Is your feature request related to a problem? Please describe.
Nuvla placement policies shall be based on Docker labels -> Nuvla resource tags. Thus the system-manager should be able to parse all Docker labels assigned to the NuvlaBox deployment and automatically push those into Nuvla, for an easier and complementary tagging mechanism

Describe the solution you'd like
Simply add a procedure in the system manager, at start-up, that looks for all Docker labels that are relevant to NuvlaBox

• Make use of build-push-actions, qemu, and buildx to build images in parallel and accelerate the build process.
• Use cache in docker base images, python (for unittests) and intermediate builds
• Create and independent workflow for development (nuvladev) images
  - [ ] #53 descoped

and consider using the openssl version (a bit smaller and allow more algorithms).

Allow the system manager to use the memory monitoring to perform memory limit controls to prevent memory overloads that runs OOM exception in a system host level.

Feature set:

• set node labels
• reboot
• execute command? (ssh nuvla credential vs system-manager bind mount .ssh/authorized_keys)
• start data_gateway routers
• publish local dashboard temporarily

Is your feature request related to a problem? Please describe.

In case minimum requirements for installation are not met, the system manager prints the following and bails out.

system-manager_1   | ERROR - run.py/run/requirements_check - System does not meet the minimum requirements!
system-manager_1   | Cannot continue...

The actual requirement that is not met is not specified in the error message.

Describe the solution you'd like

The requirements that are not met are logged.

Additional context

system-manager_1   | ERROR - run.py/run/requirements_check - System does not meet the minimum requirements!
system-manager_1   | Cannot continue...

Import CI configuration so that the validation is triggered when PR against main.

docker run -it --net host --pid host --userns host --cap-add audit_control -e DOCKER_CONTENT_TRUST=$DOCKER_CONTENT_TRUST -v /var/lib:/var/lib -v /var/run/docker.sock:/var/run/docker.sock -v /usr/lib/systemd:/usr/lib/systemd -v /etc:/etc --label docker_bench_security \ docker/docker-bench-security

Needs discussion.

Describe the bug
when docker.info() contains special chars, the current quoting will not properly stringify those

... so that the NuvlaEdge can report the minimum requirements that are not met to Nuvla.

Also make "Swarm mode enabled" not a requirement but only a warning or an info.

Remove exposed SYSTEM_MANAGER_PORT. system-manager dashboard is only for local usage and can be accessed within the nuvlaedge internal network. No need to bind an external port

For example, do not stop NuvlaBox Engine if Swarm mode is not enabled but instead continue in a degraded mode and possibly warn the user.

Sometime I see this exception on my RPi4:

Traceback (most recent call last):
  File "/opt/nuvlabox/./run.py", line 103, in <module>
    self_sup.write_container_stats_table_html()
  File "/opt/nuvlabox/system_manager/Supervise.py", line 189, in write_container_stats_table_html
    container_stats = json.load(cstats)
  File "/usr/local/lib/python3.9/json/__init__.py", line 293, in load
    return loads(fp.read(),
  File "/usr/local/lib/python3.9/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/usr/local/lib/python3.9/json/decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/local/lib/python3.9/json/decoder.py", line 355, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

Describe the bug
Because of the supervision checks, the datagateway is restarted when down.
During a shutdown, if unlucky, the datagateway (if stopped first) might be restarted by the system-manager, thus making the down command fail the 1st time.

To Reproduce
docker-compose down

Is your feature request related to a problem? Please describe.
containers that are started as part of the data gateway might end up not being well cleaned, if the corresponding peripheral is removed and network is unstable

Describe the solution you'd like
When getting all containers, check for those labelled as data-gateway specific and double check whether respective peripherals (matching nuvla resource id) are still in the system

Describe the bug

system-manager_1   | Traceback (most recent call last):
system-manager_1   |   File "./app.py", line 127, in <module>
system-manager_1   |     supervisor.build_content()
system-manager_1   |   File "/opt/nuvlabox/system_manager/Supervise.py", line 89, in build_content
system-manager_1   |     self._get_stats_table_html())
system-manager_1   |   File "/opt/nuvlabox/system_manager/Supervise.py", line 127, in _get_stats_table_html
system-manager_1   |     cpu_system = float(container_stats["cpu_stats"]["system_cpu_usage"])
system-manager_1   | KeyError: 'system_cpu_usage'

To Reproduce

Deploy a NB with a restarting container (like vpn-client)

Expected behavior
Simply do not account that container

atm the operational-status is set randomly, possibly inducing a wrong status.

instead, improve the mechanism in such a way that different checks in the system can contribute to the final assessment

Describe the bug
API-SERVER expect for status-notes a vector without blank strings.
When there is multiple lines, some empty lines are sent to api-server that will return a spec validation failure.
Perhaps we should relax the spec to accept vector of strings even if empty to be less error prone

To Reproduce
When we have a status-notes like this one:
cat /var/lib/docker/volumes/nuvlabox_nuvlabox-db/_data/.status-notes

System does not meet the minimum requirements!
	* Your device only provides 977.61 MBs of memory. MIN REQUIREMENTS: 1024 MBs

Cannot connect nuvlabox_agent_1 to Data Gateway network

Steps to reproduce the behavior:

1. Deploy with '...'
2. Go to '....'
3. Click on '....'
4. See error

Expected behavior
No spec issue

Allow to set the encryption flag for the nuvlabox-shared-network network via en environment variable.