[ntp:questions] Decreasing Frequency Error in Linux 2.6
Hello, I'm currently running an NTP server using a Garmin 18x LVC as a stratum-0 time source. Before I installed Linux, I used FreeBSD. Its frequency error was quite low, usually around 1 ppm. This allowed me to have stable, desirable times such as: remote refid st t when poll reach delay offset jitter == +navobs1.gatech. .GPS.1 u 32 64 377 58.966 2.622 5.055 +ntp-s1.cise.ufl .GPS.1 u 18 64 377 69.260 2.001 1.998 *GPS_NMEA(1) .PPS.0 l5 16 3770.000 -0.001 0.002 Even under heavy load, the offset for FreeBSD always seemed to be under 20 microseconds. Now, I'm trying to run a similar time server with Linux 2.6 and Rudalfo's LinuxPPS patch (or shmpss or gpsd, doesn't really matter to me). I've installed Debian Etch (4.0) stable, and recompiled the testing 2.6.28 Linux kernel with Debian patches and the newest LinuxPPS patch. At the moment, I'm using the ATOM driver in ntpd and am experiencing undesirable frequency error and offsets. According to today's drift file, Min offset: -68 microseconds, Max offset: 112 microseconds. Frequency ppm Min: -18.1570, Max: -17.5150. a sample ntpq output is remote refid st t when poll reach delay offset jitter == +navobs1.gatech. .GPS.1 u 21 64 377 57.631 1.882 2.206 oPPS(0) .PPS.1 l 10 16 3770.000 -0.018 0.001 Jitter is often as high as 0.006. It appears there is no difference between LinuxPPS, shmpps, and gpsd for these timings. Also, this was not tested under load (besides any normal load for a debian machine). So, does anyone have any suggestions for reducing frequency error in Debian (or the linux kernel)? Also, is there anyway to keep timings low while producing load? I'd like to run other things on this server, but not at the expense of time precision. Thanks for all the support this group has provided me so far via other posts, Chris ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
In article 9bca36-5p@mail.specsol.com j...@specsol.spam.sux.com writes: Unruh unruh-s...@physics.ubc.ca wrote: j...@specsol.spam.sux.com writes: Have you never heard of calling ntpdate before starting the NTP daemon? uh, ntpdate is severely depricated, and ntpd -g is what is supposed to be used. If ntpd -g fails it is a bug. Uhh, lots of mainline 'nix's don't have a -g option to ntpd and still have ntpdate, e.g. Solaris 10. FWIW, and still quite irrelevant to the original question, while the version shipped with Solaris 10 (xntpd actually) is indeed based on ancient code, it does have a -g option: # uname -sr SunOS 5.10 # /usr/lib/inet/xntpd -v /usr/lib/inet/xntpd: option requires argument -v usage: /usr/lib/inet/xntpd [ -abdgmx ] [ -c config_file ] [ -e e_delay ] [ -f freq_file ] [ -k key_file ] [ -l log_file ] [ -p pid_file ] [ -r broad_delay ] [ -s statdir ] [ -t trust_key ] [ -v sys_var ] [ -V default_sysvar ] It's not documented, and may not do exactly what the current version of -g does, but something pretty close. --Per Hedeland p...@hedeland.org ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] interpreting the allan deviation graph of the ntp loopstats-file
Hi, Somewhere I read that one can evaluate their ntp-server by creating an allan deviation graph of the loopstats-file. Now I succeeded in creating them: http://keetweej.vanheusden.com/stats/allan-deviation-plots/ using this octave script: http://www.wraith.sf.ca.us/ntp/allan.oct But: now what? What do these graphs tell me? Folkert van Heusden -- Multitail es una herramienta flexible que permite visualizar los log file y seguir la ejecución de comandos. Permite filtrar, añadir colores, combinar archivos, la visualización de diferencias (diff- view), etc. http://www.vanheusden.com/multitail/ -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
David J Taylor david-tay...@blueyonder.neither-this-bit.nor-this-bit.co.uk wrote: j...@specsol.spam.sux.com wrote: [] You do understand there are lots of environments where it takes an act of God to be allowed to replace vendor utilities with self compiled versions, don't you? Not a problem with Windows, fortunately. G David What planet do you people live on? I have one client that will not even allow Windows critical security updates to be installed until a extensive formal test is done to prove the updates won't effect operations. This is hardly a unique operation. -- Jim Pennino Remove .spam.sux to reply. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
