What are folks using these days for smaller organizations, that need to dole out time from an internal source?
I use the $300 GPS-based TM1000A from TimeMachinesCorp.com. Gets Stratum-1 time from GPS satellites and distributes it. Usually I relay this through a handful of local time servers to spread out the load, but it can handle hundreds of queries per minute, so it’s reasonable to use as a primary source even in moderate-sized data centers. I’ve put in a ton of them, and in most installations I buy two for redundancy. The GPS antenna works from a window in most instances . -mel beckman
On Jun 20, 2019, at 7:53 AM, David Bass <davidbass570@gmail.com> wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
On Thu, Jun 20, 2019 at 11:00 AM Mel Beckman <mel@beckman.org> wrote:
I use the $300 GPS-based TM1000A from TimeMachinesCorp.com. Gets Stratum-1 time from GPS satellites and distributes it. Usually I relay this through a handful of local time servers to spread out the load, but it can handle hundreds of queries per minute, so it’s reasonable to use as a primary source even in moderate-sized data centers.
I’ve put in a ton of them, and in most installations I buy two for redundancy. The GPS antenna works from a window in most instances .
I recently fell down the high precision time rabbithole, and now have 3 GPS units (a Truetime, a Symmetricom S250 and a LeoNTP), 3 Cesuim Primary Reference sources (an FTS4060, and 2 PRS-50s), and an assortment rubidium units. One of the "standard" solutions is one of the Microsemi (Symmetricom) SyncServer's, but these can be expensive -- I've been much happies with the LeoNTP ( http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=272 ) -- they are small, they are cheap, and they fast, they are "accurate enough", and they just work. I've got one on my desk, with a cheap (car) GPS antenna dangling out the window, and it syncs and runs happily. A friend of mine has stuffed one in an IP68 box and it's hanging happily on the side of a TV tower in the elements with no issues... I get mine from airspy.us - $349 + antenna. W
-mel beckman
On Jun 20, 2019, at 7:53 AM, David Bass <davidbass570@gmail.com> wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
-- I don't think the execution is relevant when it was obviously a bad idea in the first place. This is like putting rabid weasels in your pants, and later expressing regret at having chosen those particular rabid weasels and that pair of pants. ---maf
Warren, I like the cheap price of the LeoNTP. The only reason I prefer the Tm1000a is that it has an embedded web server, which lets me monitor the satellite constellation visibility. Otherwise, except for oven-controller time clocks, it seems obvious that the $2000+ GPS NTP servers are overpriced overkill :) -mel via cell
On Jun 20, 2019, at 8:31 AM, Warren Kumari <warren@kumari.net> wrote:
On Thu, Jun 20, 2019 at 11:00 AM Mel Beckman <mel@beckman.org> wrote:
I use the $300 GPS-based TM1000A from TimeMachinesCorp.com. Gets Stratum-1 time from GPS satellites and distributes it. Usually I relay this through a handful of local time servers to spread out the load, but it can handle hundreds of queries per minute, so it’s reasonable to use as a primary source even in moderate-sized data centers.
I’ve put in a ton of them, and in most installations I buy two for redundancy. The GPS antenna works from a window in most instances .
I recently fell down the high precision time rabbithole, and now have 3 GPS units (a Truetime, a Symmetricom S250 and a LeoNTP), 3 Cesuim Primary Reference sources (an FTS4060, and 2 PRS-50s), and an assortment rubidium units.
One of the "standard" solutions is one of the Microsemi (Symmetricom) SyncServer's, but these can be expensive -- I've been much happies with the LeoNTP ( http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=272 ) -- they are small, they are cheap, and they fast, they are "accurate enough", and they just work. I've got one on my desk, with a cheap (car) GPS antenna dangling out the window, and it syncs and runs happily. A friend of mine has stuffed one in an IP68 box and it's hanging happily on the side of a TV tower in the elements with no issues...
I get mine from airspy.us - $349 + antenna.
W
-mel beckman
On Jun 20, 2019, at 7:53 AM, David Bass <davidbass570@gmail.com> wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
-- I don't think the execution is relevant when it was obviously a bad idea in the first place. This is like putting rabid weasels in your pants, and later expressing regret at having chosen those particular rabid weasels and that pair of pants. ---maf
On Thu, Jun 20, 2019 at 11:42 AM Mel Beckman <mel@beckman.org> wrote:
Warren,
I like the cheap price of the LeoNTP. The only reason I prefer the Tm1000a is that it has an embedded web server, which lets me monitor the satellite constellation visibility. Otherwise, except for oven-controller time clocks, it seems obvious that the $2000+ GPS NTP servers are overpriced overkill :)
Yup, that is a very good point -- the LeoNTP has a small LED interface and a rotary encoder for configuration and monitoring, but it doesn't have a web UI.
From the FAQ: "Q/ Can I configure it via HTTP/Telnet ? A/ No. Running a web server on this device although entirely possible would reduce the performance of the unit. Therefore we took the decision to just do configuration via the front panel."
This is indeed a really annoying limitation - in the past I've ended up pointing a webcam at the LCD, but that is (obviously) suboptimal. I also forgot to mention that it doesn't (yet) do v6, but that might be added in future firmware versions... W
-mel via cell
On Jun 20, 2019, at 8:31 AM, Warren Kumari <warren@kumari.net> wrote:
On Thu, Jun 20, 2019 at 11:00 AM Mel Beckman <mel@beckman.org> wrote:
I use the $300 GPS-based TM1000A from TimeMachinesCorp.com. Gets Stratum-1 time from GPS satellites and distributes it. Usually I relay this through a handful of local time servers to spread out the load, but it can handle hundreds of queries per minute, so it’s reasonable to use as a primary source even in moderate-sized data centers.
I’ve put in a ton of them, and in most installations I buy two for redundancy. The GPS antenna works from a window in most instances .
I recently fell down the high precision time rabbithole, and now have 3 GPS units (a Truetime, a Symmetricom S250 and a LeoNTP), 3 Cesuim Primary Reference sources (an FTS4060, and 2 PRS-50s), and an assortment rubidium units.
One of the "standard" solutions is one of the Microsemi (Symmetricom) SyncServer's, but these can be expensive -- I've been much happies with the LeoNTP ( http://www.leobodnar.com/shop/index.php?main_page=product_info&products_id=272 ) -- they are small, they are cheap, and they fast, they are "accurate enough", and they just work. I've got one on my desk, with a cheap (car) GPS antenna dangling out the window, and it syncs and runs happily. A friend of mine has stuffed one in an IP68 box and it's hanging happily on the side of a TV tower in the elements with no issues...
I get mine from airspy.us - $349 + antenna.
W
-mel beckman
On Jun 20, 2019, at 7:53 AM, David Bass <davidbass570@gmail.com> wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
-- I don't think the execution is relevant when it was obviously a bad idea in the first place. This is like putting rabid weasels in your pants, and later expressing regret at having chosen those particular rabid weasels and that pair of pants. ---maf
-- I don't think the execution is relevant when it was obviously a bad idea in the first place. This is like putting rabid weasels in your pants, and later expressing regret at having chosen those particular rabid weasels and that pair of pants. ---maf
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick. -- Jay Hennigan - jay@west.net Network Engineering - CCIE #7880 503 897-8550 - WB6RDV
On 2019-06-20 20:18, Jay Hennigan wrote:
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
https://www.ntpsec.org/white-papers/stratum-1-microserver-howto/ RPi + GPS Hat because time across USB has much jitter. Patrick
Patrick <nanog@haller.ws>:
On 2019-06-20 20:18, Jay Hennigan wrote:
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
https://www.ntpsec.org/white-papers/stratum-1-microserver-howto/
RPi + GPS Hat because time across USB has much jitter.
