log2http watches log files and sends new contents to a specified http endpoint. That's it.
You will need Python >= 3.6.
Install the application like this (preferably in a virtualenv):
pip install log2http
Place a YAML config file like the following at a location of your preference:
-
logfile: /Users/demo/example.log
endpoint: http://endpoint1.example.com/tag/sample
min_lines: 5
-
logfile: /var/log/system.log
endpoint: http://endpoint2.example.com/tag/syslog
min_lines: 2
logfilespecifies the file to watchendpointis the http endpoint you want to send the collected log lines to (as POST). Usually, this will be the url of a logging service, e.g.http://logs-01.loggly.com/bulk/token/tag/example/min_linessets the minimum of lines that must be collected until a http request is made (use this to minimize http overhead per logged event)
Once installed, you can launch log2http from your terminal:
log2http --config=/path/to/the/config.yml
When log2http is running, create some sample events to see if they are being collected. For example in bash:
for i in {1..6}; do echo "hello world" >> example.log; done
The output should tell you if events are being collected and sent.
You could also start the log collector from Python like this:
from log2http import load_config, LogCollector
config = load_config('/your/path/to/config.yml')
collector = LogCollector(config)
with collector:
collector.start() # will run until interrupted
Watch out: log lines always need to end with a newline character (\n). If a line does not end with a newline character, log2http will wait indefinitely until it is written to the file.
See requirements-dev.txt for development requirements.
Run tests with pytest.
Run static type checking with mypy --ignore-missing-imports log2http.