Clarity Log Search Tool
This is a Radial Wheel repository for the Clarity Log Search Tool.
Clarity is a simple log viewing tool with a web interface for following log
Since this is a log viewing application, running by itself is pretty useless as
it will only view it's own logs. So if you already use fig to start up a
collection of images (let's assume other Radial Spokes) then you may combine the
contents of all the
fig.yml files to produce a hybrid Wheel. If all the
images keep to the Radial topology guidelines, then they should work
Tunable environment variables; modify at runtime. Italics are defaults.
- $PORT: ["80"] Port to access web server.
- $USER: ["daemon"] User to run web server as in container.
- $LISTEN_ADDRESS: ["0.0.0.0"] Addresses to allow connections from.
- $HTTP_USER: ["anonymous"] User used to login to the web app.
- $HTTP_PASS: [random] Password to use for login user. Default will
generate a random password and display it in the logs.
Radial is a Docker container topology strategy that
seeks to put the canon of Docker best-practices into simple, re-usable, and
scalable images, dockerfiles, and repositories. Radial categorizes containers
into 3 types: Axles, Hubs, and Spokes. A Wheel is a repository used to recreate
an application stack consisting of any combination of all three types of
containers. Check out the Radial documentation for more.
One of the main design goals of Radial containers is simple and painless
modularity. All Spoke (application/binary) containers are designed to be run by
themselves as a service (a Wheel consisting of a Hub container for configuration
and a Spoke container for the running binary) or as part of a larger stack as a
Wheel of many Spokes all joined by the Hub container (database, application
code, web server, backend services etc.). Check out the Wheel
tutorial for some more details on how this works.
Note also that for now, Radial makes use of Fig for all orchestration,
demonstration, and testing. Radial is just a collection of images and
strategies, so technically, any orchestration tool can work. But Fig was the
leanest and most logical to use for now.
How to Use
In case you need to modify the entrypoint script, the Dockerfile itself, create
your "config" branch for dynamic building, or just prefer to build your own from
scratch, then you can do the following:
- Clone this repository
- Make whatever changes needed to configuration and add whatever files
A standard feature of all Radial images is their ability to be used dynamically.
This means that since great care is made to separate the application code from
it's configuration, as long as you make your application configuration available
as a git repository, and in it's own "config" branch as per the guidelines in
the Wheel template, no building of any images will be
necessary at deploy time. This has many benefits as it allows rapid deployment
and configuration without any wait time in the building process. However:
Dynamic builds will not commit your configuration files into any
resulting images like static builds.
Static builds do a "COPY" of files into the image before exposing the
directories as volumes. Dynamic builds do a
git fetch at run time and the
resulting data is downloaded to an already existing volume location, which is
now free from Docker versioning. Both methods have their advantages and
disadvantages. Deploying the same exact configuration might benefit from a
single image built statically whereas deploying many different disposable
configurations rapidly are best done dynamically with no building.
To run dynamically:
- Modify the
fig-dynamic.ymlfile to point at your own Wheel repository
location by setting the
$WHEEL_REPOvariable. When run, the Hub container
will pull the "config" branch of that repository and use it to run the Spoke
container with your own configuration.
fig -f fig-dynamic.yml up