I wrote that white paper, and a good big chunk of the software in the recipe is mine. The rest is about 25% percent of Dave Mills's reference implementation of NTP. USB jitter isn't too bad, actually. Unacceptable if you're doing pgysics experiments but an order of magitude below the expected accuracy of WAN time synchronization. That said, my recipe *is* better. And a fun, simple, dirt-cheap build. -- <a href="http://www.catb.org/~esr/">Eric S. Raymond</a>
On Jun 20, 2019, at 10:18 PM, Jay Hennigan <jay@west.net> wrote:
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
Not sure how accurate you need, but I just use a Raspberry Pi as a pool.ntp.org node. I thought about going the GPS route with it but didn’t want to mess with it. -Andy
Yep, went through the same route until I figured out that GPS time is a bit ahead of UTC. We simply use a windows NTP server for internal use at work, and I won't recommend doing so, because it went off the rails once for a while despite of having several upstream servers pointed to. Also there's a community out there dedicated for atomic clocks for civil use, it'd be fun to have one, set it up, and watch it ticking. On 6/21/2019 11:44 AM, Andy Ringsmuth wrote:
On Jun 20, 2019, at 10:18 PM, Jay Hennigan <jay@west.net> wrote:
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source? If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick. Not sure how accurate you need, but I just use a Raspberry Pi as a pool.ntp.org node. I thought about going the GPS route with it but didn’t want to mess with it.
-Andy
On 21 Jun 2019, at 10:57, Quan Zhou <quan@posteo.net> wrote:
Yep, went through the same route until I figured out that GPS time is a bit ahead of UTC.
The clocks on the GPS satellites are set to GPST which I think (I'm not a time geek so this is going to make someone cringe) is UTC without leap seconds or other corrections relating to rotation of the earth. However, the messages sent to GPS receivers include the offset between GPST and UTC as well as the GPST timestamp. The receivers can use both together to obtain a measure of UTC accurate to about 100 nanoseconds. Seems to me (again, not time geek, stop throwing things) that the use of GPST is an internal implementation detail chosen because it's easier to adjust an offset that rarely changes than it is to adjust atomic clocks floating in space. The system (including the system-internal adjustment of GPST with the offset) still produces a reasonably accurate measure of UTC. Also I imagine occasional leap seconds causing GPS navigators to jump spontaneously to the left which is probably more amusing in my imagination than in real life. Joe
On 6/21/19 07:57, Quan Zhou wrote:
Yep, went through the same route until I figured out that GPS time is a bit ahead of UTC.
The data from GPS includes the offset value from UTC for leap-second correction. This should be easily included in your time calculation. It's presently 18 seconds. -- Jay Hennigan - jay@west.net Network Engineering - CCIE #7880 503 897-8550 - WB6RDV
Once upon a time, Jay Hennigan <jay@west.net> said:
The data from GPS includes the offset value from UTC for leap-second correction. This should be easily included in your time calculation.
Not only that, but at least some GPS receivers/protocols notify of pending leap seconds, so software can properly distribute the notification in advance. -- Chris Adams <cma@cmadams.net>
I would submit that the proper use of a GPS receiver is for alignment of the start of the second to a more precise value than can be distributed across an asymmetric network like the Internet. The actual 'time label' for that second doesn't necessarily need to come from GPS at all. For security reasons, it's probably a good thing to make sure you validate the data received from GPS in any case. On Mon, Jun 24, 2019 at 8:23 PM Chris Adams <cma@cmadams.net> wrote:
Once upon a time, Jay Hennigan <jay@west.net> said:
The data from GPS includes the offset value from UTC for leap-second correction. This should be easily included in your time calculation.
Not only that, but at least some GPS receivers/protocols notify of pending leap seconds, so software can properly distribute the notification in advance.
