On Sep 17 09:03, Corinna Vinschen wrote: > ----- Forwarded message from "Pierre A. Humblet" ----- > > | Sorry, an earlier answer was rejected due to inappropriate subject.
In theory, that should be fixed. > > | After thinking about it, I don't like mixing calls to the Windows > > resolver > > | for res_query and contacting DNS servers directly with res_send, > > | as proposed. So I have a patch where res_send also uses the Windows > > | resolver. > > | > > | Unfortunately although I can build cygwin1.dll fine, it's broken, > > | something to do with a NULL stdout (this is from /bin/date) > > | 1 thread 2588.0x4c0 fputc (ch=10, file=0x0) > > | at ../../../../../src/newlib/libc/stdio/fputc.c:101 > > | > > | Not sure what to do, I already did make clean for cygwin. I don't know either. I have no such problem with CVS HEAD. > > | Also minires used to come with a README explaining the effect of > > | an optional /etc/resolv.conf (e.g. to bypass the Windows resolver). > > | That information is not present currently [ and nobody asks for it :) ] > > | I wonder if we should add it and where to place it. One option is the > > | User's Guide. Another option is a custom resolv.conf man page. > > | What do you think? The User's Guide would be a good place, IMHO. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat