Node dont start after server crash

We run ES 7.13.2 on k8s, with Docker engine. This is simple 3 nodes cluster with 10 indices and less than 100Go of data.

The ES cluster is used to monitore another ES cluster which is working fine (via metricbeat).

Our super infra team have restarted all Docker engine at same time ....

Now, all ES pod container are throwing:

Failed to submit a listener notification task. Event loop shut down?

Most of the time, I manually remove all pods, but now, I really want ES recover by it-self, is that possible?

Capture d’écran 2021-07-08 à 17.10.51

Full logs here: elasticsearch crash · GitHub

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.