I am seeing a case where one board in my system keeps switching its peer back and forth between its local clock and a remote ntp server for about 40 minutes when it first comes up. Eventually, it correctly starts using the remote ntp server.
My setup includes a stratum 1 server (40.14.0.254) that is connected to one board in my system (144.190.31.2). The second board that is swapping its peer back and forth is connected to this first board. Both boards are running linux. A snippet of some ntpq output is shown below: [09:36:02] remote refid st t when poll reach delay offset jitter [09:36:02] ============================================================================== [09:36:02] +144.190.31.2 40.14.0.254 2 u 5 16 17 0.365 0.022 0.004 [09:36:02] *LOCAL(1) .LCL. 14 l 5 16 17 0.000 0.000 0.004 [09:37:02] remote refid st t when poll reach delay offset jitter [09:37:02] ============================================================================== [09:37:02] *144.190.31.2 40.14.0.254 2 u 11 16 177 0.365 0.022 0.004 [09:37:02] LOCAL(1) .LCL. 14 l 1 16 377 0.000 0.000 0.004 [09:37:12] remote refid st t when poll reach delay offset jitter [09:37:12] ============================================================================== [09:37:12] +144.190.31.2 40.14.0.254 2 u 3 16 377 0.345 0.012 0.011 [09:37:12] *LOCAL(1) .LCL. 14 l 11 16 377 0.000 0.000 0.004 … Any ideas why this would be happening, and what is the best approach to debug? Is there a specific ntpq command that can provide enough information? Is there a debug level that I can set when I start the ntp daemon that will dump this information … if so, what should I be looking for in the debug? Many thanks, SteveW _______________________________________________ questions mailing list [email protected] http://lists.ntp.org/listinfo/questions
