[opensuse] ntp and wireless
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!) If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp. -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
-- Roger Oberholtzer
Roger - if you get a good answer to your question I will be interested also. That has been my biggest complaint also about the Network Manager, is that it does not make networks available until after booting up is complete. I not only need ntp myself, but also must automount a number of network devices during bootup for a number of scripts I have custom made, and my backup (Bacula) client needs the network available during boot up also. So I configure the network management to use the traditional ifup and ifdown methods instead. Those commands do work during boot up. I far prefer graphical tools, so I don't have to remember command line arguments and a well designed GUI will guide the user to the solution(s) he/she is looking for. But some GUI based applications, like Network Manager, need to be designed to be robust enough to work even when the user cannot interact with them, such as during boot up time, and I am not sure that it has (at least I have not been able to figure out how either)... Marc Chamberlin -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Marc Chamberlin pecked at the keyboard and wrote:
Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
-- Roger Oberholtzer
Roger - if you get a good answer to your question I will be interested also. That has been my biggest complaint also about the Network Manager, is that it does not make networks available until after booting up is complete.
Actually not until someone logs in. It is exactly this crock of crap why I also use ifup on all my systems. -- Ken Schneider SuSe since Version 5.2, June 1998 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 28 October 2009 19:48:53 Ken Schneider - openSUSE wrote:
Actually not until someone logs in. It is exactly this crock of crap why I also use ifup on all my systems.
If you have any connections configured in YaST before switching to NM, NM brings these up when it starts before login. nm-system-settings is responsible for reading the config files in /etc/sysconfig/network and telling NetworkManager about them. Otherwise it's a bug. -- The Crock Cleaner -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Will Stephenson wrote:
On Wednesday 28 October 2009 19:48:53 Ken Schneider - openSUSE wrote:
Actually not until someone logs in. It is exactly this crock of crap why I also use ifup on all my systems.
If you have any connections configured in YaST before switching to NM, NM brings these up when it starts before login. nm-system-settings is responsible for reading the config files in /etc/sysconfig/network and telling NetworkManager about them. Otherwise it's a bug.
-- The Crock Cleaner
Will - I think you are missing the point of this thread. Even if the NetworkManager does start up the connections before login, the problem is that it does NOT start them up early enough in the boot up process. An early startup of the network connections is needed so that other services which are also starting up, during boot, and dependent on having the network connections up and running, can also start. NTP is just one of several daemons that need access to the network, and if it is not available then it fails to start up. So far as I know, ifup is the only method that works for these scenarios. So IMHO the crock is still quite dirty! Marc Chamberlin -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 28 October 2009 21:00:29 Marc Chamberlin wrote:
Will - I think you are missing the point of this thread. Even if the NetworkManager does start up the connections before login, the problem is that it does NOT start them up early enough in the boot up process. An early startup of the network connections is needed so that other services which are also starting up, during boot, and dependent on having the network connections up and running, can also start. NTP is just one of several daemons that need access to the network, and if it is not available then it fails to start up.
So far as I know, ifup is the only method that works for these scenarios. So IMHO the crock is still quite dirty!
I see your point, but in theory dispatcher.d scripts should hup or restart these services after NM brings up the network (since the /etc/init.d/network script does not wait for NM to complete this, unlike when it is in ifup mode). Also in theory (mine at least) you shouldn't really be using static network mounts or running servers on NetworkManager machines as I doubt that those dispatcher.d scripts, or the services they start/stop, get enough testing to be robust enough in the face of all the dynamism that NetworkManager introduces. Consider a slow-starting server that might not be finished starting in the time between NM bringing a network connection up, and bringing it down again or changing the default route because another network connection it prefers comes up. I'd be sticking to 'desktoppy' network services (SMB ioslaves/GIO) that have more chance of being adapted to NM's signals. Anyway, I'm just the NM client developer here, not its maintainer. I may be wrong. Will -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Will Stephenson wrote:
On Wednesday 28 October 2009 21:00:29 Marc Chamberlin wrote:
Will - I think you are missing the point of this thread. Even if the NetworkManager does start up the connections before login, the problem is that it does NOT start them up early enough in the boot up process. An early startup of the network connections is needed so that other services which are also starting up, during boot, and dependent on having the network connections up and running, can also start. NTP is just one of several daemons that need access to the network, and if it is not available then it fails to start up.
So far as I know, ifup is the only method that works for these scenarios. So IMHO the crock is still quite dirty!
I see your point, but in theory dispatcher.d scripts should hup or restart these services after NM brings up the network (since the /etc/init.d/network script does not wait for NM to complete this, unlike when it is in ifup mode).
Also in theory (mine at least) you shouldn't really be using static network mounts or running servers on NetworkManager machines as I doubt that those dispatcher.d scripts, or the services they start/stop, get enough testing to be robust enough in the face of all the dynamism that NetworkManager introduces. Consider a slow-starting server that might not be finished starting in the time between NM bringing a network connection up, and bringing it down again or changing the default route because another network connection it prefers comes up. I'd be sticking to 'desktoppy' network services (SMB ioslaves/GIO) that have more chance of being adapted to NM's signals.
