[1946] in Kerberos-V5-bugs
Why make check fails under Ultrix....
daemon@ATHENA.MIT.EDU (epeisach@MIT.EDU)
Sun Jun 2 18:50:14 1996
From: epeisach@MIT.EDU
Date: Sun, 2 Jun 1996 18:50:02 -0400
To: krbdev@MIT.EDU
Cc: krb5-bugs@MIT.EDU
I consider this important enought that it should be included in the
release, but I am not sure what machine to test with...
The failure is to include to compile our own version of
getdtabliesize.o in appl/gssftp/ftpd.
The fix is to conditionalize the building of getdtablesize.o on the
need to build it... If you do not include it, ftpd works under ultrix....
We can change the configure.in - similarly to how the setenv is
defined, but I am not sure what plaform actually needs this for proper
testing...
Why does Ultrix break in the sysconf call is another issue... sysconf
when looking for OPEN_MAX calls getdtablesize.... So our call to
getdtablesize calls sysconf, which calls getdtablesize, repeat...
Sooner or later, your stack is exhausted....
So - does this deserve to be fixed before the release - I think so...
This is my list of thinks to patch then before beta-6
appl/bsd/login.c Compile w/o krb4 - Ted has patch
appl/bsd/telnet/*/configure.in Compile w/o krb4 - Ted has patch
appl/gssftp/ftpd/configure.in, Makefile.in: This problem....
Ezra
