Re:Microsoft blames monster email outage on software defect ...
by hinkley on Tuesday November 17, 2026 @11:57AM
(#50534380)
This is the “common sense intuition isn’t” situation. Most people don’t even get this far, as is evidenced by their surprise at why things take so long to get back to normal.
Also “9 minutes to clear” only tells how long the congestion is, NOT what the customer experiences. On a typical response time of .6s, after the delay the first customer sees 101x the usual response time tapering off to 2x almost 9 minutes later. That is a very different story than, “things will be slow for 9 minutes”. It’s everything will be an absolute shitshow for 9 minutes.
Re:Microsoft blames monster email outage on software defect ...
by Choad_20261104 on Tuesday November 17, 2026 @02:44PM
(#50537082)
thanks guys
Re:Microsoft blames monster email outage on software defect ...
by cowlevel on Tuesday November 17, 2026 @09:48PM
(#50541237)
Related: if a network link is at 90% utilisation, then on average each packet will be stuck behind 9 other packets, in a simplistic model. There's a 90% chance the link is already occupied when the new packet arrives, an 81% chance that slot and the next one are both already occupied, a 72.9% chance that slot and the next two are already occupied, etc. the average delay is 1/(1-r) which is 10 at 90%, 100 at 99%, etc. this works for any type of queue, such as a line to buy movie tickets.
If you get 990 with a capacity of 1000, it takes 99 minutes to clear that same disruption. And if you get 1000 with a capacity of 1000, average time to recover from a little random fluctuation is infinity.
I believe normal practice is to upgrade fiber-optic lines when they're somewhere between 50% and 80% utilisation.