Anyway, I'm just the NM client developer here, not its maintainer. I may be wrong.
Will
Aw! So there is a bit of documentation/insight that I wish I had, when asked to choose which method I wanted to use (in YaST) to manage my network connections! So if I am reading you right, then it appears the goal of the NetworkManager will be to manage network connections in the face of a dynamic network. Somewhat akin to what Windoz Network Neighborhood does now. And yes I would dearly love to be able to browse dynamic networks like I can from Windoz (although I want something far more robust and reliable!!!) I agree that trying to start and stop, or restart, some services too quickly will cause some of them to fail. (Tomcat immediately comes to mind! LOL) So... let me propose a Grand Unified Network Management project! One that can handle both server systems as well as desktoppy systems! Uh maybe I should leave now before a shoe hits me in the face.... Marc Chamberlin -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 28 October 2009 10:10:34 am Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
-- Roger Oberholtzer
There are 2 problem with the current SuSE wireless setup. (1) starting ntp before the wireless link is up, and (2) shutting down the wireless link before the unmount of smbfs. As you point out, since ntp is started (and fails) before your wireless link is activated by network manager, ntp is dead for most laptop configurations. The workaround in to restart ntp after your wireless link comes up with 'rcntp restart' (as root or sudo). The smbfs unmount issue causes shutdown and reboots to hang if you have smb shares mounted due to the wireless connection being shut down before the shares are unmounted leaving no way to unmount the smb shares -- and prompting a 180 or 300 second timeout period before your box will shut down. (I haven't timed it exactly, but I know it is long...) I think both are probably the result of network manager (1) starting the wireless link later (when kde is started), and (2) shutting down the wireless link earlier (I don't know when this occurs exactly) in the current setup where wireless is handled by network manager. Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown. That's what I have found by trial and error. -- David C. Rankin, J.D.,P.E. Rankin Law Firm, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 Telephone: (936) 715-9333 Facsimile: (936) 715-9339 www.rankinlawfirm.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
David C. Rankin wrote:
On Wednesday 28 October 2009 10:10:34 am Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
-- Roger Oberholtzer
There are 2 problem with the current SuSE wireless setup. (1) starting ntp before the wireless link is up, and (2) shutting down the wireless link before the unmount of smbfs. As you point out, since ntp is started (and fails) before your wireless link is activated by network manager, ntp is dead for most laptop configurations. The workaround in to restart ntp after your wireless link comes up with 'rcntp restart' (as root or sudo).
The smbfs unmount issue causes shutdown and reboots to hang if you have smb shares mounted due to the wireless connection being shut down before the shares are unmounted leaving no way to unmount the smb shares -- and prompting a 180 or 300 second timeout period before your box will shut down. (I haven't timed it exactly, but I know it is long...)
I think both are probably the result of network manager (1) starting the wireless link later (when kde is started), and (2) shutting down the wireless link earlier (I don't know when this occurs exactly) in the current setup where wireless is handled by network manager.
Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown.
That's what I have found by trial and error.
I just checked my notebook and see that ntpd is running with a WiFi connection. The clock appears to be in sync with my desktop computer, which is also running ntpd. Using "ps aux|grep ntp" shows similar results on both computers. Unless I'm mistaken, ntpd is running properly on my notebook. I'm running OpenSUSE 11.0 on both computers and NetworkManager on my notebook. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
James Knott wrote:
I just checked my notebook and see that ntpd is running with a WiFi connection. The clock appears to be in sync with my desktop computer, which is also running ntpd. Using "ps aux|grep ntp" shows similar results on both computers. Unless I'm mistaken, ntpd is running properly on my notebook. I'm running OpenSUSE 11.0 on both computers and NetworkManager on my notebook.
I just checked the ntpd log and it shows it's setting the time correctly. The last sync was about 2h 28 m ago. This is on the notebook running OpenSUSE 11.0 and connecting via WiFi and using NetworkManager. No problem here. Also, looking back through the log, I can see times when it couldn't find the ntp server and used the local clock instead. Later it again shows using the configured ntp server. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday, 2009-10-28 at 15:02 -0400, James Knott wrote:
I just checked the ntpd log and it shows it's setting the time correctly. The last sync was about 2h 28 m ago. This is on the notebook running OpenSUSE 11.0 and connecting via WiFi and using NetworkManager. No problem here.
Also, looking back through the log, I can see times when it couldn't find the ntp server and used the local clock instead. Later it again shows using the configured ntp server.
That's what it should do. However, if there is a big initial time difference, it will abort. If you start the ntp service after the network is up, first it will jump-adjust the time, then it starts the daemon to keep it. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkrol9gACgkQtTMYHG2NR9WFIwCeMGn39uXxnb19zHy3zT+wklmE 2iMAn0LB9inpdABdngcHNDlVrZfYKxCh =jfrm -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Carlos E. R. wrote:
On Wednesday, 2009-10-28 at 15:02 -0400, James Knott wrote:
I just checked the ntpd log and it shows it's setting the time correctly. The last sync was about 2h 28 m ago. This is on the notebook running OpenSUSE 11.0 and connecting via WiFi and using NetworkManager. No problem here.
Also, looking back through the log, I can see times when it couldn't find the ntp server and used the local clock instead. Later it again shows using the configured ntp server.
That's what it should do. However, if there is a big initial time difference, it will abort. If you start the ntp service after the network is up, first it will jump-adjust the time, then it starts the daemon to keep it.
One ntpd option is "-g" which will allow it to ignore a big time difference. According to ps aux, that is enabled, so that big error shouldn't be a problem, at least once. A possibility to fix this would be to have a script started a few minutes after boot, to ensure ntpd is running. There are a couple of ways that could be done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday, 2009-10-28 at 15:45 -0400, James Knott wrote:
Carlos E. R. wrote:
That's what it should do. However, if there is a big initial time difference, it will abort. If you start the ntp service after the network is up, first it will jump-adjust the time, then it starts the daemon to keep it.
One ntpd option is "-g" which will allow it to ignore a big time difference. According to ps aux, that is enabled, so that big error shouldn't be a problem, at least once.
Yes... but not quite. A few minutes difference can take hours (even days) to get corrected by that method, because what it does is speed up or slow down the clock just a trifle, so that timing operations in the running processes are not affected (noticeably). When the clock catches with the real world time, it readjusts the speed.
A possibility to fix this would be to have a script started a few minutes after boot, to ensure ntpd is running. There are a couple of ways that could be done.
Things like this are designed to work with the "traditional" setup, and do work. Graphical tools in linux have some catch up to do. That is, if you have a network and want some things like ntp, nfs, samba, etc, to come up and down properly, just use the traditional network setup. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkropQoACgkQtTMYHG2NR9VFewCdHU0JdbartQJ++XuPIlHKqO+M 32gAoJB7YFCXXMvXR4HoQEu3HmCZ6xmv =dGlP -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 28 October 2009 01:52:21 pm James Knott wrote:
David C. Rankin wrote:
On Wednesday 28 October 2009 10:10:34 am Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
-- Roger Oberholtzer
There are 2 problem with the current SuSE wireless setup. (1) starting ntp before the wireless link is up, and (2) shutting down the wireless link before the unmount of smbfs. As you point out, since ntp is started (and fails) before your wireless link is activated by network manager, ntp is dead for most laptop configurations. The workaround in to restart ntp after your wireless link comes up with 'rcntp restart' (as root or sudo).
The smbfs unmount issue causes shutdown and reboots to hang if you have smb shares mounted due to the wireless connection being shut down before the shares are unmounted leaving no way to unmount the smb shares -- and prompting a 180 or 300 second timeout period before your box will shut down. (I haven't timed it exactly, but I know it is long...)
I think both are probably the result of network manager (1) starting the wireless link later (when kde is started), and (2) shutting down the wireless link earlier (I don't know when this occurs exactly) in the current setup where wireless is handled by network manager.
Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown.
That's what I have found by trial and error.
I just checked my notebook and see that ntpd is running with a WiFi connection. The clock appears to be in sync with my desktop computer, which is also running ntpd. Using "ps aux|grep ntp" shows similar results on both computers. Unless I'm mistaken, ntpd is running properly on my notebook. I'm running OpenSUSE 11.0 on both computers and NetworkManager on my notebook.
James, Mine too has started acting OK over the past few months: [00:08 alchemy:/srv/www/download] # rcntp status remote refid st t when poll reach delay offset jitter ============================================================================== LOCAL(0) .LOCL. 10 l 5 64 377 0.000 0.000 0.001 *ntp3.tamu.edu 206.77.62.136 2 u 61 1024 377 28.940 -4.798 0.391 Checking for network time protocol daemon (NTPD): running I'm now sure when or where. On the other had, the CIFS unmount hang occurs every time I lease a drive mounted...:( -- David C. Rankin, J.D.,P.E. Rankin Law Firm, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 Telephone: (936) 715-9333 Facsimile: (936) 715-9339 www.rankinlawfirm.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
David C. Rankin wrote:
Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown.
That's what I have found by trial and error.
What would be nice is if NetworkManager supported running a script on connection and perhaps on disconnection too. This feature was in the modem dialer IIRC. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 28 October 2009 20:08:38 James Knott wrote:
David C. Rankin wrote:
Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown.
That's what I have found by trial and error.
What would be nice is if NetworkManager supported running a script on connection and perhaps on disconnection too. This feature was in the modem dialer IIRC.
See /etc/NetworkManager/dispatcher.d/. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Will Stephenson wrote:
On Wednesday 28 October 2009 20:08:38 James Knott wrote:
David C. Rankin wrote:
Two work-arounds. (1) Use traditional ifup for networking instead of network manager (good choice unless you use a lot of different APs), or (2) remember to restart ntp and unmount smb drives before you logout/shutdown.
That's what I have found by trial and error.
What would be nice is if NetworkManager supported running a script on connection and perhaps on disconnection too. This feature was in the modem dialer IIRC.
See /etc/NetworkManager/dispatcher.d/.
I've looked at what's in there. Next question is how to make use of it. If I wanted to restart ntpd, how would I go about it? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wed, 2009-10-28 at 20:28 +0100, Will Stephenson wrote:
See /etc/NetworkManager/dispatcher.d/.
Ahh. I think this is the thing I am looking for. I think a script like this might work: #! /bin/sh case "$2" in up) rcntp start ;; down) rcntp stop ;; *) exit 0 ;; esac Note, I have not tried it yet. But I will shortly -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wed, 2009-10-28 at 22:26 +0100, Roger Oberholtzer wrote:
On Wed, 2009-10-28 at 20:28 +0100, Will Stephenson wrote:
See /etc/NetworkManager/dispatcher.d/.
Ahh. I think this is the thing I am looking for. I think a script like this might work:
#! /bin/sh case "$2" in up) rcntp start ;; down) rcntp stop ;; *) exit 0 ;; esac
Note, I have not tried it yet. But I will shortly
I have tried this, and it does not seem to work. I tried: 1. run the script by hand to be sure there is nothing odd there. It runs as expected. 2. With ntp running, turn off wireless via NetworkManager. (This is what the icon in my KDE system tray is eventually talking to, right?) I expected the ntp daemon to go away. Nope. It is still running. 3. With ntp not running, connect to my wireless access point. I expected the ntp daemon to be started. Nope. It is not running. I read the man page, and the script permissions are as it says (and the same as the scripts that are already there). So, either the kde system tray icon is not connected to this, or something is amiss. This is with openSUSE 11.2 RC1 As suggested elsewhere, I can of course do by hand: rcntp start|stop|restart But that is not the point. -- Roger Oberholtzer OPQ Systems / Ramböll RST Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden Office: Int +46 8-615 60 20 Mobile: Int +46 70-815 1696 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2009-10-29 at 08:34 +0100, Roger Oberholtzer wrote: ...
I have tried this, and it does not seem to work.
The script would need to run as root, to be able to start services. Does NetworkManager run as root? :-? - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkrpTr4ACgkQtTMYHG2NR9WS/wCfUfqPuOfWEiYL/sZ3cYXd1rJx rJsAn0krRpi9gFAdrHxUxNv7oYhQCYoU =PBCU -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Carlos E. R. wrote:
I have tried this, and it does not seem to work.
The script would need to run as root, to be able to start services. Does NetworkManager run as root? :-?
I don't know, but it's simple enough to use sudo to run a script, as I have done for other functions. Also, I just started up this computer and here's the ntpd log: 29 Oct 11:09:45 ntpd[2367]: host name not found: time.nrc.ca 29 Oct 11:09:45 ntpd[2367]: couldn't resolve `time.nrc.ca', giving up on it 29 Oct 11:12:54 ntpd[2364]: synchronized to LOCAL(0), stratum 10 29 Oct 11:12:54 ntpd[2364]: time slew +0.000000 s 29 Oct 11:12:55 ntpd[3174]: Deleting interface #2 lo, ::1#123, interface stats: received=0, sent=0, dropped=0, active_time=1 secs 29 Oct 11:12:55 ntpd[3174]: Deleting interface #3 WiFi, fe80::20e:35ff:fe4c:dfb8#123, interface stats: received=0, sent=0, dropped=0, active_time=1 secs 29 Oct 11:16:10 ntpd[3174]: synchronized to LOCAL(0), stratum 10 29 Oct 11:16:10 ntpd[3174]: kernel time sync status change 0001 29 Oct 11:17:15 ntpd[3174]: synchronized to 132.246.168.164, stratum 2 I booted the computer around 11:09 and logged in at 11:12 and 5 minutes after doing so, ntpd was talking to the server via WiFi. At 11:09, I hadn't yet logged in and the WiFi link not yet established. At that time, ntpd could not find the server, as expected. Based on this, it appears ntpd works fine with NetworkManager, at least on this computer. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Thu, 2009-10-29 at 11:29 -0400, James Knott wrote:
I booted the computer around 11:09 and logged in at 11:12 and 5 minutes after doing so, ntpd was talking to the server via WiFi. At 11:09, I hadn't yet logged in and the WiFi link not yet established. At that time, ntpd could not find the server, as expected. Based on this, it appears ntpd works fine with NetworkManager, at least on this computer.
Yesterday I ran ntp by hand after NetworkManager started, because it had not been starting. That was seen because the time has bee off by over an hour since I installed 11.2 RC1. I have added ntp to the NetworkManager scripts. Yesterday I reported that fiddling with NetworkManager from the KDE system tray had no effect. The script seemed not to run. I have now rebooted, and ntp is running. I do not know who started it. As the only change from all my previous boots where ntp did not start are listed above, one of them must be causing my improved behavior. Next reboot, I will try without the NetworkManager script. That will tell it is is that, or at least one successful ntp run first was the trigger. -- Roger Oberholtzer Ramböll RST/OPQ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden Office: Int +46 8-615 60 20 Mobile: Int +46 70-815 1696 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2009-10-29 at 23:17 +0100, Roger Oberholtzer wrote:
On Thu, 2009-10-29 at 11:29 -0400, James Knott wrote:
I booted the computer around 11:09 and logged in at 11:12 and 5 minutes after doing so, ntpd was talking to the server via WiFi. At 11:09, I hadn't yet logged in and the WiFi link not yet established. At that time, ntpd could not find the server, as expected. Based on this, it appears ntpd works fine with NetworkManager, at least on this computer.
Yesterday I ran ntp by hand after NetworkManager started, because it had not been starting. That was seen because the time has bee off by over an hour since I installed 11.2 RC1.
Exactly what I said would happen. NTP runs "happily" with no network, but it quits, aborts, when it finds the time difference is too big. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkrqIrUACgkQtTMYHG2NR9U1PwCfWqGbtdxzloSlxXoNgEuzS0LX MbUAmwacuKk7hcK8I2dzG13AgZqMGVBm =KOYz -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, 2009-10-30 at 00:18 +0100, Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Thursday, 2009-10-29 at 23:17 +0100, Roger Oberholtzer wrote:
On Thu, 2009-10-29 at 11:29 -0400, James Knott wrote:
I booted the computer around 11:09 and logged in at 11:12 and 5 minutes after doing so, ntpd was talking to the server via WiFi. At 11:09, I hadn't yet logged in and the WiFi link not yet established. At that time, ntpd could not find the server, as expected. Based on this, it appears ntpd works fine with NetworkManager, at least on this computer.
Yesterday I ran ntp by hand after NetworkManager started, because it had not been starting. That was seen because the time has bee off by over an hour since I installed 11.2 RC1.
Exactly what I said would happen. NTP runs "happily" with no network, but it quits, aborts, when it finds the time difference is too big.
I have yet to confirm this is what happened. I suspect it is. My test/verification will be this evening. I was busy fiddling with a new mobile last night... -- Roger Oberholtzer OPQ Systems / Ramböll RST Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden Office: Int +46 8-615 60 20 Mobile: Int +46 70-815 1696 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, 2009-10-30 at 00:18 +0100, Carlos E. R. wrote:
Exactly what I said would happen. NTP runs "happily" with no network, but it quits, aborts, when it finds the time difference is too big.
It must be the presence of both factors at the same time (no network, difference too big). After I had the network, ntp happily corrected the big difference. -- Roger Oberholtzer OPQ Systems / Ramböll RST Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden Office: Int +46 8-615 60 20 Mobile: Int +46 70-815 1696 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Am Mittwoch, 28. Oktober 2009 16:10:34 schrieb Roger Oberholtzer:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
https://bugzilla.novell.com/show_bug.cgi?id=253399 It should work even without having a network up and running while booting. If not you should comment on the bug(s). Maybe it is an update interval issue. https://bugzilla.novell.com/show_bug.cgi?id=549903 Sven -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Roger Oberholtzer wrote:
I have a laptop that uses wireless controlled by Network Manager. (11.2 RC1 - and it seems to be working again!)
If I want to keep time in sync with ntp, it seems that the network must be up when booting. Which is not the case with Network Manager. How does one get ntp working when using Network Manager. I suspect there are other network services that do not function. But I am interested specifically in ntp.
Maybe not really a cure to your problems, but I'm using the standard ifup methods, and then SCPM for different networks - I need that anyhow because I do start/stop quite some daemons depending in which network I am and SCPM doeas all that automagically ;^> Pit -- Dr. Peter "Pit" Suetterlin http://www.astro.su.se/~pit Institute for Solar Physics Tel.: +34 922 405 590 (Spain) P.Suetterlin@royac.iac.es +46 8 5537 8534 (Sweden) Peter.Suetterlin@astro.su.se -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Why do you need ntp? Is the system clock severely sloppy about keeping correct time? Is the CMOS clock a P.O.S.? Is there something critical that requires zero delta time with some other system(s)? Or is it so there is an excuse to neglect the system and CMOS clocks? If you just want the clock to be adjusted on startup or network connection, netdate or rdate will do the same thing. Since they're not daemons, they will not take away any system resources. I seem to recall ntpd blocking if it tried to reach timeservers when the network was down or didn't yet exist, among other problems that prompted me to use rdate and netdate instead. A script to periodically test the network connection and run netdate or rdate when the network is up works just fine. The script can also be run under cron at some suitably long time interval to update system time. The script can also run hwclock to set the CMOS clock as well. I used to have it on some boxen with P.O.S. CMOS clocks but the newer ones have very accurate clocks (approx +/-0.2sec/month) so it's only necessary to adjust clocks every couple months or so. So all ntp would do is tie up resources that could better be used for other things. ntp was created because of the need to keep system clocks accurately synchronised on a network for some time-critical applications. Most people do not need it. Even the author discouraged it's use. There was once an easy to follow adjtime tutorial that made it easy even for n00bs to adjust the system clock accurately. Anyone know where it went to? (It's not in the HOW-TO.) == jd Celebrate Hannibal Day this year. Take an elephant to lunch. -- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 j debert wrote:
Why do you need ntp?
Synchronising time has been a pain since the early rail services needed to ensure that trains ran on an identifiable schedule, missing your connection or being on one of two trains trying to use same track at the same time would ruin anyone's day. The early British rail companies instigated a common time across their rail networks because local times were so variable and effectively created the first time zone to try and eliminate this cause of such problems. Certain network OSs and services do have operations that tend to be time critical in a similar manner, and use the OS clock rather than the hardware clock (which is just a simple time counter with little additional info) for timestamping operations. The important relative time is then the network time and status, not the local hardware clock time, or even actual local time for that matter. Keeping the OS clock in sync with the network time becomes more important than being in sync with the hardware clock in this case.
Or is it so there is an excuse to neglect the system and CMOS clocks?
To be honest have not bothered much with these in years because except in the initial boot and standalone scenario they are now largely irrelevant in any networked environment. see... http://www.linuxselfhelp.com/quick/clock.html for one take on the clock, and... http://tldp.org/HOWTO/Clock-2.html for something different
A script to periodically test the network connection and run netdate or rdate when the network is up works just fine. The script can also be run under cron at some suitably long time interval to update system time.
Hmm... isnt this what ntpd does, occasionally check the upstream server and initiate a resync process if things have drifted too far. Using cron is effectively making time synchronisation process dependant on another time related process. It is not a particularly resource intensive daemon anyway.
ntp was created because of the need to keep system clocks accurately synchronised on a network for some time-critical applications. Most people do not need it. Even the author discouraged it's use.
Netdate uses ntp protocols to synchronise with a ntp server. (as does rdate)... so.... see above - -- ============================================================================== I have always wished that my computer would be as easy to use as my telephone. My wish has come true. I no longer know how to use my telephone. Bjarne Stroustrup ============================================================================== -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/ iEYEARECAAYFAkrsG6sACgkQasN0sSnLmgLwvgCfcwDPxtkP747O3jdlBjpJjrY5 IlcAn0uj48TU9pzWtaBd/nVyRcbmw1Bv =d+NN -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Sat, 2009-10-31 at 11:12 +0000, G T Smith wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
j debert wrote:
Why do you need ntp?
There are many reasons, such as: To be a member of a windows active directory, you need precise time (compared to the AD controller). That is a big part of how permissions in AD are maintained. A few minutes difference between the client and AD controller are enough to make authentication fail. -- You can't just ask customers what they want and then try to give that to them. By the time you get it built, they'll want something new. -- Steve Jobs Roger Oberholtzer Ramböll RST/OPQ Ramböll Sverige AB Krukmakargatan 21 P.O. Box 17009 SE-104 62 Stockholm, Sweden Office: Int +46 8-615 60 20 Mobile: Int +46 70-815 1696 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Roger Oberholtzer wrote:
On Sat, 2009-10-31 at 11:12 +0000, G T Smith wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
j debert wrote:
Why do you need ntp?
There are many reasons, such as:
To be a member of a windows active directory, you need precise time (compared to the AD controller). That is a big part of how permissions in AD are maintained. A few minutes difference between the client and AD controller are enough to make authentication fail.
There are also a lot of things when the time of a transaction is important. For example, imagine a 911 call center. Or a stock trading system etc. Not to mention not missing a show on TV!!! ;-) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Saturday, 2009-10-31 at 11:12 -0000, G T Smith wrote: ...
Certain network OSs and services do have operations that tend to be time critical in a similar manner, and use the OS clock rather than the hardware clock (which is just a simple time counter with little additional info) for timestamping operations. The important relative time is then the network time and status, not the local hardware clock time, or even actual local time for that matter. Keeping the OS clock in sync with the network time becomes more important than being in sync with the hardware clock in this case.
The CMOS clock was not part of the original PC, it was an add-on card invented later. I think they still use the same chip, or an equivalent. And read access to it was (is?) slow, so it can not be used for timing operations.
Or is it so there is an excuse to neglect the system and CMOS clocks?
To be honest have not bothered much with these in years because except in the initial boot and standalone scenario they are now largely irrelevant in any networked environment.
If the CMOS clock fails, the filesystem check at boot may trigger, and waste your time :-p
A script to periodically test the network connection and run netdate or rdate when the network is up works just fine. The script can also be run under cron at some suitably long time interval to update system time.
Hmm... isnt this what ntpd does, occasionally check the upstream server and initiate a resync process if things have drifted too far. Using cron is effectively making time synchronisation process dependant on another time related process.
It does more than that. It also adjusts (slowly) the speed of the clock, for example, and never "jumps" the time, so that processes that rely on the clock are not affected during the adjustments.
It is not a particularly resource intensive daemon anyway.
ntp was created because of the need to keep system clocks accurately synchronised on a network for some time-critical applications. Most people do not need it. Even the author discouraged it's use.
Netdate uses ntp protocols to synchronise with a ntp server. (as does rdate)... so....
But the clock is jumped, and the speed is not altered. It is not the same method. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkrs10MACgkQtTMYHG2NR9WOWQCgkpqKBDVRb4u0BPel899Ioqus 1FQAoIHMaV70cmrrxNJjNMQEkTIaHO6X =0EQO -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Carlos E. R. wrote:
On Saturday, 2009-10-31 at 11:12 -0000, G T Smith wrote:
The CMOS clock was not part of the original PC, it was an add-on card invented later. I think they still use the same chip, or an equivalent. And read access to it was (is?) slow, so it can not be used for timing operations.
The IBM PC (1981) did not have a CMOS hardware clock, but from the PC/AT (1984) onwards it was standard. The situation with intel based non-IBM kit (it did exist in 1981) is a bit more complex (but I do remember powering up such early machines without needing to enter the date). The IBM machines were technically solid (in more ways than one :-) ), but not necessarily bleeding edge for the time.
Hmm... isnt this what ntpd does, occasionally check the upstream server and initiate a resync process if things have drifted too far. Using cron is effectively making time synchronisation process dependant on another time related process.
It does more than that. It also adjusts (slowly) the speed of the clock, for example, and never "jumps" the time, so that processes that rely on the clock are not affected during the adjustments.
Should have put either graduated or controlled before resync.. it also IIRC allows for things like network latency (which is a bit problematic to handle well with a sequence of unrelated one off time checks)...
ntp was created because of the need to keep system clocks accurately
Netdate uses ntp protocols to synchronise with a ntp server. (as does rdate)... so....
But the clock is jumped, and the speed is not altered. It is not the same method.
Me being pedantic :-) ntp the protocol is not quite the same as ntp(d) the program(s). As for clock speed alteration I do recall people telling me they used sequences of Null OP instructions to ensure accurate timing in some circumstances (!??)... Depends really what you mean by changing the clock speed, changing the hardware clock speed itself could have unfortunate consequences, changing the translation between cpu or clock ticks and calender increments is another matter. I assume the latter is meant, but I cannot find any direct reference to this in the ntp (the program) docs...
-- Cheers, Carlos E. R.
- -- ============================================================================== I have always wished that my computer would be as easy to use as my telephone. My wish has come true. I no longer know how to use my telephone. Bjarne Stroustrup ============================================================================== -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/ iEYEARECAAYFAkrus4gACgkQasN0sSnLmgIKoACgn6zgc8fVQ31bx4tVvf2xc3dQ AuEAoKaijRw/vP1n/LIYzPCF2pFhXoRN =umVs -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Hi! Am Montag 02 November 2009 schrieb G T Smith:
As for clock speed alteration I do recall people telling me they used sequences of Null OP instructions to ensure accurate timing in some circumstances (!??)... Depends really what you mean by changing the clock speed, changing the hardware clock speed itself could have unfortunate consequences
Nowadays this happens all the time. Therefore such practice should be phased out anyways. And yes, programs relying on a constant clock speed have problems with that. You can observer that in Unreal Tournament 99, e.g. Regards, Matthias -- Matthias Bach www.marix.org „Der einzige Weg, die Grenzen des Möglichen zu finden, ist ein klein wenig über diese hinaus in das Unmögliche vorzustoßen.“ - Arthur C. Clarke
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Monday, 2009-11-02 at 10:25 -0000, G T Smith wrote:
Carlos E. R. wrote:
On Saturday, 2009-10-31 at 11:12 -0000, G T Smith wrote:
The CMOS clock was not part of the original PC, it was an add-on card invented later. I think they still use the same chip, or an equivalent. And read access to it was (is?) slow, so it can not be used for timing operations.
The IBM PC (1981) did not have a CMOS hardware clock, but from the PC/AT (1984) onwards it was standard. The situation with intel based non-IBM kit (it did exist in 1981) is a bit more complex (but I do remember powering up such early machines without needing to enter the date). The IBM machines were technically solid (in more ways than one :-) ), but not necessarily bleeding edge for the time.
I think that the AT added the external clock card, but putting the chip in the mainboard instead (the chip had a small cmos memory maintained with a small battery... used for keeping bios config as well). But not all machines at the time had that, I think, I remember that the commands date and time where typically at the start of the autoexec.bat, and most people ignored it, which meant that all files were dated Jan 1 1980. ...
As for clock speed alteration I do recall people telling me they used sequences of Null OP instructions to ensure accurate timing in some circumstances (!??)... Depends really what you mean by changing the clock speed, changing the hardware clock speed itself could have unfortunate consequences, changing the translation between cpu or clock ticks and calender increments is another matter. I assume the latter is meant, but I cannot find any direct reference to this in the ntp (the program) docs...
Yes, I remember using that timing method. It worked very well... in the eighties. Turbo Pascal used it in the "delay" routine. Right at program start it measured one tick of the timer in noops loops, and recorded a constant for time adjustment. At some version (6? 7? around 1998?) the variable that held the initial count overflowed, or wrapped back, causing many problems in problems already written. There were patches for the run time library, and also a method to patch released binaries. I guess that borland c had a similar problem, but I dunno. But it surprises me a lot to think that people may be using a method of that kind to time anything nowdays. Perhaps for microseconds :-? - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.9 (GNU/Linux) iEYEARECAAYFAkrvU1sACgkQtTMYHG2NR9Vb4QCfVMaz+yzBrSjqDJV/jPepBO7h pbwAn3nuAcquwlnkhR8iwVIvtqGEe58o =Rt6Q -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
j debert wrote:
Why do you need ntp?
