wolf@grasp2.univ-lyon1.fr (Christophe Wolfhugel) (02/20/91)
Still trying to make seteuid work on AIX 3.1 (RS/6000). I've got a Sun POSIX man about this function, and it speaks about a certain syscall that should be issued in order to be autorized to use this function. Another person told me about extented file attributes. Can anyone help me? (I halas don't have enough disk space to have the C man on the machine). - Chris (wolf@grasp1.univ-lyon1.fr)
jfh@greenber.austin.ibm.com (John F Haugh II) (02/21/91)
In article <1991Feb20.131737.3780@rusmv1.rus.uni-stuttgart.de> wolf@grasp2.univ-lyon1.fr (Christophe Wolfhugel) writes: >Still trying to make seteuid work on AIX 3.1 (RS/6000). I've got a Sun POSIX >man about this function, and it speaks about a certain syscall that should >be issued in order to be autorized to use this function. Another person told >me about extented file attributes. > >Can anyone help me? (I halas don't have enough disk space to have the C man on >the machine). You don't describe exactly what isn't working about it. The requirement for seteuid is that you must have the new effective UID as either your real or saved UID. Being root won't cut it, as there is an AIX requirement that your effective UID be either your real or saved UID. This can be gotten around by setting the effective and real, or the effective and real and saved all at the same time. The behavior of seteuid wasn't specified by POSIX the last time I worked on that system call. I did get to see a draft copy of the new POSIX specification for seteuid(), but I doubt that Sun has implemented that behavior yet - I only saw the draft copy four or so months ago. [ I think Doug Gwyn or Guy Harris was the person who sent me the copy, and thanks are due to them ] -- John F. Haugh II | I've Been Moved | MaBellNet: (512) 838-4340 SneakerNet: 809/1D064 | AGAIN ! | VNET: LCCB386 at AUSVMQ BangNet: ..!cs.utexas.edu!ibmchs!auschs!snowball.austin.ibm.com!jfh (e-i-e-i-o)