going thru hellhole of system updates ...
MongoDB, NodeJS, Elasticsearch, Java, Docker, RabbitMQ, epel, plus deployment of ~ 600 Tensorflow containers. All clustered, of course. And expected to be a 0 downtime thingie.
After our IT frontline gorillas rebooted all of the database nodes over the week end, crashed the message queueing system (beyond easy recovery) and 2 country heads decided to call me directly ...
Worn out and tired.
I want a 24V lead battery, soap, clamps to teach those gorillas how to take care of a delicate system (it's handling several ten thousand translations/day ...)