Eagle Eye Tester

Eagle Eye Tester is a tool for testing the performance of Eagle Eye live streams. The project was created when we were having latency issues with cameras when they were being accessed via Carson. In an effort to try to narrow down the problem, I created this tool that uses the Eagle Eye API to determine the latency across our cameras. With this information, Eagle Eye admitted that they had an issue on their end and migrated our account to a different data center, which fixed the issue entirely. I'm providing the program I used in the event it is helpful for others in the same situation.

Requirements

• Python 3.8 or later
• An Eagle Eye account
• An Eagle Eye API key

Installation

1. Clone this repo.
2. Enter the directory created.
3. Install the required libraries by running

pip3 install -r requirements.txt

Configuration

The tool expects a JSON config file to be provided via the -c or --config command-line argument. This JSON file has the following options:

Key	Type	Required?	Description
email	String	Y	The email address of the Eagle Eye account
password	String	N	The password for the Eagle Eye account. If not provided, the tool prompts for it.
auth_token	String	Y	The auth token associated with the Eagle eye account.
cameras	Object	Y	The set of cameras to test. The key is a unique name for the camera with no spaces (String). The value is the ESN (String). The ESN for a camera can be found in the Eagle Eye Dashboard by looking at the information in the camera settings.
delay_between_runs_seconds	Integer	N	The amount of time to delay between latency tests. If not specified, defaults to 60 seconds.

A sample config file is provided in eagle_eye.json.

Running The Tool

The tool can be run with the following command run from the directory created. This shows shows a typical execution:

python3 eagle_eye_tester.py --config eagle_eye.json latency 5

As executed, this command would load the config from eagle_eye.json, run 5 latency tests, and print out the results:

Load time for front_door:
	Minimum: 535ms
	Average: 622ms
	Median:  615ms
	Maximum: 685ms
Load time for back_door:
	Minimum: 551ms
	Average: 779ms
	Median:  588ms
	Maximum: 1,466ms

The two possible commands to run are latency <number of runs> and stream <camera name>.

The latency command tries to fetch the first byte from the live stream for each camera specified in the config <number of runs> times, with delay_between_runs_seconds delay between each run. The delay can be useful in the event the act of opening the stream caches any state in the Eagle Eye servers that may affect results (e.g., streams to the on-site bridge that may remain open) and you want to wait for that state to clear. Once all runs are complete, the tool prints out the minimum, average, median, and maximum load time for each stream across all runs.

The stream command continually reads bytes from the stream for <camera name>, which is a camera that must be specified in the config file. If the stream breaks, the tool reopens it. The tool continues to read from the stream until you press Ctrl+C.

There are several command-line options available in the tool:

Option	Required?	Description
--config <file> or -c <file>	Y	The JSON configuration file
-v or --verbose	N	Prints out more detailed information when executing
-h or --help	N	Prints out information about running the tool.

License

See LICENSE.

Communication

Please enter an issue for any questions or problems you have.