There's a typo in the return output for /send_command/<job_id>. It returns app: "nass", instead of app: "naas"

There is currently a rudimentary authentication failure lockout mechanism for a single user, however this does not catch the instance where different usernames are spamming the same device. We need to catch these and also temporarily lockout all access to that device.

Mechanism will be similar to existing user based lockout, just with the device's IP as the key.

Need to create initial set of documentation for starting/running the API.

This would include updated Readme and Contributing docs, as well as setting up a Swagger based API doc.

Currently, for every front end API call a user makes, we instantiate a new backend Netmiko connection.

There needs to be a way to pipeline multiple requests by the same user/device pair down an existing connection if any.

Currently, while there is some support for using your own cert, there's no built way to actually get that cert/key into the container. Need to plumb that up.

Need to provide some additional Redis options:

1. Ability to specify Username/PW as well as other options on the redis.Redis redis-py object.
2. Ability to use existing Redis instance and not spin one up with Docker unless needed.

When a job is fetched from the queue, it prints the entire function signature, including any arguments passed to the function. This can include passwords (And does today).

We need to rework credential handling to obfuscate this information in INFO level log output.

hey, trying to get prompt(devices name) out of send_commands function. sometimes it works and returns the name and some times it returns none.

any help?

If I submit a POST to send_command with no Authorization in the header, I get a 500 error returned. This should return a 400-level error.

We're currently using simply logging.basicConfig to setup the loggers, this is making log searches complicated as they're not in the same format as the API container logs.

Need to enhance the logging config to match that of the API via a log formatter.

Today in payload, we ask for device_type however in Netmiko, this is a platform.

See netmiko.platforms for example.

Need to modify the payload parameter to platform, but as this is a client breaking change, am waiting for v1.0.0 release.

If I run POST send_command with an incorrectly spelled "commands" variable in the payload, or if it's missing, I get a proper 400 error with a proper response.

If I run the same command, but instead incorrectly spell or omit the "ip" variable in the payload, I get a 500 error returned, with only the following:

{
    "message": "Internal Server Error"
}