For instance to keep multiple systems in sync such that it is possible to correlate log events across them. /Per -- Per Jessen, Zürich (10.4°C) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Per Jessen wrote:
j debert wrote:
Why do you need ntp?
For instance to keep multiple systems in sync such that it is possible to correlate log events across them.
/Per
I often set up telecom equipment and whenever possible I configure it to use ntp, as I occasionally have to compare logs. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Thu, 29 Oct 2009 07:15:21 -0700 j debert <jdebert@garlic.com> wrote:
Why do you need ntp?
All computer clocks drift. Some more than others.
Is the system clock severely sloppy about keeping correct time?
The system clock is only sometimes accurate when no other interrupts have the CPU. Otherwise, a timer interrupt updates the number of ticks, which updates the system time.
Is the CMOS clock a P.O.S.?
No. The CMOS clock is only used when the computer is powered off.
Is there something critical that requires zero delta time with some other system(s)?
Left to their own devices, without NTP any two computer clocks will drift apart.
[ .... ] A script to periodically test the network connection and run netdate or rdate when the network is up works just fine.
This is just a poor re-implementation of the NTPd client. And assumes available connectivity to a static host.
[ .... ] necessary to adjust clocks every couple months or so. So all ntp would do is tie up resources that could better be used for other things.
I don't know where you got this mis-information. NTP is extremely light weight. It does not tie up resources.
ntp was created because of the need to keep system clocks accurately synchronised on a network for some time-critical applications. Most people do not need it. Even the author discouraged it's use.
Again, this is pure nonsense. NTP is the correct method to keep computer clocks in sync with the world clock. If you want to have accurate time on your computer, you want to configure and run NTP.
There was once an easy to follow adjtime tutorial [ ... ] Anyone know where it went to?
Before I used openSuse, configuring NTP on server was as simple as editing /etc/ntp.conf and changing the server line to point to the hostname(s) of local ntp server(s). I just fired up yast and see that it just puts up a radio box without explaining what the different options mean. Obviously, mobile computing has definitely upset the ease of setting up network settings, NTP being only one part of a bigger configuration issue.
== jd
-- Rich Coe rcoe@wi.rr.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Rich Coe wrote:
Again, this is pure nonsense. NTP is the correct method to keep computer clocks in sync with the world clock.
While NTP is an excellent method for syncing clocks, there are others, such as receivers for WWVB or other time signals. Another accurate source is the CDMA cell phone network. Also back in the analog TV days, PBS in the United States had an accurate time signal in the vertical blanking interval, which was used by many TVs and VCRs to obtain the time. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
James Knott wrote:
Rich Coe wrote:
Again, this is pure nonsense. NTP is the correct method to keep computer clocks in sync with the world clock.
While NTP is an excellent method for syncing clocks, there are others, such as receivers for WWVB or other time signals. Another accurate source is the CDMA cell phone network. Also back in the analog TV days, PBS in the United States had an accurate time signal in the vertical blanking interval, which was used by many TVs and VCRs to obtain the time. Forgot to mention, there's also GPS which provides one of the best time signals going.
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, 06 Nov 2009 09:49:17 -0500 James Knott <james.knott@rogers.com> wrote:
James Knott wrote:
Rich Coe wrote:
Again, this is pure nonsense. NTP is the correct method to keep computer clocks in sync with the world clock.
While NTP is an excellent method for syncing clocks, there are others, such as receivers for WWVB or other time signals. Another accurate source is the CDMA cell phone network. Also back in the analog TV days, PBS in the United States had an accurate time signal in the vertical blanking interval, which was used by many TVs and VCRs to obtain the time. Forgot to mention, there's also GPS which provides one of the best time signals going.
The ntp code supports these as time sources for the purpose of syncing computer clocks. -- Rich Coe rcoe@wi.rr.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* James Knott (james.knott@rogers.com) [20091106 15:45]:
While NTP is an excellent method for syncing clocks, there are others,
You still need devices that sync your computer clock to those other sources and ntpd supports quite a few of them, among them GPS or WWMV. So you don't need internet access to use ntpd. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* j debert (jdebert@garlic.com) [20091031 02:28]:
ntp was created because of the need to keep system clocks accurately synchronised on a network for some time-critical applications. Most people do not need it. Even the author discouraged it's use.
Which one? The main author of ntpd certainly does not do so. And synchronised time isn't only needed for time-critical applications. Even things like compiling code that resides on another machine in a network may fail because of missing synchronisation. Bottom line is: if you're part of a local network you definitely need synchronised time. If not, it's more or less nice-to-have like a radio clock and unlike your statement ntpd is very lightweight in it's resource needs. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
participants (15)
-
Carlos E. R. -
David C. Rankin -
G T Smith -
j debert -
James Knott -
Ken Schneider - openSUSE -
Marc Chamberlin -
Matthias Bach -
Per Jessen -
Philipp Thomas -
Pit Suetterlin -
Rich Coe -
Roger Oberholtzer -
Sven Burmeister -
Will Stephenson