From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27661 invoked by uid 1002); 23 Jan 2003 19:39:46 -0000 Mailing-List: contact gentoo-dev-help@gentoo.org; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Id: Gentoo Linux mail X-BeenThere: gentoo-dev@gentoo.org Received: (qmail 10960 invoked from network); 23 Jan 2003 19:39:46 -0000 From: Dylan Carlson Reply-To: absinthe@pobox.com To: Dewet Diener , Date: Thu, 23 Jan 2003 14:34:17 -0500 User-Agent: KMail/1.5 References: <200301232105.50700.gentoo@dewet.org> In-Reply-To: <200301232105.50700.gentoo@dewet.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200301231434.17672.absinthe@pobox.com> Subject: Re: [gentoo-dev] Gentoo-sources vs "stock" kernels X-Archives-Salt: 72193aad-eae5-47e0-aa72-801f35ed66d9 X-Archives-Hash: 1670cb3af0180a91590ce32171b5f67f On Thursday 23 January 2003 02:05 pm, Dewet Diener wrote: > Hi all > > I'm wondering what the general status of gentoo-sources is compared to > the more "stock" kernels, like vanilla and -ac? Is it being used in > production-class setups without hitches? I've had only one problem > with gentoo-sources so far, and it seemed to have been > grsecurity-related (random processes getting killed at boot time, from > init onwards), which disappeared as soon as I disabled it from > building. > > I'm preparing to switch a RedHat production-class system over to > Gentoo, and have everything running in a chroot. The only thing I > haven't been able to test is the kernel... I've had trouble with the other kernels along 2.4.20 related to NVIDIA drivers... which is why I'm forced to use vanilla at the present time. Get everything working under vanilla first, and make sure it works correctly under normal usage (and it should). That way you have a working config to fall back on if you have trouble with a different kernel. It's also easier to troubleshoot new kernels and find out exactly what's breaking if you have everything working under vanilla already. If your system doesn't require any closed-source drivers that change a lot (like Nvidia's) other kernels may work fine. Always test them thoroughly before rolling them into production. Put them under load and see how they handle it... almost any kernel will boot up a system fine. If you want to see where the kernel defects are, you need to stress it under lots of memory and cpu load. My bad experiences with kernels in Linux almost always have to do with the VM doing bizarre stuff under load. If you don't have slack in your schedule to do that kind of testing, my advice is to stick with the vanilla kernel. Cheers, Dylan Carlson [absinthe@pobox.com] -- gentoo-dev@gentoo.org mailing list