I believe you've nailed it. I was overthinking the problem (see previous comment), then failing to consider that my first setuid($u) was negating my setuid/sgid bits that were allowing me to setuid(0) in the first place.
In reply to Re^2: An odd failure of setuid(0)
by Llew_Llaw_Gyffes
in thread An odd failure of setuid(0)
by Llew_Llaw_Gyffes
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |