hi,

without any apparent reason, and without anything new from my side, the n8n workflow crashed in prod during the night.

Here are the last logs of the worker:

Worker finished execution 389 (job 971)

[runnner:js] Received SIGTERM signal, shutting down...

[runnner:js] Task runner stopped

[Redis client] read ECONNRESET

Queue errored

read ECONNRESET

[Redis client] read ECONNRESET

Lost Redis connection. Trying to reconnect in 1s... (0s/10s)

Lost Redis connection. Trying to reconnect in 1s... (2.7s/10s)

Lost Redis connection. Trying to reconnect in 1s... (3.7s/10s)

Lost Redis connection. Trying to reconnect in 1s... (6.1s/10s)

Lost Redis connection. Trying to reconnect in 1s... (7.5s/10s)

Lost Redis connection. Trying to reconnect in 1s... (8.5s/10s)

Unable to connect to Redis after trying to connect for 10s

Exiting process due to Redis connection error

And Redis database: "The database container is starting up or transitioning. Please wait a moment and try again."

How is it possible?

How could I fix this so it doesn't happen again?