Please check this Stackoverflow answer
To run all of haystack and its components, it is suggested to change the default in docker settings from 2GiB to 4GiB
docker-compose -f docker-compose.yml \
-f traces/docker-compose.yml \
-f trends/docker-compose.yml \
-f service-graph/docker-compose.yml \
-f agent/docker-compose.yml up
The command above starts haystack-agent as well. Give a minute or two for the containers to come up and connect with each other. Haystack's UI will be available at http://localhost:8080
Finally, one can find a sample spring boot application @ https://github.com/ExpediaDotCom/opentracing-spring-haystack-example to send data to Haystack via haystack-agent listening in port 34000.
docker-compose -f docker-compose.yml \
-f zipkin/docker-compose.yml \
-f trends/docker-compose.yml \
-f service-graph/docker-compose.yml up
The command above starts Pitchfork to proxy data to Zipkin and Haystack.
Give a minute or two for the containers to come up and connect with each other. Once the stack is up, one can use the sample application @ https://github.com/openzipkin/brave-webmvc-example and send some sample data to see traces (from Zipkin), trends and service-graph in haystack-ui @ http://localhost:8080
Note the two commands above add a series of docker-compose.yml files.
- Haystack needs at least one trace provider ( traces/docker-compose.yml or zipkin/docker-compose.yml ) and one trends provider ( trends/docker-compose.yml )
- One can remove service-graph/docker-compose.yml if
service graphis not required - Staring the stack just with with base docker-compose.yml, will start core services like kafka, cassandra and elastic-search along with haystack-ui with mock backend
docker-compose -f docker-compose.yml up