-- Chris Adams <cma@cmadams.net>
-- - Forrest
Once upon a time, Forrest Christian (List Account) <lists@packetflux.com> said:
I would submit that the proper use of a GPS receiver is for alignment of the start of the second to a more precise value than can be distributed across an asymmetric network like the Internet. The actual 'time label' for that second doesn't necessarily need to come from GPS at all. For security reasons, it's probably a good thing to make sure you validate the data received from GPS in any case.
If you don't trust the GPS receiver's idea of the time, why do you trust its start of the second? It seems really odd to trust one and not the other. -- Chris Adams <cma@cmadams.net>
It's about minimizing the impact of the attack vector. And you shouldn't implicitly trust the second alignment either. In a potential spoofing attack, if you trust the GPS for all of the data exclusively, then someone who can spoof your GPS (not as hard/expensive as one would think) can fully control what time you think it is. This is obviously bad. If instead you take the time data from another source, and only take the second from the GPS, at most you're going to be off a second. This is less bad but still bad in some cases. Fortunately, we can easily do better than this. NTP itself provides the solution. Ideally you'd get your time from multiple sources and use some sort of algorithm to determine what the most likely correct time is. NTP has this functionality built in. If you take a stratum 2 or 3 server, and add multiple, geographically diverse stratum 1 and 2 servers to it, the stratum 2 or 3 server will look at all of the views of time including second alignment that it is receiving, and will determine which servers can be trusted and which can't. If a stratum 1 server is being spoofed, the stratum 2 or 3 server will notice that it is out of alignment and ignore it. In this way, you don't trust what is coming down the GPS of one or two stratum 1 servers. For most people just running a stratum 2 or 3 server with a well-curated set of stratum 1 or 2 servers scattered around the internet will be accurate enough, and will provide robust, not easily spoofed time. The limitation here is that this is limited by RTT/2 in the worst case, so if you're a long ways away from your closest stratum 1 server, your clock may be offset by up to RTT/2 plus whatever systemic errors are inherent in the stratum 1 server (cable delays, etc). If you need better alignment, a local stratum 1 server can be used, but it should just be added to your local stratum 2 or 3 server to improve the alignment of the second. Once this is added, the stratum 2 or 3 server will typically notice that it's really close and will start to follow it's second alignment, but only if it is within the window that it has determined is likely to be valid by the 'voting' of all of the other stratum 1 and 2 servers which are scattered around. One other note: There are some stratum 1 servers out there which do not generally rely on GPS for time transfer from their stratum 1 clocks. For instance, the NIST and USNO ntp servers, along with others around the world in various standards organizations. It might pay to include some of these in your mix as well. On Mon, Jun 24, 2019 at 8:36 PM Chris Adams <cma@cmadams.net> wrote:
Once upon a time, Forrest Christian (List Account) <lists@packetflux.com> said:
I would submit that the proper use of a GPS receiver is for alignment of the start of the second to a more precise value than can be distributed across an asymmetric network like the Internet. The actual 'time label' for that second doesn't necessarily need to come from GPS at all. For security reasons, it's probably a good thing to make sure you validate the data received from GPS in any case.
If you don't trust the GPS receiver's idea of the time, why do you trust its start of the second? It seems really odd to trust one and not the other. -- Chris Adams <cma@cmadams.net>
-- - Forrest
* jay@west.net (Jay Hennigan) [Fri 21 Jun 2019, 05:19 CEST]:
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
Have you tried this? Because I have, and it's absolutely terrible. GPS doesn't give you the correct time, it's supposed to give you a good 1pps clock discipline against which you can measure your device's internal clock and adjust accordingly for drift due to it not being Cesium-based, influenced by room temperature etc. You're unlikely to get the 1pps signal across USB, and even then there'll likely be significant latencies in the USB stack compared to the serial interface that these setups traditionally use. -- Niels.
