Re: [time-nuts] Linux time servers
M. Warner Losh wrote: I'd try the FreeBSD distribution of Linux. Thanks for the suggestions to use something else, but not an option. Like I say, I *have* fedora, I'm not planning to uninstall it as it was installed for me and I don't want the hassle. HI Dave (G0DJA) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 4a0d0fa0.3020...@tiscali.co.uk Dave Ackrill dave.g0...@tiscali.co.uk writes: : M. Warner Losh wrote: : : I'd try the FreeBSD distribution of Linux. : : Thanks for the suggestions to use something else, but not an option. : : Like I say, I *have* fedora, I'm not planning to uninstall it as it was : installed for me and I don't want the hassle. HI Then just install ntpd and be done with it. If you want sub-microsecond sync to a PPS, you'll be disappointed. If you want about 10ms over a LAN, then you'll be happy. Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Dave, Fedora will be fine as a basic network time server. If NTP isn't already installed on the system, all you have to do is type (as root): yum install ntp The default config usually queries the NTP Pool. Depending on where you live you might want to edit the file to use only your country's NTP Pool servers, or even some local servers that you know of. You can then set your windows machines on your network to query your fedora machine for time. If they can't connect then the firewall is probably blocking UDP/123. Jason ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
- Original Message - From: M. Warner Losh i...@bsdimp.com To: time-nuts@febo.com; dave.g0...@tiscali.co.uk Sent: Friday, May 15, 2009 3:02 AM Subject: Re: [time-nuts] Linux time servers In message: 4a0d0fa0.3020...@tiscali.co.uk Dave Ackrill dave.g0...@tiscali.co.uk writes: : M. Warner Losh wrote: : : I'd try the FreeBSD distribution of Linux. : : Thanks for the suggestions to use something else, but not an option. : : Like I say, I *have* fedora, I'm not planning to uninstall it as it was : installed for me and I don't want the hassle. HI Then just install ntpd and be done with it. If you want sub-microsecond sync to a PPS, you'll be disappointed. If you want about 10ms over a LAN, then you'll be happy. Warner You might consider switching to FreeBSD for more reasons than just timing, It's much faster than Fedora and I found the new 7.1 version easy to install. I have 30 some servers and switching all to FreeBSD. Phil ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
You might consider switching to FreeBSD for more reasons than just timing, It's much faster than Fedora and I found the new 7.1 version easy much faster in what respect? tested how? Folkert van Heusden -- www.vanheusden.com/multitail - multitail is tail on steroids. multiple windows, filtering, coloring, anything you can think of -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
2009/5/16 phil fort...@bellsouth.net - Original Message - From: M. Warner Losh i...@bsdimp.com To: time-nuts@febo.com; dave.g0...@tiscali.co.uk Sent: Friday, May 15, 2009 3:02 AM Subject: Re: [time-nuts] Linux time servers In message: 4a0d0fa0.3020...@tiscali.co.uk Dave Ackrill dave.g0...@tiscali.co.uk writes: : M. Warner Losh wrote: : : I'd try the FreeBSD distribution of Linux. : : Thanks for the suggestions to use something else, but not an option. : : Like I say, I *have* fedora, I'm not planning to uninstall it as it was : installed for me and I don't want the hassle. HI Then just install ntpd and be done with it. If you want sub-microsecond sync to a PPS, you'll be disappointed. If you want about 10ms over a LAN, then you'll be happy. Warner You might consider switching to FreeBSD for more reasons than just timing, It's much faster than Fedora and I found the new 7.1 version easy to install. I have 30 some servers and switching all to FreeBSD. For servers, possibly, but he may be running a workstation. 73, Steve -- Steve Rooke - ZL3TUV G8KVD JAKDTTNW A man with one clock knows what time it is; A man with two clocks is never quite sure. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Anyone got any good Linux time systems for PCs ? Make sure you have a recent kernel installed with the PPS patch applied: remote refid st t when poll reach delay offset jitter == *127.127.20.1.GPS.0 l5 16 3770.000 -0.014 0.006 That system is only up for 1 day so the offset (and jitter) will get better later on. Typically the offset outputted by ntpq is around 0.001 and the jitter 0.000. This is with a Garmin 18LVC. http://wiki.enneenne.com/index.php/LinuxPPS_support I also have a DCF77 receiver connected to a system: remote refid st t when poll reach delay offset jitter == *127.127.8.0 .DCFa. 0 l 14 64 3770.000 -2.193 7.711 and an MSF receiver (using my http://vanheusden.com/lpc-ntpd/lindy_precision_clock.php driver), it is a lindy precision clock connected via USB: remote refid st t when poll reach delay offset jitter == x127.127.28.0.SHM0. 