posix

conformance-posix

This binary runs an HTTP web server that accepts POST HTTP requests to an /add endpoint. This endpoint takes arbitrary data and adds it to a file-based log.

Bring up a log

This will create a directory in your filesystem to store a log, and start a personality binary that can add entries to this log.

First, define a few environment vaiables:

export LOG_PRIVATE_KEY="PRIVATE+KEY+example.com/log/testdata+33d7b496+AeymY/SZAX0jZcJ8enZ5FY1Dz+wTML2yWSkK+9DSF3eg"
export LOG_PUBLIC_KEY="example.com/log/testdata+33d7b496+AeHTu4Q3hEIMHNqc6fASMsq3rKNx280NI+oO5xCFkkSx"
export LOG_DIR=/tmp/mylog

Then, start the personality:

go run ./cmd/conformance/posix \
  --storage_dir=${LOG_DIR} \
  --initialise \
  --listen=:2025 \
  --v=2

Add entries to the log

Manually

Head over to the codelab to manually add entries to the log, and inspect the log.

Using the hammer

In another terminal, run the hammer against the log. In this example, we're running 32 writers against the log to add 128 new leaves within 1 minute.

go run ./internal/hammer \
  --log_public_key=example.com/log/testdata+33d7b496+AeHTu4Q3hEIMHNqc6fASMsq3rKNx280NI+oO5xCFkkSx \
  --log_url=http://localhost:2025 \
  --max_read_ops=0 \
  --num_writers=32 \
  --max_write_ops=64 \
  --max_runtime=1m \
  --leaf_write_goal=128 \
  --show_ui=false

Optionally, inspect the log on the filesystem using the woodpecker tool to see the contents. Note that this reads only from the files on disk, so none of the commands above need to be running for this to work.

go run github.com/mhutchinson/woodpecker@main \
  --custom_log_type=tiles \
  --custom_log_url=file:///${LOG_DIR}/ \
  --custom_log_origin=example.com/log/testdata \
  --custom_log_vkey=${LOG_PUBLIC_KEY}