Update Redis backend

[gaurav]

We've been having some issues with the Redis servers behind NodeNode recently. In order to debug this issues, it might help to fix some issues in the Kubernetes file we use to set this up at https://github.com/helxplatform/translator-devops/tree/develop/helm/redis-r3-external. Some possible improvements:

• [ ] redis-server is currently the top-level process within these nodes. It would simplify debugging and testing if instead we used something like tini as the init process, as we could then restart the Redis instance without rebooting the node.
• [x] As far as I can tell, the current setup is hardcoded to use redis-12.7.4.tgz, which sets up a master-slave setup. We should either replace that with a production version of Redis so we can include improvements and bugfixes as we go, or set up some procedure for updating redis-r3-external at regular intervals.

Note that we don't need to worry about this at all for ITRB, because they run and manage their own Redis cluster instance. At some point it might make sense for us to do this as well, but the Redis-on-Kubernetes setup we currently use seems to work fine as long as it doesn't have random issues.

[gaurav]

the current setup is hardcoded to use redis-12.7.4.tgz

Not true -- it uses the helm dependency system, so you can use https://helm.sh/docs/helm/helm_dependency_update/ to upgrade it.

[gaurav]

We have upgraded Redis in RENCI Dev. I'll keep this open until we've shown that the code continues to work on ITRB.

[gaurav]

This seems to be generally working everywhere. We do need to upgrade Redis to Redis 7 at some point, which we should coordinate with upgrading the databases on ITRB, but that's not particularly urgent. Closing.