Refs. swh/meta#5048

Refs. swh/meta#5045

This allows to override it for the staging-cassandra-next-version deployment and make it
stop faster.

Prior to this, only the swh application records were properly pushed to the right index.
The system logs were pushed in the unfinished 'staging-logs-' index.

Refs. swh/infra/sysadm-environment#4955

Otherwise, it's not storing records in the right index.

Refs. swh/infra/sysadm-environment#4955

It was an old test which should never have been committed.

Refs. swh/infra/sysadm-environment#4955

Refs. swh/infra/sysadm-environment#4955

After testing this within minikube (branch wip-minikube), this adds the required parts
so the oltp collector agents can push logs in daily index (depending on the data
records).

This first update the chart to a version supporting the dynamic index. Then enables the
dynamic index configuration. It configures a suffix and a prefix so the index to store
data is computed out of those.

Internally, the attribute field record 'elasticsearch.index.prefix' is configured with
the environment value (e.g. staging, production, ...). The attribute field
'elasticsearch.index.suffix' is computed out of date time stored in the record field
'asctime'. It's using the day of the year with format '%Y-%m-%d'.

swh/infra/sysadm-environment#4955

Refs. swh/infra/sysadm-environment#4986

Refs. swh/infra/sysadm-environment#4984

Related to swh/infra/sysadm-environment#4879

Related to swh/infra/sysadm-environment#4983

Related to <TBD>

Related to swh/infra/sysadm-environment#4879

Related to swh/infra/sysadm-environment#4879

Something went went wrong in the incoming-tag > build-docker-images > build-docker-image
tooling. What happened is not clear.

Refs. swh/infra/sysadm-environment#4982

Related to swh/infra/sysadm-environment#4879

Related to swh/infra/sysadm-environment#4879

and align the configuration for production

and disable the postgresql first instance

Related to swh/infra/sysadm-environment#4879

Refs. swh/infra/sysadm-environment#4971

The replayer's code is not calling storage.flush() so a part
of the replayed content is lost when the replayer restart
or is taking a loooonnggg time to be replayed, specially low
occurences objects like snapshot with a default buffer size of 25 000.

Related to swh/infra/sysadm-environment#4879