Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Hi,
My router (and other machines) get their time from my dns-323 running ntpd from ffp 0.5 and I have just noticed this in the router log
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] Requesting time from [IP Address of the DNS-323]
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] ntpd: Executing ntpdate -s 0.0.0.0
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] ntpdCmdExec: exec'ing -> ntpdate -s 0.0.0.0_
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] ntpdCmdExec: ntpdate -s 0.0.0.0 -> cmdStatus: 512_
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] Received KOD packet from [IP Address of the DNS-323]
Could the KOD packet be because the router requests time every 2 hours and ntpd on the 323 thinks this is too frequent? I don't mind the two hour poll though as it's only a home network. One other thing I should say is that the 323 gets its time through its connection to the outside world via the router - so to start with when everything boots up (I don't leave them on overnight) the router may request time from the 323 before ffp/ntpd is up or if the 323 is up first it won't be able to get to the outside world to update its time.
I have just looked at the log further, so it looks like the router isn't getting a DNS server (see below). That would make sense as the cable modem is also booting/getting a connection at the same time and may not be up when the router fails to get time from the 323 and rolls over to the alternate ntp server (uk.pool.ntp.org). Assuming that the router can get to ntpd on the 323 (on the basis that the router log is showing that it has received a KOD packet from the 323) why is the KOD packet being sent? Is it the case that ntpd won't give anybody time data if it hasn't updated its own time from another time server?
1970 Jan 1 00:00:13 [ROUTER] [ntpdate] Looking Up uk.pool.ntp.org [the alternate ntp server address specified in the router]
- Last output repeated twice -
1970 Jan 1 00:00:20 [ROUTER] [root] [SYSTEM START-UP] System Started
1970 Jan 1 00:00:23 [ROUTER] [wand] [FW] Firewall Restarted
1970 Jan 1 00:00:23 [ROUTER] [ntpdate] Looking Up uk.pool.ntp.org
- Last output repeated twice -
1970 Jan 1 00:00:29 [ROUTER] [wand] [FW] Firewall Restarted
1970 Jan 1 00:00:33 [ROUTER] [ntpdate] Looking Up uk.pool.ntp.org
- Last output repeated twice -
1970 Jan 1 00:00:40 [ROUTER] [wand] [FW] Firewall Restarted
1970 Jan 1 00:00:43 [ROUTER] [ntpdate] Looking Up uk.pool.ntp.org
- Last output repeated 2 times -
1970 Jan 1 00:00:56 [ROUTER] [wand] [IPSEC] IPSEC Restarted
1970 Jan 1 00:00:58 [ROUTER] [ntpdate] Looking Up uk.pool.ntp.org
1970 Jan 1 00:01:03 [ROUTER] [ntpdate] Next DNS Retry after 1 hour
Another point: I have noticed that sometimes ntpd crashes on the 323. I only see this when I try and get console access and it doesn't work as the ssh/telnet daemons run after ntpd. I note that adjtimex.sh and ntpd.sh run early on during ffp boot. Is there any reason why adjtimex.sh and ntpd.sh can't be made to run last in the sequence (I assume using an AFTER directive in the scripts). Also is there any way to flag that this has happened - the only way I can think of off the top of my head is to alter the scripts so they, say, touch /mnt/HD_a2/ntpdup and externally check for that file from say a PC and issue an alert if it isn't there. Any thoughts?
Offline