Date: Mon, 11 Apr 2005 09:10:08 +0900 From: Jacques Vidrine <jacques@vidrine.us> To: Jacques Vidrine <nectar@FreeBSD.org> Cc: =?ISO-8859-1?Q?Dag-Erling_Sm=F8rgrav?= <des@des.no> Subject: Re: cvs commit: src/libexec/rexecd rexecd.c Message-ID: <db55a80b0b145835c59ebb5b6f460786@vidrine.us> In-Reply-To: <d48970b320355d1b5337567f0aa4e1e7@FreeBSD.org> References: <200504081119.j38BJoZf036618@repoman.freebsd.org> <6715759dfc36ac3b48dc3efe8cfcd07d@FreeBSD.org> <8664yx2wwd.fsf@xps.des.no> <d48970b320355d1b5337567f0aa4e1e7@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Apr 11, 2005, at 9:07 AM, Jacques Vidrine wrote: > On Apr 8, 2005, at 10:58 PM, Dag-Erling Sm=F8rgrav wrote: >> Jacques Vidrine <nectar@FreeBSD.org> writes: >>> Thanks for fixing, DES. Only, you have re-introduced some of the >>> bugs. You need to use getpwnam_r or copy the contents of the=20 >>> returned >>> struct passwd before continuing. >> >> Hmm, actually it should be sufficient to re-getpwnam() after the call >> to pam_setcred() in the child. There's nothing else that has the >> potential to screw up pwd. > > Yes, it would be sufficient, but wasteful (^_^). But, I'm not too=20 > concerned. (BTW, getpwnam is just a wrapper for getpwnam_r). Also, once resolved, RELENG_5 and RELENG_5_4 need to be sync'd with=20 HEAD before 5.4-RELEASE. Unfortunately, I'm not in a position to do so=20= right now. :-/ Cheers, --=20 Jacques A Vidrine / NTT/Verio nectar@celabo.org / jvidrine@verio.net / nectar@freebsd.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?db55a80b0b145835c59ebb5b6f460786>