2 l 58 6470.000 11.120 6.601 Folkert van Heusden -- To MultiTail einai ena polymorfiko ergaleio gia ta logfiles kai tin eksodo twn entolwn. Prosferei: filtrarisma, xrwmatismo, sygxwneysi, diaforetikes provoles. http://www.vanheusden.com/multitail/ -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 20090515142600.gg2...@vanheusden.com Folkert van Heusden folk...@vanheusden.com writes: : You might consider switching to FreeBSD for more reasons than just : timing, It's much faster than Fedora and I found the new 7.1 version easy : : much faster in what respect? tested how? The usual benchmark that's cited here is the mysql tps scaling better than Linux. See for example http://people.freebsd.org/~kris/scaling/mysql.html. The numbers in this paper are a little dated (being for 7.0 and a little over a year old), they show good scaling. Of course, this isn't the place to debate that in extreme detail (for example, there are other pages I can't find right now that show newer versions of Linux, some better, some worse as changes to the scheduler help or hurt performance). It is no longer the case that you can automatically assume Linux performs better. You have to measure things and make sure you use the system that best matches your performance requirements. Also, on a 7.1R system, you need to make sure that you enable tagged queueing. A last minute change botched it. 7.2R is out now too. Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
: You might consider switching to FreeBSD for more reasons than just : timing, It's much faster than Fedora and I found the new 7.1 version easy : : much faster in what respect? tested how? The usual benchmark that's cited here is the mysql tps scaling better than Linux. See for example http://people.freebsd.org/~kris/scaling/mysql.html. The numbers in this paper are a little dated (being for 7.0 and a little over a year old), they show good scaling. Of course, this isn't the place to debate that in extreme detail (for example, there are other pages I can't find right now that show newer versions of Linux, some better, some worse as changes to the scheduler help or hurt performance). It is no longer the case that you can automatically assume Linux performs better. You have to measure things and make sure you use the system that best matches your performance requirements. Also, on a 7.1R system, you need to make sure that you enable tagged queueing. A last minute change botched it. 7.2R is out now too. You can't say that freebsd is faster than linux; you specifically need to specify what version of freebsd and what version of linux you're using. Also the hardware platform matters as well as the compiler (and version) used to compile mysql and numerous other parameters. What would be interesting is how a specific linux-kernel with the pps patches by rodolpho compare to a specific freebsd version with the same ntpd compiled using the same gcc and such. Folkert van Heusden -- Feeling generous? - http://www.vanheusden.com/wishlist.php -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 20090515152610.gi2...@vanheusden.com Folkert van Heusden folk...@vanheusden.com writes: : : You might consider switching to FreeBSD for more reasons than just : : timing, It's much faster than Fedora and I found the new 7.1 version easy : : : : much faster in what respect? tested how? : : The usual benchmark that's cited here is the mysql tps scaling better : than Linux. See for example : http://people.freebsd.org/~kris/scaling/mysql.html. The numbers in : this paper are a little dated (being for 7.0 and a little over a : year old), they show good scaling. Of course, this isn't the place to : debate that in extreme detail (for example, there are other pages I : can't find right now that show newer versions of Linux, some better, : some worse as changes to the scheduler help or hurt performance). It : is no longer the case that you can automatically assume Linux performs : better. You have to measure things and make sure you use the system : that best matches your performance requirements. : Also, on a 7.1R system, you need to make sure that you enable tagged : queueing. A last minute change botched it. 7.2R is out now too. : : You can't say that freebsd is faster than linux; you specifically need : to specify what version of freebsd and what version of linux you're : using. Also the hardware platform matters as well as the compiler (and : version) used to compile mysql and numerous other parameters. I didn't say FreeBSD was faster than Linux. Please read what I said carefully (note, the quotes stuff at the top isn't me). I said for some work loads, it is faster. : What would be interesting is how a specific linux-kernel with the pps : patches by rodolpho compare to a specific freebsd version with the same : ntpd compiled using the same gcc and such. Yes. All my evaluations of Linux pre-date this patch. However, even with the patch, I can still say that Linux performs worse than FreeBSD on NTP because the patch hasn't been committed to the kernel.org tree. I guess this is the difference between Linux can be made to perform better with this out-of-tree patch and Out of the box, Linux performs well. When FreeBSD