Andrej Podzimek wrote: >>>> Hi, >>>> >>>> Do you have any search domains in /etc/resolv.conf that have a wildcard >>>> AAAA record? >>> >>> No, I only have one line there: >>> >>> nameserver 2002:5f52:81a4::1 >> >> I'd use snoop or wireshark to find out if the bad responses are coming >> from that name server. > > Well 'dig' returns correct answers. So the bad responses do not come > from the name server. The bad responses are only produced by > gethostbyname()/getaddrinfo(), probably due to incorrect configuration. > (All works fine on Linux and FreeBSD on the same machine.)
Yes, I realize you indicated all of that with your first posting, but it still seems so very surprising that random results would show up in the address list returned by those functions that bad data on the wire (perhaps due to a difference between the actual query that 'dig' makes and the one that getaddrinfo() makes) that it makes some sense to me to try to isolate the components. As part of troubleshooting, the query from the application normally goes through nscd, which then makes the query on the wire, caches the result, sends it to the kernel for sorting, and then returns it to the application. It's not yet clear where in that process something is going awry, so knowing for _certain_ that the data are good at some known point and then bad afterwards would probably help. Plus, I've been using the IPv6 code for a while, and I can't say I've seen anything like that. Do you see the same anomalous results if you do "getent ipnodes host.name.here"? There's nothing odd in /etc/inet/ipnodes or /etc/nsswitch.conf, is there? -- James Carlson 42.703N 71.076W <carls...@workingcode.com> _______________________________________________ networking-discuss mailing list networking-discuss@opensolaris.org
