tail2kinesis - Like 'tail -f', outputs to Amazon Kinesis, optionally transforms the data
Usage:
tail2kinesis -h | --help
tail2kinesis -v | --version
tail2kinesis --stream-name=STREAM_NAME [--auth=AUTH_TYPE] [--role-arn=ROLE_ARN] [--region=REGION] [--transform=TRANSFORM] [--endpoint=ENDPOINT] [--log-level=LOG_LEVEL] <filename>
Options:
-h --help
-v --version
--auth=AUTH_TYPE
Get the credentials from the environment (AWS_ACCESS_KEY_ID/AWS_SECRET_ACCESS_KEY/AWS_SESSION_TOKEN),
the EC2 metadata service or STS AssumeRole. For 'metadata', the calling instance to be attached to
an instance profile. For 'assumerole', you need to pass '--role-arn'.
Allowed: {env,metadata,assumerole} - [default: env].
--role-arn=ROLE_ARN
The Amazon Resource Name (ARN) of the role to assume.
--region=REGION
AWS region. If not specified, the AWS_DEFAULT_REGION environment variable is used.
No region is required when using the --endpoint option.
--stream-name=STREAM_NAME
Name of the Kinesis stream that will receive the data.
It must exist.
--transform=TRANSFORM
Every input line is passed to the given TRANSFORM function.
Allowed: {identity,accesslog-query} - [default: identity].
--endpoint=ENDPOINT
Kinesis API endpoint, mostly useful in testing.
--log-level=LOG_LEVEL
Log level.
Allowed: {trace,debug,info,warn,error,fatal,panic} - [default: error].
Arguments:
<filename> Name of the file being tailed, it does not need to exist.
Notes:
* Lines are batched. Batches are sent to Kinesis no more than once per second, or when the maximum
batch size is reached (default: 100 lines).
* There's no "-n" option, files are always tailed from EOF.
make docker-build
make docker-shell
kinesalite &
. /src/kinesis.sh
kCreateStream test
find /src -name '*.go' | entr -d /src/dockerdevsetup.sh
- Configurable partition key
- Configurable flush interval