RE: 2 of 36 test failed RCS5.8 - final update

"Mike Dooley (midooley)" <[email protected]> Mon, 9 Jan 2012 12:58:04 -0800
Newsgroups gmane.comp.version-control.rcs.bugs
Message-ID <0E84D1EE8EC3C441839996AE4F7882030F3F22C2@xmb-sjc-216.amer.cisco.com>
Thien,

I apologize for not responding sooner.  I worked around the issue and
have everything running.  I also apologize because I am not going to be
able to run the make && make check as a non-root user as you have
requested on the system.  I was under a serious time crunch to get this
done.  Now that it is running I don't want to break anything.  But,
should I re-install (which may happen, we'll see what the users think) I
will run all of this as non-root and advise everyone of the outcome.

Thanks for the assist!

Thanks

Mike Dooley - Customer Support Engineer
Cisco Heartland TAC
972-204-8298

CCNA CCOS
OCP Database 10g
OCA Application Server 10g


-----Original Message-----
From: Thien-Thi Nguyen [mailto:[email protected]] 
Sent: Saturday, January 07, 2012 2:02 PM
To: Karl Berry
Cc: Mike Dooley (midooley); [email protected]
Subject: Re: 2 of 36 test failed RCS5.8 - final update

() [email protected] (Karl Berry)
() Fri, 6 Jan 2012 15:10:20 -0800

       That's correct; "make check" is not designed to be run as root.

   Then how about having it complain in that case?
   No real need for perfect detection, seems to me, so something like
this
   could suffice:
   if test x`id -u 2>/dev/null` = x0; then
     echo "make (check) in rcs cannot be run as root." >&2
     exit 1
   fi

   Of course, it would be better to simply support running make check
   as root, but I don't know what the issue is with that.

Thanks for the raising this point.  I think i will try to identify those
tests that are sensitive to being run as root and skip them (exit 77).