(use na when API docs (Release notes, etc) do not need to be updated)
📝 Summary
In the logs there are requests that got ECONNRESET in the moment there API gets SIGTERM signal. And we know that after that there are successful finished requests. So probably requests are reset because they were dialed during the sigterm (their request time is always arount 1-2ms)
So we should allow requests to finish dialling.
And at the same time the pod is excluded from the kubernetes service, so no new requests will come.
📸 Examples
Put screenshots or response/request examples here!
🛑 Problems
Write any discovered & unresolved problems (link to created Jira issues)
✅ DoD
(use na when API docs (Release notes, etc) do not need to be updated)
📝 Summary
In the logs there are requests that got ECONNRESET in the moment there API gets SIGTERM signal. And we know that after that there are successful finished requests. So probably requests are reset because they were dialed during the sigterm (their request time is always arount 1-2ms) So we should allow requests to finish dialling. And at the same time the pod is excluded from the kubernetes service, so no new requests will come.
📸 Examples
Put screenshots or response/request examples here!
🛑 Problems
💡 More ideas
Write any more ideas you have