> Le 31 août 2015 à 07:05, ※踏梦狼※ <121156...@qq.com> a écrit : > > hello, > could you have related the ntp log article in your ntp website ? i want > to understand the meaning of the below log. > > 31 Aug 02:55:46 ntpd[10817]: 0.0.0.0 061c 0c clock_step -0.193855 s ***** > system clock > 128ms fast so step it back. pb seen more than once. > 31 Aug 02:55:46 ntpd[10817]: 0.0.0.0 0615 05 clock_sync > ***** report we are sync’d > 31 Aug 02:55:47 ntpd[10817]: 0.0.0.0 c618 08 no_sys_peer > ***** no server is considered healthy enough for selection > 31 Aug 05:33:44 ntpd[10817]: 0.0.0.0 0628 08 no_sys_peer > ***** thing are better for a while, then it happens again > 31 Aug 05:49:05 ntpd[10817]: 0.0.0.0 0613 03 spike_detect +0.203705 s ***** > single clock offset greater than 128ms , report and forget it > 31 Aug 06:06:35 ntpd[10817]: 0.0.0.0 061c 0c clock_step +0.193691 s ***** > step threshold passed repeatedly so step again.
drift of 194ms in 11489s which is less than the free clock drift if the later drift file value is to be believed. (would be about 384ms). Curious.. it is almost exactly half?????????? I don’t know how to interpret that. > 31 Aug 06:06:35 ntpd[10817]: 0.0.0.0 0615 05 clock_sync > 31 Aug 06:06:36 ntpd[10817]: 0.0.0.0 c618 08 no_sys_peer > 31 Aug 07:56:00 ntpd[10817]: 0.0.0.0 0613 03 spike_detect +0.130901 s > 31 Aug 08:22:16 ntpd[10817]: 0.0.0.0 061c 0c clock_step +0.129834 s drift of 130ms in 8141s which is less than the free clock drift if the later drift file value is to be believed. (would be about 298ms). About half again? What’s wrong with my sums? > 31 Aug 11:07:31 ntpd[10817]: 0.0.0.0 0613 03 spike_detect -0.131656 s > 31 Aug 11:16:12 ntpd[10817]: ntpd exiting on signal 15 > ***** ntpd is restarted > 31 Aug 11:16:12 ntpd[44046]: 0.0.0.0 c016 06 restart > 31 Aug 11:16:12 ntpd[44046]: 0.0.0.0 c012 02 freq_set kernel -36.649 PPM > **** loads drift file sets initial frequency, to adjust for a fast local > oscillator > 31 Aug 11:19:30 ntpd[44046]: 0.0.0.0 c615 05 clock_sync I’m up a stump. It looks like the system clock is being corrected for just half its frequency drift. So there is some time when the clock is being sync’d to the servers. I still would look at your config first and try and find out what clients are seeing issues and which are not and list the differences between them. Something will pop out of that to point to a probably cause. > > thanks > > ------------------ 原始邮件 ------------------ > 发件人: "Mike Cook";<michael.c...@sfr.fr>; > 发送时间: 2015年8月29日(星期六) 下午4:41 > 收件人: "Mike Cook"<michael.c...@sfr.fr>; > 抄送: "※踏梦狼※"<121156...@qq.com>; "questions"<questions@lists.ntp.org>; > 主题: Re: [ntp:questions] what's the matter with my ntp > > > <snip> > > For me this would not be considered acceptable. I would NEVER have ntp > > servers in virtual machines. Especially if they are LAN primary servers. > > The figures for delay and > > Oops… I meant offset. delay is not an issue unless it is highly variable, > indicating asymmetric path length and induces jitter. > > > jitter are abnormally high. > > > "The bugs are notably apparent when the PC is disconnected from the mains". > PC Shop director "The main function of a modern police force is filling in forms." _______________________________________________ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
