Re: [ntp:questions] What is the best synchronization possible over the network?
Hal Murray wrote: In article 49920e17.6050...@tla.org, n...@tla.org (John Ioannidis) writes: The problem setup: two locations, both within the United States, neither has roof access so no GPS reception is possible. How do you synchronize them with better than 50-microsecond accuracy? Straight NTP over the Internet doesn't do the trick. They don't need to actually be synchronized to real time, they only need to be synchronized to each other. Assume reasonably unlimited resources (running a private fiber plant across the continent *is* unreasonable). I've looked into slaving something off a voice-carrier time base, but that's not accurate enough. Maybe something over raw SONET would do the trick? SONET gets you frequency, not time. The tick marks (frames) are not synchronized across various SONET links. Even the frequency isn't locked if the two ends come from different phone companies. I think you have several basic approaches. If cell phones work where you need the time, that may be an alternate to GPS. I don't know what flavor of cell service they use. Gene Miller suggested these devices: http://www.endruntechnologies.com/network-time-server.htm They are very reasonably priced: under $4K for the full-featured time server, plus $3K for the Rb oscillator. So long as neither VZ nor S go out of business, we're in good shape! One approach is to get good clocks at both ends and let them coast. Atomic clocks are $50K to $100K ballpark. They have long term accuracy of 1E12. So you get to coast for roughly 50*1E12 microseconds or 50E6 seconds. That's 58 days, so it will be a lot of work to keep them synchronized. I haven't worried about factors of two. You might get better than the spec accuracy if your air conditioning is good... Actually, 50E6 seconds is 578 days. It's not out of the question to have a third clock always sync-ed to GPS, fed off a 48V battery, and drive it to the data centers twice a year to slave the local oscillators. Another approach would be to build an extension cord to a place where you can get an antenna on the roof, setup a GPS unit there, and send the PPS signal to where you need it. This may require more research/testing than you want to do. Is roof access really impossible? How much would it cost? It involves problems at layer 9 :( Do you have any windows? Have you tried a GPS receiver there? With a good timing receiver, you only need 1 satellite. (You need several for setup, only one to keep going.) No windows, I use Solaris :) There is, however, cellphone reception in the datacenters, so the EndRunTechnologies devices should work. Thanks again /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] What is the best synchronization possible over the network?
The problem setup: two locations, both within the United States, neither has roof access so no GPS reception is possible. How do you synchronize them with better than 50-microsecond accuracy? Straight NTP over the Internet doesn't do the trick. They don't need to actually be synchronized to real time, they only need to be synchronized to each other. Assume reasonably unlimited resources (running a private fiber plant across the continent *is* unreasonable). I've looked into slaving something off a voice-carrier time base, but that's not accurate enough. Maybe something over raw SONET would do the trick? Thanks /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Answer: PPS on a rack server
A few days ago I had asked this list for assistance with the following problem: my rack-mounted IBM x3650 servers only have one serial port, and the data I was feeding them from my GPS clock were interfering with the serial console feature, the BIOS in particular. Thanks to several people who replied to my original message, I now have a working setup. Here it is: 1. A Spectracom 9283 netclock feeds *only* the PPS signal to the servers. Pins 1 (DCD) and 5 (GND) of a DB-9F connector are connected to the core and shielding of the coax (through T connectors, and the cable is terminated with a 50 Ohm terminator from the old thinlan days!). The signal seems to be strong enough to drive the DCD pin, but for longer runs I would use this: http://www.tapr.org/kits_tadd-3.html. 2. NTP is used to get the time reference from the Spectracom. 3. The harder part was figuring out what to put in /etc/ntp.conf. Here is what I have (the address of the Spectracom is not the real one, of course) server 192.2.200.2 minpoll 5 maxpoll 5 prefer # 9283 server 127.127.22.0 minpoll 5 maxpoll 5# PPS fudge 127.127.22.0 flag3 1 Of course, the kernel is configured with option PPS, and /dev/cuad0 is symlinked to /dev/pps0. That's it! The machines appear to be within a few microseconds of each other, so I'm happy. 4. I can still use the serial console, after a fashion -- the machines support IPMI, so I can use serial-over-lan with ipmitool to get console i/o (for the BIOS, which is all I care about) without actually needing to plug anything to the serial ports. Of course, what I really should have done is paid the extra $1K or so and gotten the RSA-II cards for the boxes, but that's another story. Thanks again to all who replied /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Getting PPS on a modern rack-mounted server?
This may sound like an embarrassment of riches, but here is my problem: I have some rack-mounted servers (IBM System x3650 to be precise) that I need to synchronize with a PPS signal. The problem is, of course, that the machines have only one serial port, which I want to use as a serial console, and of course they have no parallel ports. Any suggestions? Can I use a PCI serial card? Will that have acceptable jitter? Thanks, /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Using a PPS source without a GPS receiver?
The setup: O(10) FreeBSD 6.2 machines in a rack, a PPS source, and an NTP server somewhere on the same network. Is it possible (and if so, how?) to configure ntpd on these machines so they get the rough time over NTP from the network's NTP server, and use the PPS source so they stay better synchronized to each other (with less than a foot of coax between each machine, I'm not worried about the extra nanosecond!) Thanks, /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Inexpensive OEM GPS units?
Out of curiosity: what is wrong with the Garmin GPS 18LVC that someone would like to look at an alternative? At $70, it's practically free. /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] How do I know my GPS-based NTP server is actually working properly?
Thanks for the reply, but this is not answering my question. I wasn't asking how to *configure* the beast (I've already done this successfully), I was asking how to verify that it actually works as documented! What measurements do I have to take that will show the difference between a setup that's actually using the PPS signal from the GPS receiver and one that's not (because, for example, the DCD line on the motherboard is cracked, to give a stupid example). Cheers, /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] How do I know my GPS-based NTP server is actually working properly?
Kevin Oberman wrote: Date: Fri, 02 Nov 2007 12:25:46 -0400 From: John Ioannidis [EMAIL PROTECTED] Thanks for the reply, but this is not answering my question. I wasn't asking how to *configure* the beast (I've already done this successfully), I was asking how to verify that it actually works as documented! What measurements do I have to take that will show the difference between a setup that's actually using the PPS signal from the GPS receiver and one that's not (because, for example, the DCD line on the motherboard is cracked, to give a stupid example). I guess I was not clear. 'ntpq -p' will provide the output I showed and this will tell you whether the PPS is working. (Note: I don't see that it will as your configuration is not correct.) By working I mean actually having an effect. Not simply that the kernel takes notice, but that I'm actually getting better results. Hence my insistence on statistical measurements. I'm also curious as to how much better my accuracy is when I use the PPS (and hence whether it's worth the extra trouble for my particular application). AFAICT, my configuration is correct (modulo the polling intervals). The flag3 1 in the fudge line tells the NMEA driver (notice the .20. in the third octet of the fake IP address) to actually use the PPS signal. For the record, here is the output of ntpq -p # ntpq -p remote refid st t when poll reach delay offset jitter == *GPS_NMEA(1) .PPS.0 l 53 64 3770.0000.019 0.014 Am I missing something? Thanks, /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] How do I know my GPS-based NTP server is actually working properly?
Yes, I know this sounds weird. I've successfully set up my NTP server: an old 850MHz P3 box running FreeBSD 6.2-STABLE (kernel built with option PPS_SYNC), and a Garmin GPS-18LVC hokey puck, powered off the USB port, with its PPS wire connected to the serial port's DCD line (pin 1 on the DB9 connector). /etc/ntp.conf reads: server 127.127.20.1mode 1 prefer fudge 127.127.20.1time1 0.000 flag3 1 refid PPS So far so good. It seems to be keeping time and synchronizing to GPS. But... What's the proper way of telling whether the PPS signal is actually having an effect? That is, what behavior/measurements would show me that the PPS signal is or is not being used? What should I be monitoring (my guess would be offset and jitter from the netq -c rv output), and what kind of statistical analysis would tell me whether PPS is or is not being used? Thanks, /ji PS: the easy way to disable the use of the PPS signal is to set flag3 to 0, right? Reloading a kernel and rebooting is a bit of a pain on antiquated hardware. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] FreeBSD 6.2 with Spectracom 9283 GPS units?
I need to syncronize several FreeBSD boxes off a Spectracom NetClock 9283 unit. For reasons that I can't elaborate here, I can only use the RS485 and PPS signals off the 9283, not their built-in NTP servers. So far, it looks like my best bet is to run STP and 50Ohm coax wiring from the Spectracom to the FreeBSD boxes, get RS485-RS232 converters, and using some leftover BNC T connectors from the days of thinlan connect the PPS signal that's coming down the coax to the signal ground and DCD pins of the RS232 connector. Questions: 1. Has anyone on the list actually interfaced a 9283 to a unix box this way? ANything I should watch out for? 2. Does there already exist a box that will do all this, so I don't have to put it together myself? Thanks, /ji ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions