Skip to content
This repository has been archived by the owner on Aug 23, 2023. It is now read-only.

Latest commit

 

History

History
81 lines (49 loc) · 2.84 KB

docker.md

File metadata and controls

81 lines (49 loc) · 2.84 KB

docker

tags

metrictank on dockerhub

You can use these tags:

  • latest: the latest official stable release, corresponding to a git tag.
  • master: latest build from master. these versions typically bring improvements but possibly also new bugs

extra docker stacks

You may have already followed our docker quickstart guide but in fact, the metrictank repository contains a variety of docker stacks for different purposes.

They can be started by cd'ing into their directory and running docker-compose up --force-recreate -V

Here's an overview. Note:

  • not everything is detailed here. Plus this needs some more work. For specifics on any setup, please refer to the sources.
  • every environment comes with grafana and cassandra unless otherwise noted
  • every environment automatically provisions useful datasources and dashboards unless otherwise noted

docker-standard

plain, basic environment. uses metrictank docker image (with baked in binaries and configs), statsdaemon. Comes with 1 dashboard: the metrictank dashboard

see our quickstart guide

docker-dev

Similar to docker-standard, but custom metrictank build is loaded, custom configs from scripts/config also: tag support, main+extra dashboards, jaeger tracing and prometheus monitoring for the system and docker containers

docker-dev-bigtable

Same as docker-dev but instead of using cassandra as the backend store and index store it uses Bigtable via a bigtable emulator.

docker-dev-scylla

Same as docker-dev but using scylladb instead of cassandra

docker-dev-debug

Like docker-dev but uses debug image (build with make debug) and exposes port 40000 for delve.

docker-dev-custom-cfg-kafka

Similar to docker-dev in terms of build and configuration. Also:

  • kafka input and clustering backend
  • short chunkspan & numchunks (benefits: easy to trigger mem and mem_and_cass requests, frequent cass saves, notifier messages etc)
  • caddy-proxied endpooints/datasources to check with different org-id auth headers
  • tsdb-gw

docker-cluster

Similar to docker-dev-custom-cfg-kafka

  • uses a cluster of 4 MT's (with 2x replication)
  • separate graphite monitoring server

see our docker-cluster quickstart guide

docker-cluster-query

A cluster just like docker-cluster, but with an extra query layer (of 1 metrictank instance) in front.

docker-chaos

Similar to docker-cluster but used for chaos testing

docker-cosmosdb

Used for testing against cosmosDB. Note: expects you to have a reachable cosmosdb and adjust settings accordingly. Does not auto-load datasources or dashboards.