In the master branch there is a code with functions that is under development. The latest release for each network can be found in the Releases section, you can switch to the corresponding tag and build the application.
Please read repo wiki article.
It is only possible to create deb and fat jar packages.
For Ubuntu/Debian:
echo "deb https://dl.bintray.com/sbt/debian /" | sudo tee -a /etc/apt/sources.list.d/sbt.list
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 2EE0EA64E40A89B84B2DF73499E82A75642AC823
sudo apt-get update
sudo apt-get install sbt
You can install sbt on Mac OS X using Homebrew.
Clone this repo and execute
sbt packageAll
.deb and .jar packages will be in /package folder. To build testnet packages use
sbt -Dnetwork=testnet packageAll
sbt test
Note
If you prefer to work with SBT in the interactive mode, open it with settings:
SBT_OPTS="${SBT_OPTS} -Xms512M -Xmx1536M -Xss1M -XX:+CMSClassUnloadingEnabled" sbt
For Java9 it should be:
SBT_OPTS="${SBT_OPTS} -Xms512M -Xmx1536M -Xss1M -XX:+CMSClassUnloadingEnabled --add-modules=java.xml.bind --add-exports java.base/jdk.internal.ref=ALL-UNNAMED" sbt
to solve the Metaspace error
problem.
- Make sure you have Docker and SBT.
sbt it/test
By default, it/test
will do the following:
- Build a container image with the fat jar and a template.conf. The newly-built image will be registered with the local Docker daemon. This image is built with sbt-docker plugin.
- Run the test suites from
src/it/scala
, passing docker image ID viadocker.imageId
system property.
By default all logs are written to the STDOUT. If you want to write logs, for example,
to JSON files, you should define your own logging configuration and specify a path to it in conf/application.ini
:
-Dlogback.configurationFile=/path/to/your/logback.xml
Integration tests run in a forked JVM. To debug test suite code launched by SBT, you will need to add remote debug
options to javaOptions
in IntegrationTest
configuration:
javaOptions in IntegrationTest += "-agentlib:jdwp=transport=dt_socket,server=y,suspend=n,address=5005"
Debugging a node inside a container is a little more complicated: you will need to modify the WAVES_OPTS
environment
variable before starting a container.
You can run integration test suites from your preferred IDE. The only requirement is to have Docker image pre-built and
have docker.imageId
system property defined for the run configuration. The easiest way to build an image is to issue
sbt it/docker
command. You'll find the image ID in the SBT output:
...
[info] Step 5/5 : ENTRYPOINT /opt/waves/start-waves.sh
[info] ---> Using cache
[info] ---> e243fa08d496
[info] Successfully built e243fa08d496
[info] Tagging image e243fa08d496 with name: com.wavesplatform/root
[success] Total time: 4 s, completed Mar 22, 2017 12:36:34 PM
In this example, e243fa08d496
is the image ID you need. Make sure to re-build the image whenever the node code (not
the tests) is changed. If you run the tests from SBT, there's no need to manually rebuild the image, SBT will handle
this automatically.
Note: all required tools will be installed though Docker for simplicity.
-
Install Graphite, a service for collecting metrics.
-
Install Grafana for beautiful graphs.
-
By default all metrics are disabled. So specify Kamon settings through Java Properties and run the node with a desired config. For example, we ran Graphite locally and it accepts StatsD information on the
9999
port:SBT_OPTS="${SBT_OPTS} -Xms512M -Xmx1536M -Xss1M -XX:+CMSClassUnloadingEnabled \ -Dkamon.modules.kamon-statsd.auto-start=yes \ -Dkamon.modules.kamon-system-metrics.auto-start=yes \ -Dkamon.statsd.hostname=localhost \ -Dkamon.statsd.port=9999" sbt waves-testnet.conf
Here:
-Dkamon.modules.kamon-statsd.auto-start=yes
enables custom metrics;-Dkamon.modules.kamon-system-metrics.auto-start=yes
enables metrics of CPU, Memory and others;- See application.conf for more options.
We use YourKit full-featured Java Profiler to make Waves node faster. YourKit, LLC is the creator of innovative and intelligent tools for profiling Java and .NET applications.
Take a look at YourKit's leading software products:
YourKit Java Profiler and
YourKit .NET Profiler.