switched from gcc 3.x to gcc 4.x, I did measurements of the ability of the kernel to track a PPS (also changes with the major revision of the kernel). I found that there was no measurable difference between the different FreeBSD kernels I tested despite being built with a number of different compilers (3.4.5, 4.2.0 and 4.1). It turns out that the algorithms for steering the time aren't dependent on how fast the results are computed, but rather dependent on the results being computed correctly. I will admit that my testing of Linux was been rather cursory over the years compared to the attention I've given to FreeBSD. Of course, we're mixing up problems a little bit here. The ntpd with pps performance issue is somewhat different than the claims another writer was making about FreeBSD being faster for his servers... Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message 20090515152610.gi2...@vanheusden.com, Folkert van Heusden writes: What would be interesting is how a specific linux-kernel with the pps patches by rodolpho compare to a specific freebsd version with the same ntpd compiled using the same gcc and such. I think first of all, it would be interesting what your quality parameters are... Prescision ? Resolution ? Reliability ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message 20090515.095406.-1739011364@bsdimp.com, M. Warner Losh write s: When FreeBSD switched from gcc 3.x to gcc 4.x, I did measurements of the ability of the kernel to track a PPS (also changes with the major revision of the kernel). I found that there was no measurable difference between the different FreeBSD kernels I tested despite being built with a number of different compilers (3.4.5, 4.2.0 and 4.1). It turns out that the algorithms for steering the time aren't dependent on how fast the results are computed, but rather dependent on the results being computed correctly. [...] ... which you can read more about in my paper from 2002: http://phk.freebsd.dk/pubs/timecounter.pdf Poul-Henning -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
What would be interesting is how a specific linux-kernel with the pps patches by rodolpho compare to a specific freebsd version with the same ntpd compiled using the same gcc and such. I think first of all, it would be interesting what your quality parameters are... Prescision ? Resolution ? Reliability ? Reliability is, I think, for me the most important. E.g. that it doesn't jitter all over the place and that the jitter is also as constant as possible. Folkert van Heusden -- MultiTail na wan makriki wrokosani fu tan luku den logfile nanga san den commando spiti puru. Piki puru spesrutu sani, wroko nanga difrenti kroru, tya kon makandra, nanga wan lo moro. http://www.vanheusden.com/multitail/ -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message 20090515161538.gk2...@vanheusden.com, Folkert van Heusden writes: What would be interesting is how a specific linux-kernel with the pps patches by rodolpho compare to a specific freebsd version with the same ntpd compiled using the same gcc and such. I think first of all, it would be interesting what your quality parameters are... Prescision ? Resolution ? Reliability ? Reliability is, I think, for me the most important. E.g. that it doesn't jitter all over the place and that the jitter is also as constant as possible. That is not reliability, that is precision. Reliability means that it keeps answering. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Anyone got any good Linux time systems for PCs ? Make sure you have a recent kernel installed with the PPS patch applied: remote refid st t when poll reach delay offset jitter == *127.127.20.1.GPS.0 l5 16 3770.000 -0.014 0.006 That system is only up for 1 day so the offset (and jitter) will get better later on. Typically the offset outputted by ntpq is around 0.001 and the jitter 0.000. This is with a Garmin 18LVC. The _sore_ thing is that with 2.4.x-nano kernels offsets/jitter got that good within minutes. Now many years later it takes days... :-( -- Björn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
What would be interesting is how a specific linux-kernel with the pps patches by rodolpho compare to a specific freebsd version with the same ntpd compiled using the same gcc and such. I think first of all, it would be interesting what your quality parameters are... Prescision ? Resolution ? Reliability ? Reliability is, I think, for me the most important. E.g. that it doesn't jitter all over the place and that the jitter is also as constant as possible. Folkert van Heusden For my daytime use of ntp, convergence in less than 5 minutes is essential. Due to some unusual peripherials, that often have drivers dictating a certain kernel version its is often impossible to find a linux kernel version that support both the proprietary drivers and decent ntp-server performance. -- Björn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
