While debugging some NTP weirdness, I noticed an outlier amongst the
constraints (all except one had an offset less than 1s, the one
outlier was over 7 hours off). Unfortunately, I couldn't tell which
constraint had the outlier, because the IP's rDNS didn't correspond
to one of the constraints (it was cloudfront's), and none of the
constraints resolve to the IP address logged.
Therefore, I suggest the following patch which logs the name of the
constraint together with its IP address:
Index: constraint.c
===================================================================
RCS file: /cvs/src/usr.sbin/ntpd/constraint.c,v
retrieving revision 1.54
diff -u -p -r1.54 constraint.c
--- constraint.c 27 Nov 2022 13:19:00 -0000 1.54
+++ constraint.c 20 Apr 2023 09:03:00 -0000
@@ -679,8 +679,9 @@ constraint_msg_result(u_int32_t id, u_in
offset = gettime_from_timeval(&tv[0]) -
gettime_from_timeval(&tv[1]);
- log_info("constraint reply from %s: offset %f",
+ log_info("constraint reply from %s (%s): offset %f",
log_sockaddr((struct sockaddr *)&cstr->addr->ss),
+ cstr->addr_head.name,
offset);
cstr->state = STATE_REPLY_RECEIVED;
Cheers,
Paul
--
>++++++++[<++++++++++>-]<+++++++.>+++[<------>-]<.>+++[<+
+++++++++++>-]<.>++[<------------>-]<+.--------------.[-]
http://www.weirdnet.nl/
