Thread: pg_hba.conf and Solaris

I didn't like the Solaris bug mention in pg_hba.conf.  It seemed like
the wrong place:

+ # Note: On some Solaris systems, an IP-MASK of 255.255.255.255 is known not to work.
+ # The corresponding CIDR-MASK of /32 does work. 

I have removed it.  Should we put something in the docs instead, or add
some code to emit a log message if netmask is used on Solaris?  Is
anyone going to use netmask anymore now that the sample has /32?

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 

Bruce Momjian said:
> I didn't like the Solaris bug mention in pg_hba.conf.  It seemed like
> the wrong place:
>
> + # Note: On some Solaris systems, an IP-MASK of 255.255.255.255 is
> known not to work. + # The corresponding CIDR-MASK of /32 does work.
>
> I have removed it.  Should we put something in the docs instead, or add
> some code to emit a log message if netmask is used on Solaris?  Is
> anyone going to use netmask anymore now that the sample has /32?
>

Put something in the docs instead, if you want. Note that netmask apparently
does work in the general case - it is only this specific netmask that does
not.
If we're going to write code for it, it seems more sensible to make it work
than to emit a warning, either by using inet_aton as previuosly discussed,
or just testing for that exact mask and calling SockAddr_cidr_mask instead
of getaddrinfo_all.

cheers

andrew