The _sore_ thing is that with 2.4.x-nano kernels offsets/jitter got that good within minutes. Now many years later it takes days... :-( I think that's a bug introduced by the tickless scheduler work. For my daytime use of ntp, convergence in less than 5 minutes is essential. Due to some unusual peripherials, that often have drivers dictating a certain kernel version its is often impossible to find a linux kernel version that support both the proprietary drivers and decent ntp-server performance. Do you build your own kernels? It looks like there is a simple change that fixes the convergence. http://lkml.org/lkml/2009/5/4/373 There is also the TSC calibration mess. If you are only running on one machine, or one type of system, you can patch the kernel to use a hard-coded answer. Or you can try one of the non-TSC clock sources. -- These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
The _sore_ thing is that with 2.4.x-nano kernels offsets/jitter got that good within minutes. Now many years later it takes days... :-( I think that's a bug introduced by the tickless scheduler work. For my daytime use of ntp, convergence in less than 5 minutes is essential. Due to some unusual peripherials, that often have drivers dictating a certain kernel version its is often impossible to find a linux kernel version that support both the proprietary drivers and decent ntp-server performance. Do you build your own kernels? It looks like there is a simple change that fixes the convergence. http://lkml.org/lkml/2009/5/4/373 There is also the TSC calibration mess. If you are only running on one machine, or one type of system, you can patch the kernel to use a hard-coded answer. Or you can try one of the non-TSC clock sources. Thanks! That might be good news after a looong dark night. -- Björn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] Linux time servers
Anyone got any good Linux time systems for PCs ? I now have a PC on my home system that has Linux fedora on it and I'm keen to learn how to make it a useful new member of my network. I did dabble with Redhat Linux once before in the 1990s, and still have the scars to show for it, so please don't assume that I know what I'm doing... Thanks Dave (G0DJA) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
I now have a PC on my home system that has Linux fedora on it and I'm keen to learn how to make it a useful new member of my network. It's probably already running ntpd and setup to get time from a few pool machines out on the net. Start by doing: ntpq -p If that works, look in /etc/ntp.conf The official ntp documentation starts as html files. There may be scripts that turn them into man pages. They may be out of date, misleading or missing a lot of critical info. The html stuff may get installed on your system. If not, be sure to get the html pages that correspond to the version of ntpd that you are running. If you want better time, you need a local source of time. ntpd calls them refclocks. If you want seriously good time, you need the PPS mods to the kernel. I've lost track of where they come from and/or what the current status is. -- These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
RedHat in the 90s was terrible. It's much better now. Last thing I read about ntp was that it was kind of broken for high precision stuff on Linux and people tend to use FreeBSD. I duplicated the work of one of the time-nuts by following his site here: http://www.febo.com/pages/soekris/ Even with all of the details there it still required a great effort on my part to get things up and running to where I have them now. If you decide to go this route I will be more than happy to send you a copy of the image on my CF card so that you'll have a working system out of the box. I had to recompile ntp because the current FreeBSD distro didn't have support for something (NMEA I think, of all things!). -Bob On Thu, May 14, 2009 at 2:09 PM, Dave Ackrill dave.g0...@tiscali.co.ukwrote: Anyone got any good Linux time systems for PCs ? I now have a PC on my home system that has Linux fedora on it and I'm keen to learn how to make it a useful new member of my network. I did dabble with Redhat Linux once before in the 1990s, and still have the scars to show for it, so please don't assume that I know what I'm doing... Thanks Dave (G0DJA) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 4a0c7a74.50...@tiscali.co.uk Dave Ackrill dave.g0...@tiscali.co.uk writes: : Anyone got any good Linux time systems for PCs ? : : I now have a PC on my home system that has Linux fedora on it and I'm : keen to learn how to make it a useful new member of my network. : : I did dabble with Redhat Linux once before in the 1990s, and still have : the scars to show for it, so please don't assume that I know what I'm : doing... I'd try the FreeBSD distribution of Linux. :) Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Is there any consensus for the reasons why Linux performs poorly? I was thinking about setting up a server as well (possibly using a little ARM-based single-board computer that runs Linux). Randy. --- On Thu, 5/14/09, Robert Darlington rdarling...@gmail.com wrote: From: Robert Darlington rdarling...@gmail.com Subject: Re: [time-nuts] Linux time servers To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: Thursday, May 14, 2009, 4:23 PM RedHat in the 90s was terrible. It's much better now. Last thing I read about ntp was that it was kind of broken for high precision stuff on Linux and people tend to use FreeBSD. I duplicated the work of one of the time-nuts by following his site here: http://www.febo.com/pages/soekris/ Even with all of the details there it still required a great effort on my part to get things up and running to where I have them now. If you decide to go this route I will be more than happy to send you a copy of the image on my CF card so that you'll have a working system out of the box. I had to recompile ntp because the current FreeBSD distro didn't have support for something (NMEA I think, of all things!). -Bob On Thu, May 14, 2009 at 2:09 PM, Dave Ackrill dave.g0...@tiscali.co.ukwrote: Anyone got any good Linux time systems for PCs ? I now have a PC on my home system that has Linux fedora on it and I'm keen to learn how to make it a useful new member of my network. I did dabble with Redhat Linux once before in the 1990s, and still have the scars to show for it, so please don't assume that I know what I'm doing... Thanks Dave (G0DJA) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
At 05:24 PM 5/14/2009, M. Warner Losh wrote... I'd try the FreeBSD distribution of Linux. Ouch. In some circles, those are fightin' words. FreeBSD is _not_ Linux, in any way except being Unix-like. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Is there any consensus for the reasons why Linux performs poorly? I was thinking about setting up a server as well (possibly using a little ARM-based single-board computer that runs Linux). Consensus? I doubt it. My reading. Lots of cooks. None of them are time geeks. There are a lot of people working on Linux. A lot of them are smart. A lot of them are not plugged into the culture of key chunks of technology so they fix or clean up some code in ways that actually breaks things. There are only a few big screwups that are on my list these days. No PPS support. The TSC calibration code is broken. (and it's the default mode) The in-kernel NTP support code is broken. The last two work, just not quite correctly. They are close enough so that you probably won't notice any problems unless you are a geek. One of the things that is driving some of the changes is making things work better for low power applications. The old scheduler used to do a bit of work every clock tick (100 HZ to 1000 HZ). That chews up a lot of power if your battery powered system goes to sleep when there is nothing to do. So it seems reasonable to look ahead in the scheduler queue and figure out how long until the next time there is work to do and sleep until then. -- These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 20090514220030.68e47b...@ip-64-139-1-69.sjc.megapath.net Hal Murray hmur...@megapathdsl.net writes: : : Is there any consensus for the reasons why Linux performs poorly? I : was thinking about setting up a server as well (possibly using a : little ARM-based single-board computer that runs Linux). : : Consensus? I doubt it. : : My reading. Lots of cooks. None of them are time geeks. Many of these cooks have firmly held, strong views on how things should be done, which also gets in the way of having a reasonable discussion about why those views hurt time performance. At least that's been my experience. On FreeBSD you have phk@ and I which works better... : There are a lot of people working on Linux. A lot of them are smart. A lot : of them are not plugged into the culture of key chunks of technology so they : fix or clean up some code in ways that actually breaks things. Many times the fixes neglect edge cases, or dismiss the need to get them right at all (like: who cares if the system time is off by a second, ntpd will steer that out). : There are only a few big screwups that are on my list these days. : No PPS support. : The TSC calibration code is broken. (and it's the default mode) : The in-kernel NTP support code is broken. : : The last two work, just not quite correctly. They are close enough so that : you probably won't notice any problems unless you are a geek. True enough... If you are, you are way too much about it, but if you don't it doesn't bother you at all... : One of the things that is driving some of the changes is making : things work better for low power applications. The old scheduler : used to do a bit of work every clock tick (100 HZ to 1000 HZ). That : chews up a lot of power if your battery powered system goes to sleep : when there is nothing to do. So it seems reasonable to look ahead : in the scheduler queue and figure out how long until the next time : there is work to do and sleep until then. The problem is that these can't easily be turned off... The other problem that the tickless stuff starts to expose is that many of these platforms have counters that can be used for time keeping, but they wrap too quickly to sleep for long... Anyway, I'm totally biased on this stuff, so you should take me with a grain of salt. Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
I'd try the FreeBSD distribution of Linux. Ouch. In some circles, those are fightin' words. I interpreted it as a joke, like telling a Windows user to install Service Pack Linux. If all you want is to run a time server, FreeBSD will do a better job than Linux. In particular, the Soekris boxes are polular. If you have a Linux box that you need/use for other stuff, it can also run ntpd. That may or may not be good enough at timekeeping for a time-nut. -- These are my opinions, not necessarily my employer's. I hate spam. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
Assuming you are looking for better time than you can get from just network servers, you can add a local time source. EASY WAY If you want good time at low cost, and have a generic PC, connect something like a Garmin 18x LVC to a serial port and install FreeBSD. Easy to configure and NTPD has reported jitter in the 1 to 4 microsecond range ever since. Total elapsed time maybe 2 hours including installing FreeBSD and mounting the GPS on the roof. The FreeBSD folks do a very good job of supporting ntpd with local hardware clock sources, and document it well. HARD WAY Linux with PPS is still more of a work-in-progress and you should expect to have to patch kernels etc. You will find dozens of kernel postings, 1PPS seems to have ended up as a serial line protocol, AFAIK it is still not in any distribution base kernel. If you are doing an ARM you will likely have to modify the ARM's serial driver to do timestamping, by looking at samples in the 8250 driver PPS code. Certainly quite do-able, but not in 2 hours for most people :) -Bob p.s. It was not obvious from the Garmin instructions, but all you have to do is crimp the serial wires on the 18x LVC to a 9 pin connector, placing the yellow 1 PPS wire on pin 1, plug into the PC's serial port and FreeBSD does the rest. Avoid any PC without an actual serial port. You can pick up +5v from your PC (use a fuse). On May 14, 2009, at 2:34 PM, Randy Scott wrote: Is there any consensus for the reasons why Linux performs poorly? I was thinking about setting up a server as well (possibly using a little ARM-based single-board computer that runs Linux). Randy. --- On Thu, 5/14/09, Robert Darlington rdarling...@gmail.com wrote: From: Robert Darlington rdarling...@gmail.com Subject: Re: [time-nuts] Linux time servers To: Discussion of precise time and frequency measurement time- n...@febo.com Date: Thursday, May 14, 2009, 4:23 PM RedHat in the 90s was terrible. It's much better now. Last thing I read about ntp was that it was kind of broken for high precision stuff on Linux and people tend to use FreeBSD. I duplicated the work of one of the time-nuts by following his site here: http://www.febo.com/pages/soekris/ Even with all of the details there it still required a great effort on my part to get things up and running to where I have them now. If you decide to go this route I will be more than happy to send you a copy of the image on my CF card so that you'll have a working system out of the box. I had to recompile ntp because the current FreeBSD distro didn't have support for something (NMEA I think, of all things!). -Bob On Thu, May 14, 2009 at 2:09 PM, Dave Ackrill dave.g0...@tiscali.co.ukwrote: Anyone got any good Linux time systems for PCs ? I now have a PC on my home system that has Linux fedora on it and I'm keen to learn how to make it a useful new member of my network. I did dabble with Redhat Linux once before in the 1990s, and still have the scars to show for it, so please don't assume that I know what I'm doing... Thanks Dave (G0DJA) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/ listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/ time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Linux time servers
In message: 20090514223310.0d552b...@ip-64-139-1-69.sjc.megapath.net Hal Murray hmur...@megapathdsl.net writes: : I'd try the FreeBSD distribution of Linux. : Ouch. In some circles, those are fightin' words. : : I interpreted it as a joke, like telling a Windows user to install Service : Pack Linux. Yes. It is a joke. One of the market share firms (netcraft?) had this as one of the choices for a questions in ~1998 that went something like: Which distribution of Linux do you run redhat suse debian freebsd : If all you want is to run a time server, FreeBSD will do a better job than : Linux. In particular, the Soekris boxes are polular. : : If you have a Linux box that you need/use for other stuff, it can also run : ntpd. That may or may not be good enough at timekeeping for a time-nut. ntpd works on Linux, it just doesn't perform as well as FreeBSD. But the difference in performance typically is in the millisecond range or 10ms range (depending on the kernel version). For most people, this is more than good enough. For subscribers to time-nuts... :) Warner ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.