feat: release of infinitemonkeys helm chart (#2610)

* rerelease of infinitemonkeys helm chart, adding readme, minor refactor removing nested docker install from Dockerfile

* removing unneeded dependencies, updating vars and readme

* introducing jvm limits for monkey-server pods

* use JAVA_OPTS, Java 8 style flags

* add delay-pool argument to the controller

* decreasing waits between monkey startups, adding java-opts readme section, fixing indentation

* adding action for monkey image push to quay

* starting pods in parallel, adding ENV vars with expected app timeout settings to monkey-controller

* debugging new quay push action

* debugging new quay push action

* debugging new quay push action

* debugging new quay push action

* changes to monkey-dashboard

* disabling image build for PR branch

* updating readme on cluster resource use

---------

Co-authored-by: Mathias Staab <[email protected]>

.github/workflows/quay.yml

@@ -0,0 +1,37 @@
+name: quay.io infinitemonkeys image build & push
+
+on:
+  push:
+    branches: ['develop']
+    paths: ['monkeys/**']
+
+jobs:
+  quay_publish:
+    runs-on: ubuntu-latest
+    steps:
+      -
+        name: Checkout
+        uses: actions/checkout@v4
+      -
+        name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+      -
+        name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+      -
+        name: Login to DockerHub
+        uses: docker/login-action@v1
+        with:
+          registry: quay.io
+          username: wire+kalium_github_actions
+          password: ${{ secrets.QUAY_PASSWORD }}
+      -
+        name: Build and push
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          file: ./monkeys/docker/Dockerfile
+          platforms: linux/amd64
+          tags: quay.io/wire/infinitemonkeys
+          push: true
+

monkeys/docker/Dockerfile

@@ -6,15 +6,5 @@ RUN tar -xf /kalium/monkeys/build/distributions/monkeys.tar
 
 FROM wirebot/cryptobox:1.4.0
 
-# install docker cli
-RUN install -m 0755 -d /etc/apt/keyrings && \
-    curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc && \
-    chmod a+r /etc/apt/keyrings/docker.asc && \
-    echo \
-      "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu \
-      $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | \
-      tee /etc/apt/sources.list.d/docker.list > /dev/null
-RUN apt-get update && apt-get install -y docker-ce-cli
-
 RUN mkdir /opt/app
 COPY --from=jdk-build /monkeys /opt/app

monkeys/helm/infinitemonkeys/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

monkeys/helm/infinitemonkeys/Chart.yaml

@@ -0,0 +1,6 @@
+apiVersion: v2
+
+name: infinitemonkeys
+type: application
+version: 0.1.0
+

monkeys/helm/infinitemonkeys/Readme.md

@@ -0,0 +1,52 @@
+# infinitemonkeys - load testing tool for Wire
+
+This Helm chart provides:
+
+- A group of StatefulSets deploying one monkeyController and N monkeyServers
+- Integration Prometheus-Operator via PodMonitors and a Grafana Dashboard
+
+## Architecture
+
+The Controller will instruct the Servers what to do, via files/config.json.
+The config schema can be [viewed here](https://github.com/wireapp/kalium/blob/develop/monkeys/schema.json).
+Currently, config options for a single backend are exposed via values.yaml, but it is possible
+to add several backends and let the Servers send messages between them. In that case, it's recommended to edit
+the config.json file directly prior (re)deployment.
+
+
+## Scaling
+
+The application is capable scaling to 100s of monkeys, given sufficient cluster resources. We've determined
+that as of now (2024-03), one Controller can handle about 350 monkeys before running into concurrency issues.
+
+In order to support a deployment with 350 monkeys (plus Owner plus Controller), the cluster should have a total of at least:
+
+ * 192 GiB RAM
+ * 40 (v)CPU
+
+Equivalent to about 500 MiB RAM and per monkey and 10 monkeys per vCPU. During tests it's been determined that
+a minimum of six nodes with 32G RAM and 8 vCPU each are sufficient, with about 87% total cluster RAM usage.
+Depending on the specific actions which the monkeys are supposed to perform, more compute power might be needed down the line.
+
+
+## Log output
+
+The Servers echo all logging to stdout. The Controller writes to a local log within it's pod in `/wire/cryptobox/monkeys.log`.
+
+
+## JVM Optimizations
+
+JVM options are exposed as a config block in `values.yaml`, in order to reduce memory footprint of the monkey servers.
+At this point, the amount of RAM needed by each monkey is down about 50% compared to non-optimized defaults.
+
+
+## Limitations
+
+The total number of all users as defined via `monkeyController.userCount` (or directly in config.json backend sections) needs
+to be matched by `monkeyServer.replicaCount`, as N+1, since there's always one additional team owner monkey.
+
+
+## References
+
+1. - [Infinite Monkeys documentation](https://wearezeta.atlassian.net/wiki/spaces/ENGINEERIN/pages/878182477/Infinite+Monkeys+IM)
+2. - [JVM tuning in K8s](https://www.padok.fr/en/blog/jvm-oom)