Work Vent
Oh look a critical service with no redundancy went down in a not 24/7 datacenter, taking one of our flagship products down completely in that region.
This is the 3rd time this week it is happened.
9 months ago, the first time it happened.. I wrote remediation recommendations for the product's team.
THE BIG ONE WAS MOVING THIS SERVICE TO MODERN HARDWARE WITH IPMI AND HOT SPARE!
Guess what they just did...?
FINALLY SPIN THE SERVICE UP ON ONE OUR MODERN COMPUTE BOXES... and decide to stop using the old and busted.
re: Work Vent
I don't know... but it drives me nuts. This particular product requires a service that doesn't play nice with other instances of it running in the same region.. but there is literally no reason not to have a easily spun up hot spare ready to go especially when the hardware is in place already.