> From: Gordon Marx [mailto:[email protected]] > > On Sep 16, 2014, at 7:00 PM, "Edward Ned Harvey (blu)" > <[email protected]> wrote: > > You receive notification that your production server is down, and your > customers are being unserved and your business is losing $10k per minute. > > > > Are you going to checksum all of your system binaries before starting the > service manually? > > My production server that costs me $10k/minute for downtime is actually not > one server, or located in one datacenter. If you're interested in solving > problems at scale, talk to me about Akamai, this is basically all we do all > day.
Tru dat, but take away the exaggerated $10k/minute cost of an individual production server, and the point still stands, that the manual restart as security measure is a fallacy. I cannot say there is never a situation where manual restart is beneficial, but I *can* say that auto restart is better in most situations, because most crashes are caused by accidental conditions for which the best reaction is to restart, and then research root cause. More importantly, I can say, that even when a crash occurs as a result of malice, the manual restart is ineffective as a countermeasure. _______________________________________________ Discuss mailing list [email protected] http://lists.blu.org/mailman/listinfo/discuss
