s3-kinesis-replay

a cli for replaying historical kinesis messages via an s3 archive (e.g. s3 firehose delivery destination)

Installing

Download

Currently s3-kinesis-replay ships binaries for OSX and Linux 64bit systems. You can download the latest release from GitHub

$ wget https://github.com/cludden/s3-kinesis-replay/releases/download/{version}/{artifact}

Building From Source

$ make

Getting Started

View usage information:

$ s3-kinesis-replay --help
a cli for replaying historical kinesis messages via an s3 archive

Usage:
  s3-kinesis-replay [flags]

Flags:
      --bucket string                         s3 archive bucket name
      --delimiter string                      optional delimiter regexp
      --format string                         parser format
  -h, --help                                  help for s3-kinesis-replay
      --json-concurrency int                  json parser concurrency (default 4)
      --json-schema string                    json parser schema path
      --kinesis-backoff-interval string       kinesis backoff interval
      --kinesis-backoff-max-interval string   kinesis max backoff interval
      --kinesis-buffer-window string          kinesis buffer window size
      --kinesis-endpoint string               kinesis endpoint override
      --kinesis-region string                 kinesis region override
      --log-level string                      log verbosity level
      --partition-key string                  json parser parition key path
      --prefix string                         s3 archive prefix
      --replace string                        optional replace regexp
      --replace-with string                   optional replacement string
      --s3-concurrency int                    s3 download concurrency (default 4)
      --s3-region string                      s3 archive region
      --start-after string                    s3 archive start-after key
      --stop-at string                        s3 archive stop-at key
      --stream-name string                    target kinesis stream name

Basic usage with all required flags: (assumes aws environment is configured)

$ s3-kinesis-replay \
    --bucket my-bucket \
    --prefix 2018/01 \
    --stream-name my-stream \
    --format json \
    --partition-key path.to.partitionKey

Note: It is worthwile to use this application from within the same region as your s3 bucket & kinesis stream to avoid s3 egress charges

Configuration

s3-kinesis-replay uses viper for configuration, and accepts a configuration file, environment variables, cli flags, or a combination of all three.

Key	EnvVar	Flag	Description	Required	Default
json.concurrency	JSON_CONCURRENCY	--json-concurrency	number of parser goroutines		4
json.partition_key	JSON_PARTITION_KEY	--partition-key	path to json field holding paritition key	true
json.schema	JSON_SCHEMA	--json-schema	path to json schema file
kinesis.backoff_interval	KINESIS_BACKOFF_INTERVAL	--kinesis-backoff-interval	duration string for initial backoff		1s
kinesis.backoff_max_interval	KINESIS_BACKOFF_MAX_INTERVAL	--kinesis-backoff-max-interval	duration string for max backoff		10s
kinesis.buffer_window	KINESIS_BUFFER_WINDOW	--kinesis-buffer-window	duration string for buffer window		10s
kinesis.endpoint	KINESIS_ENDPOINT	--kinesis-endpoint	optional endpoint override
kinesis.region	KINESIS_REGION	--kinesis-region	target stream aws region, will fall back to AWS_REGION environment variable
kinesis.stream_name	KINESIS_STREAM_NAME	--stream-name	target kinesis stream name	true
log.format	LOG_FORMAT	--log-format	supports `json` or `text`		json
log.level	LOG_LEVEL	--log-level	logging verbosity		info
parser.delimiter	PARSER_DELIMITER	--delimiter	used to split s3 objects into multiple messages
parser.format	PARSER_FORMAT	--format	the parser to use. currently `json` is the only supported parser	true
parser.replace	PARSER_REPLACE	--replace	an optional replace regex patter
parser.replace_with	PARSER_REPLACE_WITH	--replace-wth	an optional replacement string
s3.bucket	S3_BUCKET	--bucket	the s3 bucket name that contains the archive	true
s3.concurrency	S3_CONCURRENCY	--s3-concurrency	the number of goroutines to use for downloading s3 objects		4
s3.endpoint	S3_ENDPOINT	--s3-endpoint	an optional s3 endpoint override
s3.prefix	S3_PREFIX	--prefix	an optional s3 key prefix
s3.region	S3_REGION	--s3-region	the s3 bucket region, will fall back to `AWS_REGION` environment variable
s3.start_after	S3_START_AFTER	--start-after	start scanning after this s3 key
s3.stop_at	S3_STOP_AT	--stop-at	stop scanning at this s3 key

Contributing

Fork it
Create your feature branch (git checkout -b my-new-feature)
Commit your changes using conventional changelog standards (git commit -am 'feat: adds my new feature')
Push to the branch (git push origin my-new-feature)
Ensure linting/tests are all passing
Create new Pull Request

Testing

Prerequisities:

Docker CE & Compose

Testing:

# run short tests
$ docker-compose run test make test-short

# run all tests
$ docker-compose run test make test-verbose

# run tests with coverage
$ docker-compose run test make test-coverage

Linting:

# run lint
$ docker-compose run test make lint

License

Licensed under the MIT License

obsecurus / s3-kinesis-replay Goto Github PK