Oliver Jowett wrote:
> Ian Pilcher wrote:
>> Did you read the bug?
> Yes, I did.
My appologies if that sounded a little short.
> Anyway, thanks for the report and workaround, it's good to get that
> information into the archives. There doesn't seem to be anything that
> postgresql or the JDBC driver can do about the problem, though.
There might be a way for the driver to force an IPv4 connection, even
when java.net.preferIPv4Stack isn't set. It doesn't seem worth the
effort to me, though.
Something I find slightly interesting is the idea of having postgresql
do its own parsing of /proc/net/tcp* (or netstat output) when it gets a
connection request over the loopback adapter. This would obviate the
need for a seperate IDENT daemon (with all of its security problems) in
the loopback case.
In the short term, however, the answer is definitely to get the IDENT
daemon authors of the world to fix their programs. (I am somewhat
concerned that Red Hat will come back with a "working as designed"
answer.)
--
========================================================================
Ian Pilcher i.pilcher@comcast.net
========================================================================