Trying to run vuegraf in a synology container

[Par-B]

First off i'm no python guru (nor am i synology guru either). So there is that...

I found vuegraph in the docker repository from the synology UI and downloaded the image.
I made the vuegraf.json file and stored it in a directory which i mounted to /opt/vuegraph/conf

{
    "influxDb": {
        "version": 2,
        "url": "http://192.168.1.200:8086",
        "org": "*****",
        "bucket": "Vuegraf",
        "token": "********",
        "reset": false
    },
    "accounts": [
        {
            "name": "Watkins",
            "email": "P***@****.com",
            "password": "*****"
        }
    ]
}

I masked my IP, API key, Organization, Email and vue password for security reasons

I should mention that both influxdb and grafana runs in separate containers on the synology device and i feed them data from hubitat. This works great. I created a new bucket for for Vuegraf on influxdb to write to. I'm using influxdb 2.6.1 if that matters.

I could not get the vuegraf python script to run properly on a container on synology. After much wrangling i logged into the synology device and attempted to run the python script directly in a shell. I found that the default python version was python3.8 and that gave me some headache. However python3.9 is installed and can be started by running python3.9.

I installed pip and from there installed the dependencies. I also installed separately the pyEmVue package using pip.

If i run this as a regular user (eg i ssh into my synology and run the command directly) i get the following error message:

$ python3.9 src/vuegraf.py vuegraf.json
Traceback (most recent call last):
  File "/volume2/homes/parb/src/vuegraf.py", line 17, in <module>
    from pyemvue import PyEmVue
ImportError: cannot import name 'PyEmVue' from 'pyemvue' (unknown location)

If i run it as root it seems to execute more but i get a different error message:

$ sudo python3.9 src/vuegraf.py vuegraf.json                  
2023-03-06 07:28:17.633548 | INFO  | Using InfluxDB version 2
2023-03-06 07:28:17.639202 | INFO  | Settings -> updateIntervalSecs: 60, detailedEnabled: False, detailedIntervalSecs: 3600
2023-03-06 07:28:20.192338 | ERROR | Fatal error: (<class 'botocore.errorfactory.UserNotFoundException'>, UserNotFoundException('An error occurred (UserNotFoundException) when calling the InitiateAuth operation: User does not exist.'), <traceback object at 0x7f95705448c0>)
Traceback (most recent call last):
  File "/volume2/homes/parb/src/vuegraf.py", line 246, in <module>
    account['vue'].login(username=account['email'], password=account['password'])
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/pyemvue/pyemvue.py", line 207, in login
    self.auth = Auth(
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/pyemvue/auth.py", line 49, in __init__
    self.cognito.authenticate(password=password)
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/pycognito/__init__.py", line 478, in authenticate
    tokens = aws.authenticate_user(client_metadata=client_metadata)
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/pycognito/aws_srp.py", line 269, in authenticate_user
    response = boto_client.initiate_auth(
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/botocore/client.py", line 530, in _api_call
    return self._make_api_call(operation_name, kwargs)
  File "/var/packages/Python3.9/target/usr/lib/python3.9/site-packages/botocore/client.py", line 960, in _make_api_call
    raise error_class(parsed_response, operation_name)
botocore.errorfactory.UserNotFoundException: An error occurred (UserNotFoundException) when calling the InitiateAuth operation: User does not exist.

Any suggestions to what i should at next to try to figure out why i can't get it started. I figured that if i first can get it started on the command line i stand a better chance of getting it working in a container. Any thoughts?

[jertel]

You're going to be better off sticking with the container for troubleshooting and long-term running. This avoids the headaches of the host machine having different/conflicting versions of python and dependencies, as well as permission issues.

So going back to the container, post a (redacted) copy of the Vuegraf container log so I can see what's going on there. Perhaps it's a simple issue of you needing to specify the correct internal container IP of the InfluxDB2 container. Which, by the way, if Grafana works against the InfluxDB2 container, you can look in Grafana's admin UI and find out what IP it's using to connect. That will help eliminate one unknown at least.