A Kubernetes rolling update with maxUnavailable: 0 still drops requests

DEV Community·11 points·remdore·about 19 hours ago·dev.to

Four replicas, maxUnavailable 0, graceful shutdown and a preStop hook. Still 20 dropped connections per deploy. The timing data showed the failures were nowhere near where I assumed.

Read the full article at dev.to →

#kubernetes #still #maxunavailable #rolling #requests #rolling update #kubernetes rolling #still drops

Related stories

Kubernetes 1.37's kyaml output closes a YAML bug that still deletes data on applydev.to
DEV Community · 7 points · 7 days ago
Stacked pull requests are now in public previewgithub.blog
Lobsters · 57 points · 2 months ago
Ask HN: Should we still learn coding?
Hacker News · 14 points · 10 days ago
Kubernetes in Dockergithub.com
Hacker News · 2 points · 4 days ago
How Kubernetes Actually Schedules Your Pod (and What the Network Does After)dev.to
DEV Community · 16 points · about 17 hours ago
When AI makes challenging last-minute requests no human client cares abouttwitter.com
Hacker News · 1 points · 9 days ago

Related topics

devops49 docker29 still97 security283 portainer3 community39 zero downtime3 secure30 hands20 early33 scheduling3 requests29