[2776] in Release_Engineering
Re: xdm
daemon@ATHENA.MIT.EDU (Richard Basch)
Sat Apr 18 20:22:50 1992
Date: Sat, 18 Apr 92 20:22:24 -0400
To: rel-eng@MIT.EDU
Cc: mar@MIT.EDU
In-Reply-To: rel-eng[2775]
From: "Richard Basch" <basch@MIT.EDU>
The following part of the patch to verify.c is BAD...
Whoever picks up the XDM changes should NOT allow this part of the patch
to be entered into the source tree without changing it!!!
! putuserattr(p->pw_name,S_SUCHK,(void *)1,SEC_BOOL);
! putuserattr(p->pw_name,S_RLOGINCHK,(void *)1,SEC_BOOL);
! putuserattr(p->pw_name,S_ADMIN,(void *)0,SEC_BOOL);
putuserattr(p->pw_name,(char *)NULL,((void *) 0),SEC_COMMIT);
+ putuserattr(p->pw_name,"athena_temp",(void *)1,SEC_INT);
+ putuserattr(p->pw_name,(char *)NULL,((void *) 0),SEC_COMMIT);
enduserdb();
You should only be doing ONE SEC_COMMIT, and that should be the final
step. This code has the potential of messing up the files in
/etc/security on the RISC/6000 (and we know what happens when that
occurs; it won't boot).
-Richard
