Exit status of vector when sink does not stop gracefully #16767
-
Hi: I am testing vector job in kubernetes using job workflow on aws eks cluster. When the eks cluster is updated and vector is not able to shutdown the sink, it kills the components as shown in the logs below. However, it looks like the job is marked a completed and pod is not restarted. It looks like the relevant line is vector/src/topology/running.rs Line 146 in 8377429 I was trying to find out what is the exit status if the vector kills a component that cannot be shutdown gracefully. Here are logs from vector:
|
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
Unfortunately it exits 0 still in that case. #13731 is tracking fixing that. Depending on your use-case, maybe your restart policy should restart Vector any time it exits rather than only when exiting non-zero? |
Beta Was this translation helpful? Give feedback.
Unfortunately it exits 0 still in that case. #13731 is tracking fixing that. Depending on your use-case, maybe your restart policy should restart Vector any time it exits rather than only when exiting non-zero?