Date: Tue, 20 Apr 1999 19:24:50 +0100 From: Stuart Henderson <stuart@eclipse.net.uk> To: "Michael R. Wayne" <wayne@staff.msen.com> Cc: freebsd-isp@FreeBSD.ORG Subject: Re: Large email installation references? Message-ID: <371CC672.501A5050@eclipse.net.uk> References: <199904201746.NAA06139@manor.msen.com>
next in thread | previous in thread | raw e-mail | index | archive | help
"Michael R. Wayne" wrote: > > Are there any good reference pointers to suggestions and issues > dealing with large (100,000+ accounts) email installations? Yes, don't use qpopper :-) (it copies the mail spool to a temp file which in most cases is not exactly optimal for performance. ;-) There is a paper on www.netapp.com about Earthlink's email system however it is a bit overly netapp-centric and doesn't give much in the way of details about their hacks to sendmail to make it work with nfs mounted spool (hint: don't bother, exim is fairly nfs-resistant anyway because one of Exim's coders is a "Core Systems Zookeeper" at a large no-monthly-charge ISP in the uk who have bought rather a lot of netapp boxes and if you're holding that much email you probably want the least possible hassle :) For tips see MX records and associated ports for <wildcard-mx?> anything.freeserve.co.uk and pop3.freeserve.net and play "guess the pop3 server" (although I think probably anything they use will be fairly seriously hacked into submission). (well I guess at 0.60GBP/minute for tech support they want to increase the confusion with dns .co.uk/.net domain names as much as possible). For anyone wondering, they make their money from call revenue (both dialup IP and tech support) and advertising on their website and use transparent netapp caches all over the place. An interesting case study ;-) As to Lockheed's alphas, is it *possible* to centrally manage 135 NT servers? Yeuuuuw! Stuart To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?371CC672.501A5050>