In the Kafka world nothing is easy, but esque
(pronounced esk) is an attempt at it.
esque
is a user-centric command line interface for Kafka administration.
Some stuff is hard, and that is okay, but listing your kafka topics shouldn't be.
While adopting kafka at real.digital we noticed the immense entry barrier it poses to newcomers.
We can't recount how often we wrote Slack messages asking for the script to check the
status of topics or consumer groups. This is partly (but not only) due to a
fragmented and unclear definition of tooling and APIs for kafka.
In a wide array of administration tools, esque
distances itself by striving to provide Kafka Ops for Humans, in a usable and natural way.
We feel that the goal of esque
embodies the principle: “keep easy things easy, and make hard things possible”.
- batteries included
- feature rich
- robust
- insightful
- by engineers for engineers
- Support for any type of Kafka deployment >1.2
- Display Resources (Topics, Consumer Groups, Brokers)
- Get detailed Overviews of Resources (Topics, Consumer Groups, Brokers)
- Create/Delete Topics
- Edit Topic Configurations
- Edit Consumer Offset for Topics
- SASL/SSL Support out of the box
- Consume and Produce to and from Avro and Plaintext Topics (including Avro Schema Resolution from Schema Registry)
- Context Switch (Easily Switch between pre-defined Clusters)
- Kafka Ping (Test roundtrip time to your kafka cluster)
$ esque
Usage: esque [OPTIONS] COMMAND [ARGS]...
esque - an operational kafka tool.
In the Kafka world nothing is easy, but esque (pronounced esk) is an
attempt at it.
Options:
--version Show the version and exit.
-v, --verbose Return stack trace on error.
--no-verify Skip all verification dialogs and answer them with yes.
--help Show this message and exit.
Commands:
apply Apply a set of topic configurations.
config Configuration-related options.
consume Consume messages from a topic.
create Create a new instance of a resource.
ctx List contexts and switch between them.
delete Delete a resource.
describe Get detailed information about a resource.
edit Edit a resource.
get Get a quick overview of different resources.
io Run a message pipeline.
ping Test the connection to the kafka cluster.
produce Produce messages to a topic.
set Set resource attributes.
urlencode Url-encode the given value.
esque
is available at pypi.org and can be installed with pip install esque
.
esque
requires Python 3.8+ to run (Python 3.11 is not yet supported).
There are no wheels for Alpine Linux, so esque
requires a few extra dependencies to build them during installation.
apk add python3-dev py3-pip librdkafka librdkafka-dev g++
The installation for Kafka is slightly different for Apple Silicon devices, so simply running pip install esque
may
result in errors.
The fix for this is to first install the librdkafka library with brew install librdkafka
(make note of which version is installed).
Then, add the following to the .zshrc
file with the correct version of librdkafka:
export C_INCLUDE_PATH=/opt/homebrew/Cellar/librdkafka/X.X.X/include/
export LIBRARY_PATH=/opt/homebrew/Cellar/librdkafka/X.X.X/lib/
After starting a new shell (or source ~/.zshrc
), you will be able to install esque
normally with pip.
The autocompletion scripts for bash
and zsh
can be generated by running esque config autocomplete
.
When starting esque
for the first time the following message will appear:
No config provided in ~/.esque
Should a sample file be created in ~/.esque [y/N]:
When answering with y
esque
will copy over the sample config to ~/.esque/esque_config.yaml
.
Afterwards you can modify that file to fit your cluster definitions.
Alternatively might just provide a config file following the sample config's file in that path.
version: 1
current_context: local
contexts:
# This context corresponds to a local development cluster
# created by docker-compose when running esque from the host machine.
local:
bootstrap_servers:
- localhost:9092
security_protocol: PLAINTEXT
schema_registry: http://localhost:8081
default_values:
num_partitions: 1
replication_factor: 1
The config for the apply command has to be a yaml file and is given with the option -f or --file.
In the current version only topic configurations can be changed and specified.
It has to use the same schema, which is used for the following example:
topics:
- name: topic_one
replication_factor: 3
num_partitions: 50
config:
cleanup.policy: compact
- name: topic_two
replication_factor: 3
num_partitions: 50
config:
cleanup.policy: compact
To setup your development environment, make sure you have at least Python 3.8 & poetry installed, then run
poetry install
poetry shell
To install pre commit hooks run:
pip install pre-commit
pre-commit install
pre-commit install-hooks
esque comes with a docker-compose based kafka stack which you can start up with make test-suite
.
You can then run the integration tests against this stack with pytest tests/ --integration --local
.
Alternatively you can go the fast way and just run the whole stack + integration tests in docker:
make integration-test
If you only want the unit tests, just run:
make test