Date: Mon, 09 Feb 2015 23:22:34 +0000 (GMT) From: Rui Paulo <rpaulo@me.com> To: Benjamin Kaduk <bjkfbsd@gmail.com> Cc: "svn-src-head@freebsd.org" <svn-src-head@freebsd.org>, "svn-src-all@freebsd.org" <svn-src-all@freebsd.org>, "src-committers@freebsd.org" <src-committers@freebsd.org>, Rui Paulo <rpaulo@freebsd.org> Subject: Re: svn commit: r278479 - in head: etc sys/kern Message-ID: <bc7b9fa5-40fd-4d87-88d7-1e7a616561d8@me.com>
next in thread | raw e-mail | index | archive | help
On Feb 09, 2015, at 03:16 PM, Benjamin Kaduk <bjkfbsd@gmail.com> wrote:=0A= =0AOn Mon, Feb 9, 2015 at 6:13 PM, Rui Paulo <rpaulo@freebsd.org> wrote:=0A= Author: rpaulo=0ADate: Mon Feb=C2=A0 9 23:13:50 2015=0ANew Revision: 27847= 9=0AURL: https://svnweb.freebsd.org/changeset/base/278479=0A=0ALog:=0A=C2=A0= Notify devd(8) when a process crashed.=0A=0A=C2=A0 This change implements= a notification (via devctl) to userland when=0A=C2=A0 the kernel produces= coredumps after a process has crashed.=0A=C2=A0 devd can then run a speci= fic command to produce a human readable crash=0A=C2=A0 report.=C2=A0 The c= ommand is most usually a helper that runs gdb/lldb=0A=C2=A0 commands on th= e file/coredump pair.=C2=A0 It's possible to use this=0A=C2=A0 functionali= ty for implementing automatic generation of crash reports.=0A=0A=C2=A0 dev= d(8) will be notified of the full path of the binary that crashed and=0A=C2= =A0 the full path of the coredump file.=0A=0AWhat advantage does putting t= his in devd have over a standalone daemon for crash reporting?=C2=A0 Is it= just the ease of implementation to leverage the existing infrastructure?=0A= =C2=A0=0AWell, I want to automatically inspect all the programs that crash= ed in a given system. =C2=A0I don't see how you can do that with a standal= one daemon. =C2=A0Or maybe I didn't understand what you meant.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bc7b9fa5-40fd-4d87-88d7-1e7a616561d8>