Day 6 results: killing a server behind a balancer
Measured on: (machine, OS, Python version)
Output
Paste part 1, part 2, part 3 and the scoreboard here.
One sentence per prediction
- P1 (fail % after kill, no health check):
- P2 (error window with health checks):
- P3 (user errors with health checks + retry):
- P4 (least connections vs round robin throughput):
The number that matters
Same crash, failed requests with nothing: __ with health checks + retry: __. In one line, what made the difference:
Part 3: the slow backend
Round robin completed __, least connections completed __, and least connections sent the slow backend __ vs __ requests. Why: