Docker image Redis auth issue/forcing redis requirement?

I’ve ran NC for a good few months (migrated from OC) and ran it with redis from the get-go. Today i notice i’m having an issue with my NC container and after a little bit of diag, i see it appears to be a redis issue (output from nextcloud.log)

{“reqId”:“nUakFPBhkkgfizv218pG”,“level”:3,“time”:“2021-01-06T10:35:37+00:00”,“remoteAddr”:“”,“user”:"–",“app”:“index”,“method”:“GET”,“url”:"/nextcloud/index.php",“message”:{“Exception”:“RedisException”,“Message”:“ERR AUTH called without any password configured for the default user. Are you sure your configuration is correct?”…

so i stop, remove and start a fresh container of both redis and NC without ‘–env REDIS_HOST=’:

podman run -d
–label “io.containers.autoupdate=image”
–name nextcloud
–publish 8085:80
–volume /mnt/nextcloud/:/var/www/html
–env MYSQL_DATABASE=nextcloud
–env MYSQL_USER=nextcloud

but it still seems to be needing redis with the above nextcloud.log error. Is there something in the DB that i need to kill thats referencing redis? i did have some issues with my home server with my volume side going offline but sql remained online at the time. thats the only issue i can think of as a possible cause.

i use the docker _/redis release that explicitly disables password usage by default so why nextcloud is seemingly having issues with redis auth, i’ve no idea.

fwiw, redis is configured as:

podman run -d
–label “io.containers.autoupdate=image”
–name redis
–publish 6379:6379
–hostname redis

How do i fix the redis auth issue? i’d prefer to keep using redis primarily for file lock handling but it seems even under testing the NC docker image seems to continue to use it even when not defined in the config.
NC is currently offline so any help would be useful :slight_smile:

looks like it’s known/fixed a few months ago: [SOLVED] Latest Docker image broke the installation (Redis password auth) - #16 by esvalot