On 2019-06-21 14:19, Niels Bakker wrote:
* jay@west.net (Jay Hennigan) [Fri 21 Jun 2019, 05:19 CEST]:
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
Have you tried this? Because I have, and it's absolutely terrible. GPS doesn't give you the correct time, it's supposed to give you a good 1pps clock discipline against which you can measure your device's internal clock and adjust accordingly for drift due to it not being Cesium-based, influenced by room temperature etc.
You're unlikely to get the 1pps signal across USB, and even then there'll likely be significant latencies in the USB stack compared to the serial interface that these setups traditionally use.
I think it depends on recipe you are using. Raspberry have low latency GPIO, and some receivers have 1pps output. https://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html
Denys Fedoryshchenko <nuclearcat@nuclearcat.com> wrote:
On 2019-06-21 14:19, Niels Bakker wrote:
Have you tried this? Because I have, and it's absolutely terrible. GPS doesn't give you the correct time, it's supposed to give you a good 1pps clock discipline against which you can measure your device's internal clock and adjust accordingly for drift due to it not being Cesium-based, influenced by room temperature etc.
You're unlikely to get the 1pps signal across USB, and even then there'll likely be significant latencies in the USB stack compared to the serial interface that these setups traditionally use.
I think it depends on recipe you are using. Raspberry have low latency GPIO, and some receivers have 1pps output. https://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html
And there are tricks for avoiding temperature-related deviations :-) https://blog.ntpsec.org/2017/02/01/heat-it-up.html Tony. -- f.anthony.n.finch <dot@dotat.at> http://dotat.at/ defend the right to speak, write, worship, associate, and vote freely
This. I've had some timing issues ( unrelated to NTP ) with certain combinations of FlightAware RTLSDR USB sticks and Pi models. IIRC USB and Ethernet share the same bus on the Pis, and that can cause bumps. GPIOs run right off the SOC, avoiding that. On Fri, Jun 21, 2019 at 7:25 AM Denys Fedoryshchenko < nuclearcat@nuclearcat.com> wrote:
On 2019-06-21 14:19, Niels Bakker wrote:
* jay@west.net (Jay Hennigan) [Fri 21 Jun 2019, 05:19 CEST]:
On 6/20/19 07:39, David Bass wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If you want to go really cheap and don't value your time, but do value knowing the correct time, a GPS receiver with a USB interface and a Raspberry Pi would do the trick.
Have you tried this? Because I have, and it's absolutely terrible. GPS doesn't give you the correct time, it's supposed to give you a good 1pps clock discipline against which you can measure your device's internal clock and adjust accordingly for drift due to it not being Cesium-based, influenced by room temperature etc.
You're unlikely to get the 1pps signal across USB, and even then there'll likely be significant latencies in the USB stack compared to the serial interface that these setups traditionally use.
I think it depends on recipe you are using. Raspberry have low latency GPIO, and some receivers have 1pps output. https://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html
On Thu, 20 Jun 2019 10:39:41 -0400, David Bass <davidbass570@gmail.com> wrote:
What are folks using these days for smaller organizations, that need to dole out time from an internal source?
If "internal" means a local NTP server independent of external network resources, the other responses are apposite. If "internal" means a stratum 2 NTP server dependent upon other servers on the network, and if you don't need accuracy in single-digit millisecond or better range, we've done well with an older ex-Windows machine that now runs FreeBSD and vanilla NTPD, and is now a pool server at ntp.org. ± six milliseconds, cost approaches the cube root of zero. mdr -- Sometimes half-ass is exactly the right amount of ass. -- Wonderella
participants (16)
-
Andy Ringsmuth -
Chris Adams -
David Bass -
Denys Fedoryshchenko -
Eric S. Raymond -
Forrest Christian (List Account) -
Jay Hennigan -
Joe Abley -
Mel Beckman -
Michael Rathbun -
Niels Bakker -
Patrick -
Quan Zhou -
Tom Beecher -
Tony Finch -
Warren Kumari