Package: ntp-simple Version: 1:4.2.0a+stable-2sarge1 Severity: important Hi,
After my the leap second got inserted, my timeservers stopped synchronizing, and they all fall back to the local clock. I've tried restarting one but that didn't help. My setup: I use 2 external ntp servers (a and b), and have 3 hosts running ntpd (1,2 and 3): They all have: server a server b On host 1 I also have: peer 2 peer 3 host 2 has: peer 1 peer 3 host 3 has: server 1 server 2 Then they all also have: server 127.127.1.0 fudge 127.127.1.0 startum 13 In my log I see at 0:59:59 UTC+1: kernel: Clock: Inserting leap second 23:59:60 UTC In my daemon.log file on host 1 I see that on Jan 1 at 2:27:03 UTC+1 it switched to local clock, which is something it never did before. At 02:27:44 it switched back to server A. Then at 3:00:31, it shows: time reset +1.005810 s At: 3:03:51 it switch to local again, and at 3:05:54 it swiched to peer 2 (which was at stratum 14). It switched peers again a few times and then stayed at the local clock. Currently it says that server A and B are about -1.6 seconds, while 2 and 3 have an offset of +135 and +45 ms. The other 2 hosts show something that is very simular, just the times are a little bit different. My guess is that the "time reset" of about +1 s caused the problem, in combination with the leap second. I'm guessing a bad interaction between ntpd and the kernel. Should I restart all my timeservers to get this working again? Kurt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
