uptime-tracker is a dead simple tool to track the uptime of a network of nodes. It is designed to be run on multiple nodes and will log the uptime of each node in the network. The logs are stored in a simple text file for easy access.
TCP sockets are used to communicate between nodes. Each node will send a message to every other node in the network. If a node does not receive a message from another node within a set time, it will increment a timeout counter. If the timeout counter reaches a set limit, the node will be considered down and the uptime will be logged.
Copy the config.example.ts file to config.ts and modify the configuration as needed.
To run the program you (TODO:) can download the release from the releases page or build it yourself.
TODO: The releases also contain a LXC template and a Dockerfile to run the program in a container.
The logs are stored in the logs directory. Each log file is named <YYYY-MM-DD>-<NODE_NAME>.log. The logs are in a simple text format and can be viewed with any text editor.
TODO: You can also use uptime-viewer to view the logs.
To build the program you will need to have Node and Bun installed.
To install dependencies:
bun installTo run:
bun run index.tsYou can also use --watch to automatically restart the program when a file changes:
bun run --watch index.ts<YYYY-MM-DD>-<NODE_NAME>.log
<> means a variable in the config and | is a seperator not OR
--- means restart logging
after restart, the first line is information about current configuration
<version>|<ISO8601>|<offlineThreshold>|<current node name>| and then every node in the network in the format <node name>:<node ip>: <node port> with nodes seperated by !
after initialization, a new line is added when a node changes state
<ISO8601>|<node name>|<timeouts> where <timeouts> is the number of timeouts the node has had (ranging from 0 to <offlineThreshold>)
the current log version is 0.0.3