Date: Wed, 10 Nov 2004 09:45:00 -0800 From: Julian Elischer <julian@elischer.org> To: Xin LI <delphij@frontfree.net> Cc: freebsd-security@freebsd.org Subject: Re: Is there any way to know if userland is patched? Message-ID: <4192539C.6040403@elischer.org> In-Reply-To: <20041110173511.GA2940@frontfree.net> References: <20041110173511.GA2940@frontfree.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Xin LI wrote: > Dear folks, > > I'm recently investigating large scale deployment and upgrading FreeBSD > RELEASE. It's our tradition to bump "RELEASE-pN" after a security patch > is applied, however, it seems that there is less method to determine > whether the userland is patched, which is somewhat important for large > site managements. > > So is "uname -sr" the only way to differencate the patchlevel of a security > branch? I have read Colin's freebsd-update script and to my best of > knowledge this is the only way (and, on condition that we have re-compiled > the kernel and installed it, and reboot'ed). Given the nature of a security > or errata branch, we can expect that no API/ABI changes will occour and it > should be safe to do make installworld/installkernel in any order, and bumping > patchlevel does not mean that a reboot must be done. > > Please correct me if I was wrong, thanks. I upgrade systems by creating packages which contain all upgraded files I have a set of makefiles etc. checked into my local CVS tree that check out a freeBSD tree at a given revision and build it (withlocal patches added) and then extracts out fies according to a list I supply. On completion I check the list in too, so I can theoretically recreate that patch.. I use the package system to keep track of which packages are loaded onto a system, and newer upgrade packages always have earlier ones as dependencies.. > > Cheers,
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4192539C.6040403>