On Sep 2, 2012, at 12:06 AM, Soren Stoutner <so...@stoutner.com> wrote:
> Matt Zimmerman <m...@csh.rit.edu> posted a suggestion in bug #96677 that > Sonar use fping instead of ping, which would not require setuid. Is there > any reason why this is not a desirable fix for this this bug? That would probably work, but it would be a lot of work to write a new sensor module for sonar that read output from fping instead of sending packets itself. If someone writes the code, I hope they send it to me. I'm not interested in writing that code myself because it's a lot of work to solve a non-existent problem. It is a pretty strange and false economy to write thousands of lines of code to avoid a single "chmod" call. For some reason you trust that fping is safe to run setuid, but do not trust sonar. I haven't read the code for fping, but I have read the code for sonar, so I trust the latter more. This is of course a non-problem on MacOS, where non-privileged datagram sockets actually work. Maybe you could get that fixed in the kernel instead. (Ha ha, only serious.) -- Jamie Zawinski http://www.jwz.org/ http://www.dnalounge.com/ -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
