https://bugzilla.novell.com/show_bug.cgi?id=778367 https://bugzilla.novell.com/show_bug.cgi?id=778367#c7 --- Comment #7 from Michael Meeks <mmeeks@suse.com> 2012-12-12 10:48:52 UTC --- This is a domain-name lookup: 20025 15:06:17.836186 poll([{fd=24, events=POLLIN}], 1, 3000) = 0 (Timeout) 20025 15:06:20.839593 socket(PF_INET, SOCK_DGRAM|SOCK_NONBLOCK, IPPROTO_IP) = 25 20025 15:06:20.839834 connect(25, {sa_family=AF_INET, sin_port=htons(53), sin_addr=inet_addr("203.192.133.9")}, 16) = 0 20025 15:06:20.839970 poll([{fd=25, events=POLLOUT}], 1, 0) = 1 ([{fd=25, revents=POLLOUT}]) 20025 15:06:20.840076 sendto(25, "Vv\1\0\0\1\0\0\0\0\0\0\10qa_yuke3\3lab\3bej\4apac\6novell\3com\0\0\1\0\1", 50, MSG_NOSIGNAL, NULL, 0) = 50 I -imagine- that (somehow) the dumb reverse-lookup bug has come back. Can you run those 'time' calls ? a few dozen times and see what the results are ? The idea that it is fast to do a reverse/forward lookup on our hostname is one that we have tried to stamp out of the code all over the desktop & in LibreOffice for many years. It keeps coming back sadly. Without that information; I'd be inclined to close this bug. Almost - better is getting a stack-trace: 20 seconds is long enough to attach gdb, and do a: thread apply all backtrace and attach the output here. Otherwise, I guess we'd need to close this ( no info ). Unless you can reproduce it Yifan ? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.