anthony-o / cmr

Dockerfile for inspectIT CMR

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Dockerfile for inspectIT APM solution

This docker file bundles the central server of the open source APM solution www.inspectit.rocks.

Quick Start

The easiest way of starting the CMR is to execute

$ docker run -d --name inspectIT-CMR -p 8182:8182 -p 9070:9070 inspectit/cmr

This starts a CMR with default configuration (meaningful for most setups). The image is created with all necessary volumes to provide persistent data storage. (see the volumes section for more detail)

Now get the inspectIT client from our download page, remember to choose the correct version (Windows, Linux or Mac, 32bit or 64bit). The release version has to match the version of your CMR (see section "Running a specific version of the CMR" below)

Configuration

Adjusting the buffer size

The inspectIT CMR keeps most of its data in a memory buffer, the default size is set to 1.5 GB. You can change the size by setting the environment variable BUFFER_SIZE:

$ docker run -d --name inspectIT-CMR -p 8182:8182 -p 9070:9070 -e BUFFER_SIZE=6000 inspectit/cmr

This will set the buffer size to 6000 MB. Warning: Never set the buffer size larger than the available system memory!

Running with additional data container

Best practice at least for the current development state of Docker is to separate the service from the data by two different containers. This approach allows to keep track of the data by the data container.

  1. Start the data container.
$ docker run -d --name inspectIT-CMR-Data -v /CMR/db -v /CMR/storages -v /CMR/config -v /CMR/ci inspectit/cmr true
  1. Start the service container and use the volumes of the data container
$ docker run -d --name inspectIT-CMR --volumes-from="inspectIT-CMR-Data" -p 8182:8182 -p 9070:9070 inspectit/cmr
  1. Backup / Access data / etc (work in progress)

Running with externally mapped volumes

Some people like to run Docker images with volumes being mounted to the host system. Personally, I think this is bad practise as this hinders one of the core strength of Docker: portability, but sure this is a possible setup. Advantage is that you have the persistent data easily accessible from your host. To run the image in this fashion, run

$ docker run -d --name inspectIT-CMR -p 8182:8182 -p 9070:9070 -v [local-folder]:/CMR/db -v [local-folder]:/CMR/storage inspectit/cmr

In addition, you can store Preferences (config folder), Environments and Profiles (ci folder) on your host system. We suggest the following two ways:

Named volumes (stored in docker/volumes)

$ docker run -d --name inspectIT-CMR -p 8182:8182 -p 9070:9070 -v config:/CMR/config -v ci:/CMR/ci inspectit/cmr

External mapped volume

$ docker run -d --name inspectIT-CMR -p 8182:8182 -p 9070:9070 -v [local-folder]:/CMR/config -v [local-folder]:/CMR/ci inspectit/cmr

Please be aware that you need to extract the data of the two folders before you can use the upper command the first time. Otherwise empty folders will be mounted which leads to a CMR startup failure. Data extraction can be done on a running container in the following manner

$ docker run -d --name inspectIT-CMR inspectit/cmr
$ docker cp inspectIT-CMR:/CMR/ci [local-folder]
$ docker cp inspectIT-CMR:/CMR/config [local-folder]

Running a specific version of the CMR

The image inspectit/cmr:latest always refers to the latest beta version of inspectIT. If you want the stable build (currently 1.5.x) or a specific version, please use the corresponding version tag instead of latest.

Important: The inspectIT agent version has to match exactly the version of inspectIT CMR! A version check is implemented since 1.6.3.

Volumes

Several volumes are created to manage the storage.

  • Metric database: All metric data is currently stored in a database on the CMR. The folder db is the root folder for data storage.
  • Storages: inspectIT allows to persist invocation sequences in so-called storages. All storages are kept on the server in the folder storage
  • Server Configuration: Configuration entries are stored below the folder config
  • Agent configurations: (not integrated yet) The central server stores the instrumentation and configuration for the agents. The folder ci is used as root folder.

Checking log output

Easiest way to check the log output is to use

$ docker logs inspectIT-CMR

If you want deeper control, connect to the running container with a shell (note: this only works with docker 1.3 and higher)

$ docker exec -it inspectIT-CMR /bin/sh

InspectIT agent

The CMR collects performance data from the inspectIT agent. There are already some preconfigured docker images available:

If you have another server, for example Weblogic, please refer to our documentation or write a comment.

Build the docker image

If you want to build the inspectIT CMR image yourself, checkout our repository and run

$ docker build -t inspectit/cmr .

You need a docker installation or the - fantastic - boot2docker if you are on Windows or Mac

Issues

If you have problems with this image or any questions feel free to create an issue or leave a comment below.

Contributions

If you wan't to contribute to this image, please look at the sources at github and send us a pull request.

About

Dockerfile for inspectIT CMR


Languages

Language:Shell 100.0%