Re: [ntp:questions] SNTP test bench
David L. Mills [EMAIL PROTECTED] writes: The rate violation is caught in the MRU list, which can be retrieved using ntpdc and the monlist command. When the number of clients is small, the list can be retrieved over the net. When the number of clients is larte, like several hundred, there are many UDP packets and one or more are usually dropped. The solution at present is to run ntpdc on the server machine and pipe the monlist output to a local file. Each time a KoD is sent a counter is increased by one. Once each second the counter is decreased by one. If an offending packet arrives and the counter is less than 2, a KoD is sent; otherwise, the packet is dropped without further action. There probably should be some triage, but not without additional complexity. This is both interesting and useful, but begs the question, which was what monitor semantics are and how the parameter should be specified (0-1, percentage, whatever) Also, it wouldn't hurt to copy-paste what you wrote above into the doc on udel.edu :) DES -- Dag-Erling Smørgrav - [EMAIL PROTECTED] ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP vs chrony comparison (Was: oscillations in ntp clock synchronization)
On Wed, 30 Jan 2008, Danny Mayer wrote: Unruh wrote: David L. Mills [EMAIL PROTECTED] writes: David, We can argue about the Hurst parameter, which can't be truly random-walk as I have assumed, but the approximation is valid up to lag times of at least a week. However, as I have been cautioned, these plots are really sensitive to spectral lines due to nonuniform sampling. I was very careful to avoid such things. But the lines I am refering to are not artifacts, they are there because of the way the computer is used. -- the temp fluctuations caused by people running the machine daily, except on weekends. These are not part of any random walk process. They are real jumps in the drift rate of the machine, large jumps, and definitely not random. Well of course. You are running Linux and losing interrupts. FreeBSD and friends don't suffer from that problem. I seem to remember setting HZ=100 mostly eliminates that problem, at the price of rebuilding the kernel. Danny No they are not lost interrupts. They are NOT jumps in the offset, they are jumps in the frequency, which will last for a few hours and then jump back. Lost interrupts do not act like that-- they would jump the offset by 10ms (or 4ms) which is definitely not happening. Andit is hard to gain interrupts. -- William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273 PhysicsAstronomy | Advanced Research | Fax: +1(604)822-5324 UBC, Vancouver,BC | Program in Cosmology | [EMAIL PROTECTED] Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/ ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] First attempt GPSD/PPS -NTP time server
Hi there Unruh wrote: This is confusing. You first say that one NMEA sentence pers second is too much data, and then that youarranged that it sent 6 sentences per second. Note that only one sentence ( which should take about 140ms at 4800Bd) is allo you need. GPRMC, GPGGA, GPGSA and three GPGSV lines (default) are max 367 chars. All sentences enabled are max 575 chars, which takes longer then one second to send. If you just want time, GPRMC is all you need. If you want to see where the sats are, you need more. See http://www.garmin.com/manuals/425_TechnicalSpecification.pdf Regards, Rob -- When the Iron Curtain fell, all of the West rejoiced that the East would become just as free as the West. It was never supposed to be the other way around. (Rick Falkvinge) ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] strange behaviour of ntp peerstats entries.
David L. Mills wrote: The result of the sort is usually the first entry on the list, but even I thought this only gave the figure head peer, but that the actual clock disciplining used a weighted average of some number of candidates aa well. that can be temporarily displaced by the anti-clockhop algorithm. Other than this wiggle, it can be truthfully said that the algorithm selects the servers at the lowest stratum and from among those at that stratum the candidate with the lowest maximum error. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP vs chrony comparison (Was: oscillations in ntp clock synchronization)
Danny Mayer wrote: Well of course. You are running Linux and losing interrupts. FreeBSD and Lost interrupts are not the problem here and nothing about FreeBSD should help (unless it runs the CPU permanently at full power). ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] strange behaviour of ntp peerstats entries.
Steve Kostecke wrote: If you wish to have a specific feature in NTP you can add it yourself using the source which is available for download from: Whilst that is often an appropriate response, it isn't here. My impression is that Unruh has already downloaded the source code and studied. My impression is that he has also already found what he considers a solution. Unfortunately for the reference implementation, it is in a competing product. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] GPS and NTP Server
noosh wrote: Thank you for kind attention. I have GPS hopf 6842 which is connected Exactly which model? There are 7 different user manuals. to the NTP Server. i want to connect these 2 device(GPS to NTP Server) by serial port and then NTP Server should get the time from GPS. how For the most accurate time, you need to connect the PPS outputs, which seem to be separate from the RS 232 interface. (It's not worth downloading the manual to check this without the exact version information.) NTP Server sychronize itself with GPS which is connecte to its serial port then? According to the NTP documentation. Alternatively, I would guess you may be able to find someone who will support you at standard consultancy rates. This may give you some clues: http://www.hopf.com/software/products/ntp/README.REDHAT ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] why is my pool server's offset so bad
Pat Farrell wrote: On Wed, 30 Jan 2008 23:40:17 -0500, Richard B. Gilbert wrote: But lately, the offsets are diving down. See http://www.pool.ntp.org/scores/70.184.242.241 Sorry about that. How about supplying some USEFUL information about your server! Knowing what O/S you are running on which hardware platform, what version of ntpd I opened it up for general queries so you can ntpq -p and see. Its a debian Etch on a dual Xeon (intel 32) platform. ntpd 4.2.2p4 Its got no real clocks, its just a load sharing device, getting good time from overworked upstream ntp servers and sharing it with the world It might be helpful to start with the ntpq -p banner. Don't know what you mean by 'banner' here is ntpq -p [EMAIL PROTECTED]:/etc$ ntpq -p remote refid st t when poll reach delay offset jitter == ntp-2.cns.vt.ed 198.82.247.402 u 31h 640 17.908 -0.296 0.000 ntp-4.cns.vt.ed 198.82.247.164 2 u 31h 640 18.5670.637 0.000 ancalagon.cede. .INIT. 16 u- 102400.0000.000 0.000 prometheus.acm. .����. 16 u 31h 6400.000 0.000 0.000 time-b.nist.gov .ACTS. 1 u 31h 640 13.5556.162 0.000 You should at least upgrade to 4.2.4. The refid garbage on the line for prometheus was fixed a long time ago. It has been my experience that ntpd, when properly configured with good servers and, optionally, a good ref clock, keeps time very well indeed. It was my experience that for four or five months, the server scored 20 constantly. Recently, not so much. Those numbers above and the ones I got when I queried your server are excellent for most purposes. You don't want to see my numbers, though admittedly I had hibernated my laptop and ntpd is still recovering right now. My stationary systems are much more stable. Danny ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] strange behaviour of ntp peerstats entries.
On 2008-01-31, David Woolley [EMAIL PROTECTED] wrote: Steve Kostecke wrote: If you wish to have a specific feature in NTP you can add it yourself using the source which is available for download from: Whilst that is often an appropriate response, it isn't here. My impression is that Unruh has already downloaded the source code and studied. My impression is that he has also already found what he considers a solution. There's nothing stopping him from implementing what he considers to be a solution himself. He could even distribute his modified version of NTP to anyone who wanted to use it. -- Steve Kostecke [EMAIL PROTECTED] NTP Public Services Project - http://support.ntp.org/ ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions