On Tue, May 14, 2019 at 10:37:06AM -0400, Dan Arena wrote: > This specifically says "only valid measurements from synchronised > sources", but It was still showing log lines for the source.
Measurements that passed the NTP tests 1-7 are considered valid. A root dispersion of 5.4 seconds is still ok at the NTP level. It's only later in the source selection when that's too much. > When I > looked at `chrony sources -v` it listed the source state as "?", which > is quite confusing because the source is not actually "unreachable". Yeah, the -v help is confusing here. '?' is a catch-all symbol for all other problems. The man page says "or whose packets do not pass all tests", but that still is not an accurate description. > Why does chrony show the source as " '?' = unreachable", instead of > like "'x' = time may be in error"? x indicates a falseticker. Large dispersion is something very different. Ideally failures in the selection tests, and there are quite a few of them, would all have different symbols. I hesitate with that as it might break compatibility with tools parsing the chronyc output. > Why are there lines in the > measurements file when the source is not synchronised? The source is synchronized, but it's not considered good enough to be used for synchronization of the local clock. > Should I just leave the maxdistance at 16.0? Yes, if you can't get access to a better server. > I am kind of inclined to > blame the person managing the NTP server; they should probably sync > its time regularly some how, either manually or from an external > source (over internet or GPS, whatever). It's probably a Windows NTP server. Large dispersion seems to be a common problem with them. The admin might want to configure a shorter polling interval or switch to ntpd. -- Miroslav Lichvar -- To unsubscribe email chrony-users-requ...@chrony.tuxfamily.org with "unsubscribe" in the subject. For help email chrony-users-requ...@chrony.tuxfamily.org with "help" in the subject. Trouble? Email listmas...@chrony.tuxfamily.org.