or in a few cases, with sophisticated XSLT schemas.

   • HTML: no transformation needed, already in XHTML; CSS can optionally be used to add visual nicies and presentation control.

   • WML: no transformation needed, unless transport buttons are required. In any case, no presentation control available, so XHTML can pretty much be used as is.

   • Postscript: probably needs an XSLT schema to ensure a predictable, printable layout. Otherwise, the source XHTML document already provides the structure, which can easily be parsed to provide an automated Table of Content page, as a bonus.

   • MAN page: the structure is already provided by XHTML, but must be filtered by troff or similar tools to convert the tagging into appropriate markup for MAN pages; a simple substitution game, easily handled by a script.

   • Newsgroup post: the document structure is already provided by XHTML, but tagging must be stripped and converted to carriage returns, tabs or spaces, to render a plain text message.

For the record, I also hated the disappearance of favorite tags and the requirement of using lowercase for everything, because I still prefer to hand-type my HTML (with the help of auto-expanders that provide tags out of a few keystrokes).

Ever since I adopted HTML Tidy, the conversion to XHTML has been rather painless. Sure, reading lowercase-only tagging took a few weeks to get used to, but nowadays I hardly notice the difference; the tags have become easy to spot, once again.

In closing, I would like to thank W3C people for their efforts, in both web standardization and providing freeware tools to implement these standards. Tidy is something I can no longer live without! ;-)

However, (and this also goes for a lot of other standards out there, such as NTP), you people really need to learn how to distill all those technicalities into more accessible documents: XHTML Basic was the very first human-readable recommendation you produced! Heck, for once, I could find a list of supported tags in one single section, instead of having to decipher an impossibly long DTD!

... And tools like RPM, dpkg, as well as scripting systems like cfengine provide ways of readily deploying upgrades and of otherwise maintaining "system cleanliness." After all, you don't want to have the box go down when a log file fills up, and then need to ship out new hard drives to fix things if they do...