I've seen the same issue recently with: chrony-3.1-4.fc26.x86_64
[seitz@storage ~]$ chronyc sources 210 Number of sources = 7 MS Name/IP address Stratum Poll Reach LastRx Last sample =============================================================================== ^* ntp-gps1.fiber.house 1 7 377 122 +799ns[-2226ns] +/- 1252us ^- some.server 1 10 377 995 +32us[-3108ns] +/- 35ms ^- another.server 1 10 377 927 -3831us[-3869us] +/- 15ms Where ntp-gps1.fiber.house was dead/down/not powered up, lastRx was 10h but it remained 'synced' to it :( Seems like correct operation would mark this server as offline, but continue to poll and mark it online again once it comes back up? I know this is how 'ntpd' works. On Thu, Oct 19, 2017 at 05:04:14PM +0000, Parker, Michael D. wrote: > I had a similar type of problem using the release Chrony 2.x under RHEL 6 > using only 2 time sources. > All looked good at the start but eventually the problem showed up. > If I recall my research, it had something to do some type of time source > window situation. > If the sources' +/- differences windows did not overlap chrony quit trying to > go back into sync. > I was polling much more frequently than you were. > > -----Original Message----- > From: Bill Unruh [mailto:[email protected]] > Sent: Thursday, October 19, 2017 8:23 AM > To: [email protected] > Subject: -EXT-Re: [chrony-users] Re: chrony losing sync with timeserver and > never recovers > > I am certainly confused by the log. The GPS and the PPS have never delivered > a valid ntp to chrony. What happened 463 days ago? > When was this system started? > Or have you edited these responses, not showing us the complete picture? > Ie, are there other servers which are being polled? It may be that if there > is no response from a server for N poll periods, chrony gives up and stops > polling. I no longer know chrony well enough to say that this is what happens > or not. > > On Thu, 19 Oct 2017, Brendan Simon (eTRIX) wrote: > > > Hi chrony-users, > > > > Anyone know why chrony would stop polling time servers? maxpoll is > > supposed to be 1024 (17 minutes) max, but I my system polls initially > > and then seems to stop polling servers completely. The example below > > shows 463 days with no response from 2 servers. > > > > Chrony 1.30 on Debian 8 (Jessie) > > > > Thanks, Brendan. > > > > ______________________________________________________________________ > > ______ > > > > On 17/10/17 2:24 pm, Brendan Simon wrote: > > I have a number embedded systems that are located in remote > > areas that need to be up 24/7 for logging of data via a 3G > > internet connection. The systems are ARM based and running > > Debian 8 (Jessie) with chrony installed as the ntp client. > > > > The systems sync with 2 ntp servers (`tic.ntp.telstra.net` and > > `toc.ntp.telstra.net`) on boot. I know this because (a) there > > is no RTC on the system, and (b) the application does not start > > until the system date is > 2015 (i.e. not the startup of default > > 1970). > > > > For some reason chrony loses sync with the servers and never > > recovers. I have system times that are out by minutes !! > > > > # chronyc sources > > 210 Number of sources = 4 > > MS Name/IP address Stratum Poll Reach LastRx > > Last sample > > =========================================================================== > > ==== > > #? GPS 0 4 0 > > 10y +0ns[ +0ns] +/- 0ns > > #? PPS 0 4 0 > > 10y +0ns[ +0ns] +/- 0ns > > ^? tic.ntp.telstra.net 2 10 0 > > 463d -14ms[ -15ms] +/- 44ms > > ^? toc.ntp.telstra.net 2 10 0 > > 463d -23ms[ -24ms] +/- 79ms > > > > > > As can be seen, the servers have a state of '?' and haven't > > recieved data in 463 days !! yet I can ping them ok, and if I > > restart chrony all is good again. > > > > The 3G modem can have problems and are reset (powered down and > > up) whenever internet connectivity is lost (detected by pings > > not responding). And 3G connectivity is not the most reliable. > > > > The interesting parts of the config are: > > It would actually be more helpful if you gave us all the information. You do > not know what is happening, so how can you be sure that these are the only > "interesting" commands in your chrony.conf? > > > > server tic.ntp.telstra.net iburst > > server toc.ntp.telstra.net iburst > > > > makestep 1000 -1 > > > > initstepslew 30 0.au.pool.ntp.org 1.au.pool.ntp.org > > 2.au.pool.ntp.org 3.au.pool.ntp.org > > > > > > What causes chrony to not retry servers? > > > > Is there a config setting I need to always try these servers? > > > > I notice the `online` and `offline` settings. Do I need to > > explicitly tag servers as `online`? I presume that's the > > default. > > > > Do I need to explicitly tag the servers as `offline` before > > powering the modem up and down? I thought leaving them online > > would be ok. The only downside is it may take a little longer > > to get the time back in sync, right? > > > > But they can only get back in sync if chrony is talking to the > > servers. > It might help to tag them as online when the 3g comes back up. -- Bryan G. Seitz -- To unsubscribe email [email protected] with "unsubscribe" in the subject. For help email [email protected] with "help" in the subject. Trouble? Email [email protected].
