It looks like bind in Red Hat 6.1 comes with incorrect init
[jes@pcitpdp17 jes]$ rpm -qlp bind-8.2.1-7.i386.rpm
named is never launched after a boot, in 6.0 rc.d had
S55named instead of K45named
This was a (regrettably, pretty much undocumented) policy change.
The general philosophy was to have more servers, especially ones
that require extra configuration, off by default, as it increases
security. This includes bind, innd, dhcpd, samba, amd, and knfsd.
In theory, user's settings with chkconfig would get preserved on
upgrade, but this didn't happen with bind because it's one of the
packages that also includes the links in the package.
Of course, to turn it on, you can do:
chkconfig --level 345 bind on
or, just use ntsysv.
*** Bug 6411 has been marked as a duplicate of this bug. ***
When you upgrade from 8.2 to 8.2.1 then your DNS will be
disabled wich is NOT the way I think.
It should leave the old settings intact and only disable or
better said not-enable when first installing this package.
It took me a serious search why some programs didn't work
anymore because the DNS wasn't started...
I get a strange msg when querying the status:
named 8.2.1 Tue Oct 12 21:49:04 EDT 1999
number of zones allocated: 64
debug level: 0
xfers running: 0
xfers deferred: 0
soa queries in progress: 0
query logging is ON
server is DONE priming
server IS NOT loading its configuration
It's this last msg about NOT loading its configuration, what
does that mean and what should I do to correct it?
The process is starting ok now that I've enabled it with