Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 734ED475A1E for ; Thu, 24 Oct 2002 17:15:23 -0400 (EDT) Received: from tomts19-srv.bellnexxia.net (tomts19.bellnexxia.net [209.226.175.73]) by postgresql.org (Postfix) with ESMTP id B80CE4759F2 for ; Thu, 24 Oct 2002 17:15:22 -0400 (EDT) Received: from cbbrowne.com ([64.229.208.36]) by tomts19-srv.bellnexxia.net (InterMail vM.5.01.04.19 201-253-122-122-119-20020516) with ESMTP id <20021024211524.WONU1527.tomts19-srv.bellnexxia.net@cbbrowne.com> for ; Thu, 24 Oct 2002 17:15:24 -0400 Received: from cbbrowne.com (localhost [127.0.0.1]) by cbbrowne.com (Postfix) with ESMTP id 3D1BEC86F for ; Thu, 24 Oct 2002 17:12:14 -0400 (EDT) From: cbbrowne@cbbrowne.com X-Draft-From: ("nntp+chvatal:comp.databases.postgresql.questions" 7801) Newsgroups: comp.databases.postgresql.general, comp.databases.postgresql.questions To: pgsql-hackers@postgresql.org Subject: Re: Hot Backup References: <20021024161128.H31360@mail.libertyrms.com> Organization: cbbrowne Computing Inc X-Home-Page: http://www.cbbrowne.com/info/ X-Emacs-Acronym: Embarrassed Manual-Writer Accused of Communist Subversion Microsoft: Where the service packs are larger than the original releases. X-Uboat-Death-Message: ANNOYED BY ATOMIC BOMB. DEPTH CHARGES. SINKING. U-854. User-Agent: Gnus/5.090008 (Oort Gnus v0.08) XEmacs/21.4 (Honest Recruiter, i386-debian-linux) X-Mailer: mh-e 6.1; nmh 1.0.4+dev; Emacs 21.4 Date: Thu, 24 Oct 2002 17:12:14 -0400 Message-Id: <20021024211214.3D1BEC86F@cbbrowne.com> X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200210/1109 X-Sequence-Number: 30878 The world rejoiced as andrew@libertyrms.info (Andrew Sullivan) wrote: > Having undertaken the exercise, I really can say that it is a little > strange to think about what would happen to data I am in charge of in > case a fairly large US centre were completely blown off the map. But > with a little careful planning, you actually _can_ think about that, > and provide strong assurances that things won't get lost. But it > doesn't pay to call such questions "silly", because they are > questions that people will demand answers to before they entrust you > with their millions of dollars of data. I was associated with one data center that has the whole "barbed-wire-fences, 40-foot-underground-bunker, retina-scanning" thing; they apparently /did/ do analysis based on the site being a potential target for nuclear attack. Realistically, two scenarios are much more realistic: a) The site resides in a significant tornado zone where towns occasionally get scraped off the map; b) The site isn't far from a small but busy airport, and they did consciously consider the possibility of aircraft crashing into the building. Presumably by accident, not by design; the company owns quite a number of jet aircraft, so that vulnerabilities involving misuse of aircraft would rapidly "fly" to mind... (Painfully and vastly moreso since 9/11, of course :-(.) When doing risk analysis, it is certainly necessary to consider these sorts of (admittedly paranoid) scenarios. It's a bit fun, in a way; you get to look for some pretty odd-ball situations; the "server room being overrun by Mongol Hordes." That particular one isn't too likely, of course :-). -- (reverse (concatenate 'string "moc.enworbbc@" "sirhc")) http://www.ntlug.org/~cbbrowne/advocacy.html "I've discovered that P=NP, but the proof is too long to fit within the confines of this signature..."