Date/Time: Sept 9, 2025 Observed Behavior: At this timestamp, I noticed an extreme delay time of ~70,722,000 ms (≈ 19.6 hours) on my pod: • P70: 70,722,309 ms • P90: 70,722,363 ms • P98: 70,722,384 ms So this was cost 35$ and i don't understand why this happened and how to avoid this in future. The endpoint worked for months, and everything was OK, so i doubt it was related to my code.
Expected: Delay times should be within normal ranges. Timeout should work to not wait until whole limit is used. Request: Could someone please investigate if this was a system-level issue (e.g., frozen jobs, infrastructure hang, etc.)?