fujiwara-api

A Meraki HUD (heads-up display, so you don't need to open Dashboard)

This project is built on the latest (August 2021) versions of the following:

• Python3
• Flask
• PostgreSQL
• Meraki Dashboard APIs
• ngrok (Required: AuthToken from ngrok.com)
• Pyngrok
• PySNMP
• Docker
• Ubuntu (on roadmap: CentOS support)

What this project gives you:

1. Automated install of all requirements for this Project (on Ubuntu or CentOS)
2. A HUD to see status of your equipment at a high level in the NOC, including:
   a. WAN Links
   b. Device Status (with name + IP)
   c. Advanced details using SNMP
3. A HUD to see events related to your sensors + cameras, we take snapshots when a sensor triggers
4. Container based run time (future)

Install and Setup

1. Clone this Github Repository on to any Ubuntu 20.04 LTS installation

    git clone https://github.com/vladivanovic/fujiwara-api.git

2. On terminal, run localrunSetup.sh (will prompt for sudo) and wait for installation to complete

    $./localrunSetup.sh

3. Open your browser and navigate to http://localhost:5000/ to complete the initial setup
   

4. Trigger a device collection process to populate the HUD device list
   

** Caution **
Docker is still under construction in our project as of September 9th and may not 100% work

** Caution 2 **
When running locally, please utilise tmux or screen or then run 'python3 app/webhook.py' and 'python3 app/engineio.py' after you have completed the initial setup in step 4 above

to-do-list

Feature 1 ()
--- Docker ---
Tasks:

1. Troubleshoot Docker SDK when main app is running in a container itself, there's a trick to be done that I'm troubleshooting through
2. use docker sdk to show the status of the container itself, not just the flask service on there

Feature 2 (pxs)

1. Use webhooks to get real time temp/humidity readings and update on the HUD page (instead of SNMP polling)
2. MV-MT Snapshot to support multiple cameras for different angles
3. Water leak sensor support (future)
4. Investigate the webhook script and fix issue of pre-existing alert profiles going missing.

random idea list

Something to be added here one day

1. Integrate alerting to Webex Teams Bot
2. Create ElasticLogstashKibana and pre-create dashboards for syslog, netflow and containerize
3. Convert Main App to use AJAX for dynamic page loading of newest alerts/device status
4. Automate the engagement of engineio.py's /device and /devicepoll to update device list in the DB and SNMP polls to update devicestatus in the DB
5. Per device create a table to collect SNMP Performance statistics (need a smart way to do this)
6. Modularize the code in app_startchecks to allow for easier container build/runs on sub-functions