The NTP Pool is built to provide a usable time service to a large number of clients, most of which aren't time servers themselves. Desktop computers are a major component of that, but so are web servers, mail servers, routers, and other things that need a good time (but not necessarily an excellent time).
The manufacturer of your reference clocks doesn't matter for pool purposes, for better or for worse. Stratum doesn't really matter, either. What does matter is your server's ability to handle an appreciable amount of traffic targeted at your IP address: I'm averaging about 150 queries per second, as an example. I don't know how much of this is "new" traffic (from recent DNS lookups) or "old" traffic (from lookups a long time ago). What I do know is that if my IP address changed, whoever received it next would not be happy. If you aren't running a time server, receiving 150 UDP packets per second looks a lot like an attack. This sort of collateral damage would outweigh any benefit of having that server in the first place. So, the line has to be drawn: if your IP address will change within the next few months, it shouldn't be in the pool. -rt On Fri, Jun 14, 2013 at 12:40 PM, Miguel Barbosa Gonçalves <[email protected]> wrote: > > NTP is a great protocol but the pool project could be improved. > > I believe that people that don't care about NTP just set it and forget it. > These ate people that turn their computer off at night. Those who run > servers don't do that. I at least don't. But maybe I am an exception. After > all I am better equipped than my country's astronomical observatory. I > count 5 stratum 1s on my network - 2 of them from Meinberg. > > As all things one may use it or not - the pool I mean. The stiffness of > the monitoring platform of the NTP pool project drove me away from it. I > have my own servers and just depend on GPS for timing. > > Quite frankly the "it's just how it is" bugs me... But again that's me. > > Cheers, > Miguel > > No dia 14/06/2013, às 17:29, Brian Rak <[email protected]> escreveu: > > Once ntpd starts using an IP address, it continues to use it until the > process exits (unless you're using the new 'pool' command, but that > requires a very recent version of ntpd). > > There are large numbers of servers and other machines that are using the > pool that are up for long periods of time. For example, pretty much every > Linux distribution defaults to pool servers. People who don't know (nor > care) about how NTP works are likely to just 'yum install ntp; service ntpd > start' and not think about it any further. I seriously doubt that most > people are manually picking some NTP servers to use. > > > On 6/14/2013 7:09 AM, Miguel Barbosa Gonçalves wrote: > > Greetings! > > I know that NTP resolves hostnames on startup and, after that, the NTP > packets are exchanged with only those IP addresses. > > I also understand that the NTP Pool Project design is based on this > premise. > > The question is: how effective is the monitoring platform in removing > the dead or bad (from the monitoring platform viewpoint of course) NTP > servers from running NTP clients on users' computers? > > This was just a rhetorical question of course. The monitoring system > can't notify the NTP daemon in the end users' computers. > > As an end user with a 24x7 powered computer I would have to check the > NTP daemon once in a while. I know the chances of all 4 of the poll > assigned NTP servers becoming bad are pretty remote. On the other hand, for > a 24x7 computer a discerning user would probably choose 4 properly > administered and maintained NTP servers close to him. I would at least. > > So it seems to me that the NTP Pool Project would be best targeted at > computers not running 24x7 or, better, might be being used mostly by > computers in these conditions. > > Why then isn't it possible to allow users willing to provide their > bandwidth to the project to use their dynamic IP based servers to support > the project? > > Something like a quick, say 30 minutes, convergence to a 10 score and > the monitoring platform would do a DNS resolve when the server stops > responding perhaps because the user changed it's IP address. > > Any arguments why this is not a good idea? > > Cheers, > Miguel > > PS: As a side note, I have a stratum 1 at home (a Trimble Acutime Gold > antenna) and use some other close by stratum 2 servers as backup. In my > company I have 2 Meinberg NTP servers and 3 more GPS based custom made > servers. > > > > > _______________________________________________ > pool mailing [email protected]http://lists.ntp.org/listinfo/pool > > > _______________________________________________ > pool mailing list > [email protected] > http://lists.ntp.org/listinfo/pool > > > _______________________________________________ > pool mailing list > [email protected] > http://lists.ntp.org/listinfo/pool > -- Ryan Tucker <[email protected]>
_______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
