Date: Wed, 27 Sep 2017 00:04:58 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 222632] Enable Capsicum for connect(2) Message-ID: <bug-222632-8-wydWwkJE8w@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-222632-8@https.bugs.freebsd.org/bugzilla/> References: <bug-222632-8@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=222632 --- Comment #5 from Shawn Webb <shawn.webb@hardenedbsd.org> --- (In reply to Brooks Davis from comment #4) That's essentially what I'm doing. I'm writing a non-capsicumized child process that handles socket and file descriptor creation in addition to file system access calls (like unlink, rename, etc.). I'm using file descriptor passing over a socketpair between the non-Capsicumized child process and its Capsicumized parent. It seems, though, that CAP_CONNECT shouldn't even exist if connect(2) doesn't support Capsicum. All of FreeBSD's official documentation leads developers like me to believe that calling connect(2) within a capabilities enabled process is possible. It seems non-intuitive that I have to wrap connect(2) via the file descriptor passing mechanism I've designed, especially with CAP_CONNECT exists. -- You are receiving this mail because: You are the assignee for the bug.help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-222632-8-wydWwkJE8w>