j...@specsol.spam.sux.com wrote: [] What planet do you people live on? I have one client that will not even allow Windows critical security updates to be installed until a extensive formal test is done to prove the updates won't effect operations. This is hardly a unique operation. Well, I can appreciate that makes sense, as I have known some updates to affect normal operations, although in those circumstances I used to recommend separate PCs for Internet operations. I don't recall ever seeing a report of NTP causing problems with normal operations. David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
I don't recall ever seeing a report of NTP causing problems with normal operations. Sorry for being anal on this but: Well almost: certain versions of the linux kernel under certain specific conditions would panic when ntp introduces a leap second. http://markmail.org/message/dhm5byrbfcarpiet?q=leap+second+list:org.kernel.vger.linux-kernel Ok it's a bug in the linux kernel but it is only triggered by ntp. Folkert van Heusden -- Multi tail barnamaj mowahib li mora9abat attasjilat wa nataij awamir al 7asoub. damj, talwin, mora9abat attarchi7 wa ila akhirih. http://www.vanheusden.com/multitail/ -- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
David J Taylor david-tay...@blueyonder.neither-this-bit.nor-this-bit.co.uk wrote: j...@specsol.spam.sux.com wrote: [] What planet do you people live on? I have one client that will not even allow Windows critical security updates to be installed until a extensive formal test is done to prove the updates won't effect operations. This is hardly a unique operation. Well, I can appreciate that makes sense, as I have known some updates to affect normal operations, although in those circumstances I used to recommend separate PCs for Internet operations. I don't recall ever seeing a report of NTP causing problems with normal operations. David Nor would I ever expect to. The point is LOTS of places have extensive procedures in place that must be followed before any software on production systems can be changed, including applying vendor supplied and recommended patches. While I have free reign to do anything I want with my systems, such is not the case for many of my client's systems. -- Jim Pennino Remove .spam.sux to reply. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
Folkert van Heusden wrote: I don't recall ever seeing a report of NTP causing problems with normal operations. Sorry for being anal on this but: Well almost: certain versions of the linux kernel under certain specific conditions would panic when ntp introduces a leap second. http://markmail.org/message/dhm5byrbfcarpiet?q=leap+second+list:org.kernel.vger.linux-kernel Ok it's a bug in the linux kernel but it is only triggered by ntp. Folkert van Heusden Thanks for that. I do wish it were easier to test these leap-second issues. David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
j...@specsol.spam.sux.com wrote: [] The point is LOTS of places have extensive procedures in place that must be followed before any software on production systems can be changed, including applying vendor supplied and recommended patches. While I have free reign to do anything I want with my systems, such is not the case for many of my client's systems. Oh, indeed, but I might ask why my client was still running, or had chosen to install in the first place, such outdated versions of software, before taking on those systems. But that's life, I suppose. Cheers, David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] ntpd IPv6 support on Windows?
I am running Meinberg's Win32 binaries on Windows Server 2003: ntpd 4.2@beijing-o Sep 01 9:15:56 (UTC+02:00) 2008 (10) It appears to me there is no support for IPv6 in these binaries. I'm guessing the work simply hasn't been done in the reference implementation to support IPv6 on Windows. Can anyone confirm that even if I build my own binaries from source, IPv6 on Windows is not supported? Incidentally, their free Windows NTP Time Server Monitor by Meinberg v1.03a is a nice piece of work. I use the NTP status tab instead of scripted ntpq -c peers in a loop, and their loopstats graphing makes it easy to see how well your timeserver is performing. If you run ntp on a platform other than Windows but can access the filesystem via a mapped network drive, you can use the Statistics tab to graph its performance by clicking the triple-dot ... button and navigating to the directory holding the loopstats file. You may need to enable stats gathering in ntp.conf first: statsdir /path/to/store/loopstats/files/ statistics loopstats Cheers, Dave Hart ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
David J Taylor david-tay...@blueyonder.neither-this-bit.nor-this-bit.co.uk wrote: j...@specsol.spam.sux.com wrote: [] The point is LOTS of places have extensive procedures in place that must be followed before any software on production systems can be changed, including applying vendor supplied and recommended patches. While I have free reign to do anything I want with my systems, such is not the case for many of my client's systems. Oh, indeed, but I might ask why my client was still running, or had chosen to install in the first place, such outdated versions of software, before taking on those systems. But that's life, I suppose. Cheers, David Solaris 10 Update 6 IS the latest release of Solaris and the provided NTP is nowhere near the latest downloadable version of NTP. I would have to check, but I am pretty sure the same is true for HP-UX. Not everyone runs Linux nor do they usually choose a OS for a rather obscure feature like NTP. To get back to the original topic, it seems to me if one really cares about absolute time, having just one hardware clock is not a very robust solution no matter what OS or version of NTP is being used. I have one client that does care and has GPS timeserver appliances at several sites with all the sites using all the NTP servers as potential sources so failing a major catastrophe at all sites, there is redundancy. -- Jim Pennino Remove .spam.sux to reply. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
Terje Mathisen wrote: Andy Helten wrote: Unruh wrote: uh, ntpdate is severely depricated, and ntpd -g is what is supposed to be used. If ntpd -g fails it is a bug. Then it is a bug because, as previously mentioned, no command line argument or tinker can disable this behavior. I suppose the solution is to skip the CLOSETIME check in clocktime() when '-g' is specified. Andy, I agree totally: With no remote reference clocks, only local hardware, said hardware has to skip the sanity checks during that initial -g adjustment. Andy, you should go on the ntp.org site and register this as a bug! Terje Before creating a bug report I searched the database for CLOSETIME and found a bug report had already been opened. The status is NEW, which seemed to imply someone following this email discussion beat me to the punch. However, the existing bug report was opened in April of 2005. There is some back-and-forth discussion between the developers (including Dave Mills), but clearly there was no consensus on how it needs to be fixed and there has been no discussion since August 2005. There were some patches submitted that either did not address the exact problem we are discussing here or the patches were never accepted. Not sure what the protocol is for this situation, but it appears the original bug report needs a bump: https://support.ntp.org/bugs/show_bug.cgi?id=417 Andy ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] My extra second ...
The linux kernel bug aparently was found and fixed: related thread: http://lkml.org/lkml/2009/1/2/276 bug analysis: http://lkml.org/lkml/2009/1/2/373 fix: http://lkml.org/lkml/2009/1/2/415 Antonio M. Moreiras. Bill Unruh escreveu: On Sat, 3 Jan 2009, Danny Mayer wrote: Unruh wrote: George R. Kasica geor...@netwrx1.com writes: On Fri, 02 Jan 2009 14:31:34 +0100, Rob van der Putten r...@sput.nl wrote: Hi there Steve Kostecke wrote: All my Linux systems had a fine time. None of them locked up / crashed / rebooted / etc. The kernels involved included: 2.6.24-etchnhalf.1-amd64 2.6.18-5-486 2.6.16-2-486 2.6.18-5-k7 2.6.18-4-powerpc 2.4.16-k7 What about the hardware (Intel /AMD)? Rob: Everything in my post above was Intel based, no AMD or otherwise. Why in the world would you then have an amd64 kerenl and two k7 kernels, and one powerpc kernel? None of those is intel. He *never* said that. Those are Steve's systems. The post even includes that fact. Ooops. My mistake-- not reading th properly Danny ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP sever on an isolated Network
Hi Harlan. Sorry for the late reply, but we had a problem with our modem and we got a tough time to find another one compatible with my hardware. I tried what you suggest : server 127.127.18.1 minpoll 12 maxpoll 17 phone atdt913034944774 atdt917195676742 The problem is different now. The ntp.log says: *refclock_open /dev/acts1: no such file or directory.* I made a symlink of /dev/acts1 to my /dev/modem, but I still get this error. Is there anything that I'm missing? Thanks. Diego On Thu, Nov 13, 2008 at 8:55 PM, Harlan Stenn st...@ntp.org wrote: Diego, Have you seen: https://support.ntp.org/bin/view/Support/ConfiguringModemRefclocks or https://support.ntp.org/bin/view/Support/ModemRefclockUsers -- Harlan Stenn st...@ntp.org http://ntpforum.isc.org - be a member! ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] ntpd IPv6 support on Windows?
