Bug #238

Something else may be firing up named.

I also notice something is checking the local time using ntpdc, which fails as ntpd isn’t running it seems, but something similar. Is it possible this script is restarting named?

Or so I hope. Fixed in rc6, will test further then. in the meantime, you can add instances = 1 to your /etc/xined.d/named file and see if the problem recurs. thx for spotting this!

The right answer is to fix ntp for dnssec, period, and move that version of ntp into the ceropackages github repo.

Which requires braincells I do not have at present. I haven’t even got so far as to make it work on x86. when I poked into the guts of ntp it was not obvious - or possible - to separate out ntp -g processig from normal processing, easily.

Why not fix NTP to start with?

ntpd does not actually start running, and the bind 9 monitoring script
to enable dnssec seems to be using ntpdc to poll.

I think there is another ntp process that uses port 123, and killing
that off allows ntpd to start.

• –Michael

—-BEGIN PGP SIGNATURE—-
Version: GnuPG v1.4.11 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk5ucA4ACgkQl6Nz7kJWYWbhNgCeOBGQg6MxusqD+qUG5GA9qdAC
0iAAn2HEtVZVgJliAN/9y0TpBHCchcMP
=Bxii
—-END PGP SIGNATURE—-

I should have fixed ntp in the first place, but this bug was long before I had a deep understanding of the right calls to make, and even then, I’d rather someone with a deep understanding of ntp and dnssec to tackle that, rather than me… if it was easy, I’d have done it….

http://huchra.bufferbloat.net/~cero1/3.3/3.3.2-8/