Front page | perl.perl5.porters |
Postings from July 2011
On Mon, Jul 04, 2011 at 02:47:07PM +0100, Paul LeoNerd Evans wrote: > On Mon, Jul 04, 2011 at 09:46:57AM +0100, Nicholas Clark wrote: > Hmm.. > > ./netdb.h:595:# define EAI_NONAME -2 /* NAME or SERVICE is unknown. */ > > Looks like a permanent failure there, but that's my glibc on Linux. That > particular error is supposed to be a permanent "the nameserver said this > doesn't exist", rather than a temporary failure of not being able to > talk to the nameserver or getting a response. > > That said, if it failed just once, it might just be e.g. a caching > problem upstream somewhere. This is the sort of small wrapper around > libc that's hard to unit test in a way that doesn't end up relying on > external dependencies to give the right answers. :/ It failed just once http://m-l.org/~perl/smoke/perl/linux/blead_g++_quick/rptd632d9afe2fe0ca4acebe9e4db95ef6ec14a299d.rpt v5.15.0-213-gd632d9a Configuration (common) -Dcc=g++ -O ----------- --------------------------------------------------------- F O -Uusenm O O -Uusenm -Duseithreads | +--------- -DDEBUGGING +----------- no debugging so from what you're saying, that means someone somewhere is saying that it's a permanent failure, when they should be saying that it's temporary? Grr. Software hateful. Other people's software, doubly so. Nicholas ClarkThread Previous