From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tim Subject: Re: re[2]: [linux-lvm] System Suggestions Message-Id: <20020306180032.B12548@connectlive.com> References: <20020306210600.JKKL21197.imf11bis.bellsouth.net@taz> <1015450169.3240.80.camel@Steven> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <1015450169.3240.80.camel@Steven>; from critch@dtsam.com on Wed, Mar 06, 2002 at 03:29:28PM -0600 Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Wed Mar 6 17:00:03 2002 List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-lvm@sistina.com Whatever other people may say about RedHat, I have found that (with rare and well-documented exceptions) their kernels (as kernel-2.x.x-source and kernel-2.x.x-headers RPMs) accept patches fairly well (if not, I'm comfortable doing it manually) and have been reliable in some extremely high-load environments, when properly customized. RH is most definitely not perfect, but 7.2 is IMHO a perfectly serviceable starting point for large systems. I have used Debian, LinuxPPC, Solaris, AIX, FreeBSD, etc. in the past and they all have their ups and downs, but my *nix systems are either OpenBSD or RedHat these days, and the reason is simply that I can get better support for them than anything else. LVM on RH7.2 has worked well for me, but as I previously mentioned, the support for LSI Logic boards is quite solid, and using hardware enclosures eliminates an enormous number of potential problems. > Although my first thought is to write RH and remind them how much we > hate it when they build things in a non standard way and therefore cause > other standard systems to break. I have several peices of hardware that > are only supported with a RH kernel. I may be braver than some in the > fact that I have removed the RH kernel and used in to build other > distros with it. RH should provide a source version of the kernel in > which you can attempt to put XFS into, then compile, and should still be > able to load the promise driver into that new kernel. It will take some > effort, but should be usable. Of course you could return it as > essentially broken and buy the 3ware card that has better support. -- "To be suspended from the legal profession is the moral equivalent of being ostracized by child molesters." --Ian Rowan