Dave Hart wrote: I am running Meinberg's Win32 binaries on Windows Server 2003: ntpd 4.2@beijing-o Sep 01 9:15:56 (UTC+02:00) 2008 (10) It appears to me there is no support for IPv6 in these binaries. I'm guessing the work simply hasn't been done in the reference implementation to support IPv6 on Windows. Can anyone confirm that even if I build my own binaries from source, IPv6 on Windows is not supported? Not yet. It's on my list of things to complete in the near future. There are two pieces that need to be changed and one of them may be tricky though I seem to remember that Martin had put something together to help with this since we need to support people without IPv6 support and don't have getaddrinfo() and friends available. getaddinfo() is necessary for IPv6 support. Now's the time to get that in. Danny ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] interpreting the allan deviation graph of the ntp loopstats-file
Folkert van Heusden wrote: Hi, Somewhere I read that one can evaluate their ntp-server by creating an allan deviation graph of the loopstats-file. Now I succeeded in creating them: http://keetweej.vanheusden.com/stats/allan-deviation-plots/ using this octave script: http://www.wraith.sf.ca.us/ntp/allan.oct But: now what? What do these graphs tell me? Folkert van Heusden You need to review the research documents on the ntp.org site or the Mills book on NTP for details. You can do a google search for allan deviation and specify site:ntp.org along with the search string. Danny ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] ntpd IPv6 support on Windows?
On Tue, Jan 6, 2009 at 7:19 PM, Danny Mayer ma...@ntp.isc.org wrote to questions@lists.ntp.org and me: Not yet. It's on my list of things to complete in the near future. There are two pieces that need to be changed and one of them may be tricky though I seem to remember that Martin had put something together to help with this since we need to support people without IPv6 support and don't have getaddrinfo() and friends available. getaddinfo() is necessary for IPv6 support. Now's the time to get that in. It sounds like late-binding the required syscalls/APIs with LoadModule and GetProcAddress is in order. One way to minimize the source disturbance when late-binding is to provide your own wrapper implementations for each function and use a macro to redirect ntp's calls to the conditionally-available functions through the corresponding runtime-binding wrappers. In this case Microsoft has already done the legwork, including support for the Windows 2000 IPv6 download (which implements these in a different DLL than WinXP and later). Quoting from their getaddrinfo documentation at http://msdn.microsoft.com/en-us/library/ms738520(VS.85).aspx begin quote The getaddrinfo function was added to the Ws2_32.dll on Windows XP and later. To execute an application that uses this function on earlier versions of Windows (Windows 2000, Windows NT, and Windows Me/98/95), then you need to include the Ws2tcpip.h and Wspiapi.h files. When the Wspiapi.h include file is added, the getaddrinfo function is defined to the WspiapiGetAddrInfo inline function in the Wspiapi.h file. At runtime, the WspiapiGetAddrInfo function is implemented in such a way that if the Ws2_32.dll or the Wship6.dll (the file containing getaddrinfo in the IPv6 Technology Preview for Windows 2000) does not include getaddrinfo, then a version of getaddrinfo is implemented inline based on code in the Wspiapi.h header file. This inline code will be used on older Windows platforms that do not natively support the getaddrinfo function. The IPv6 protocol is supported on Windows 2000 when the IPv6 Technology Preview for Windows 2000 is installed. Otherwise getaddrinfo support on versions of Windows earlier than Windows XP is limited to handling IPv4 name resolution. end quote That sounds pretty darned painless, I hope it is indeed that easy for you. Now, ask me why I care about IPv6 when it'll apparently be years before my IPv6 service isn't tunnelled over IPv4 using one lovely hack or another guaranteed to add overhead and latency. I don't have a good answer. I just want to break and enter into a few ipmasq and NAT pioneers' homes and disable every telephone ringer. Dave Hart ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Why can't clocks do inital synchronization?
Folkert van Heusden wrote: I don't recall ever seeing a report of NTP causing problems with normal operations. Sorry for being anal on this but: Well almost: certain versions of the linux kernel under certain specific conditions would panic when ntp introduces a leap second. http://markmail.org/message/dhm5byrbfcarpiet?q=leap+second+list:org.kernel.vger.linux-kernel Ok it's a bug in the linux kernel but it is only triggered by ntp. I just saw something about this in the NTP WG from the linux kernel wranglers who seem to be totally confused about NTP. They should fix the kernel bug and not start talking about TAI. Danny Folkert van Heusden ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions