Refs. swh/infra/sysadm-environment#5221

I noticed a typo in the configuration key.

The dynamic instance and not the soon-to-be decommissionned static instance.

Refs. swh/infra/sysadm-environment#5214

Refs. swh/infra/sysadm-environment#5214

It was not at the right location.

Refs. swh/infra/sysadm-environment#5214

This is to be run exclusively on saam due to the access to the pathslicing mounted
there.

This enhances the actual template to be able to provide specific node selector and
specific volume mountpoints to add to the main container.

Refs. swh/infra/sysadm-environment#5215

Refs. swh/infra/sysadm-environment#5214

Once deployed, we'll progressively make the existing clients use that instance.

Refs. swh/infra/sysadm-environment#5214

Either global to the deployment or per instance.

And then declare explicitely one for running the intance on saam only (for the zfs
pathslicing access).

Refs. swh/infra/sysadm-environment#5214

And declare explicitely one for the saam-zfs instance.

Refs. swh/infra/sysadm-environment#5214

This instance will be deployed on the saam host to continue using the
/srv/softwareheritage/objects filesystem mountpoints. Specific affinity labels will be
installed on that node to ensure the instance only runs there.

The name of the service & ingress are explicit to avoid name conflict ambiguity on the
various read-write services we have.

Refs. swh/infra/sysadm-environment#5214

According to existing documentation.

Refs. swh/infra/sysadm-environment#5110

The production workers now requires the swh.graph.http_client module.

Refs. swh/infra/sysadm-environment#5211

Activate it in production to match the previous production cooker's configuration.

Refs. swh/infra/sysadm-environment#5211

The 2 pods are not enough to reply to the current load,
mostly vault cooker.
The pods restarts regularly due to the unresponding probes.

As per the documentation already allows it for the reading part.

Refs. swh/infra/sysadm-environment#5110

(spurious push of a tagged swh-charts version)

Refs. swh/infra/sysadm-environment#5110

Except we had 2 nodes, search1 and search on moma.
Each 4 workers (1 thread).

The resources change is done according to the current use we have.

Refs. swh/infra/sysadm-environment#5110

The indexer storage sentry reports were effectively disabled, as the secret
is not mandatory and the -storage- key was missing

Ref. swh/infra/sysadm-environment#5200

For the indexers which work much better with lower latency storage

Should fix sentry issue SWH-VAULT-6E

Refs. swh/infra/sysadm-environment#5183