Graham Leggett kirjoitti viestissään (lähetysaika Saturday 04 March 2006
18:44):
Is there some kind of manual override that I can use to switch this
behaviour off? Or alternatively if this is not possible, to require
localhost so that I can run the admin server behind a reverse proxy
whose access control does work properly?
Because of the bug, you have to set
nsAdminAccessAddresses to something you
don't have and empty nsAdminAccessHosts. Well, there might be other ways to
do it, this worked for me.
I needed to allow administration from anywhere so made the following
definitions:
nsAdminAccessAddresses=255.255.255.255
nsAdminAccessHosts=
and restarted the admin server.
Regards
Kimmo Koivisto
Having changed the *.domain.com to * I am now getting this error:
[Sat Mar 04 10:42:50 2006] [notice] [client xx.xx.xx.xx]
admserv_host_ip_check: Unauthorized host ip=xx.xx.xx.xx, connection
rejected
Google finds other people with this problem, apparently "*" doesn't mean
"let everybody in", but instead it means "let everyone in whose reverse
DNS works". In this case reverse DNS does work, but I may be getting
bitten by bug 183925.
So in short, does the admin server in v1.0.2 work at all, or am I just
wasting my time? :(
Regards,
Graham
--