updates to readme

README.md

@@ -63,6 +63,9 @@ There are a number of non-intuitive kubernetes commands that are super helpful f
 
 - regenerate token join command: `kubeadm token create --print-join-command`
 
+- add pod cidr to remote nodes, sometimes this is required depending on which interface is used to add a node: `kubectl patch node <NODE_NAME> -p '{"spec":{"podCIDR":"<SUBNET>"}}'` from this really niche [github page][podcidr-help]
+`
+
 ## Links With More Information
 
 Debugging clusters can be somewhat difficult as everything is stuffed into pods, the links below have some steps to try:
@@ -71,6 +74,7 @@ Debugging clusters can be somewhat difficult as everything is stuffed into pods,
 - general steps for building images with buildx [here][docker-buildx]
 - there's some specific steps that got ARM64 builds working in my case, solution [here][docker-buildx-platform]
 
+[podcidr-help]: https://github.com/flannel-io/flannel/blob/master/Documentation/troubleshooting.md
 [crashloop]: https://managedkube.com/kubernetes/pod/failure/crashloopbackoff/k8sbot/troubleshooting/2019/02/12/pod-failure-crashloopbackoff.html
 [docker-buildx]: https://collabnix.com/building-arm-based-docker-images-on-docker-desktop-made-possible-using-buildx/
 [docker-buildx-platform]: https://github.com/docker/buildx/issues/464

clusterStartup.sh

@@ -3,10 +3,34 @@
 
 # script that starts up kubernetes cluster, and does some basic qol
 
-swapoff -a # turn off swap memory
-kubeadm init --config kubeadm-config.yml
+sudo swapoff -a # turn off swap memory
+sudo kubeadm init --config kubeadm-config.yaml
 
-mkdir -p ~/.kube
-sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
-sudo chown $(id -u):$(id -g) $HOME/.kube/config
-#tbd
+if [[ $? -ne 0 ]] ; then
+  # some error
+  echo "Error! Stopping cluster set up"
+  exit 1
+fi
+
+mkdir -p $HOME/.kube
+sudo cp -f /etc/kubernetes/admin.conf $HOME/.kube/config
+chown $(id -u):$(id -g) $HOME/.kube/config
+
+# export KUBECONFIG=/etc/kubernetes/admin.conf
+
+kubectl apply -f flannel.yml
+kubectl taint nodes --all node-role.kubernetes.io/master-
+kubectl label nodes $HOSTNAME name=master
+
+# print join command again to clipboard
+# kubeadm token create --print-join-command > ./tmpJoinCommand.txt
+# now add config command to file
+kubeadm token create --print-join-command | xclip -selection clipboard
+# save that file to clipboard
+# cat ./tmpJoinCommand.txt | xclip -selection clipboard
+# rm ./tmpJoinCommand.txt
+echo "set up done, join command copied (can ctrl-c now)"
+exit 0
+
+# dang pod cidr decided not to be assigned correctly, so follow links to assign it manually:
+# https://stackoverflow.com/questions/52633215/kubernetes-worker-nodes-not-automatically-being-assigned-podcidr-on-kubeadm-join

docker/base/Dockerfile

+ARG from=ros:noetic
+FROM ${from}
+
+RUN apt-get update && apt-get install -y python3-pip git ca-certificates
+RUN apt-get update && apt-get install -y libsdl2-dev libusb-1.0-0-dev build-essential cmake && apt-get install -y openssh-server
+
+RUN pip3 install git+https://github.com/catkin/catkin_tools.git
+
+# clone ros package repo
+ENV ROS_WS /opt/ros_ws
+RUN mkdir -p $ROS_WS/src
+WORKDIR $ROS_WS
+RUN git -C src clone https://git.caslab.ece.vt.edu/hlarkin3/base-station-ros/
+
+# install ros package dependencies
+RUN apt-get update && \
+    rosdep update && \
+    rosdep install -y --from-paths src/* --ignore-src && \
+    rm -rf /var/lib/apt/lists/*
+
+# build ros package source
+RUN catkin config \
+      --extend /opt/ros/$ROS_DISTRO && \
+    catkin build --cmake-args -DCMAKE_CXX_FLAGS="-std=c++17"
+    # \
+    #   roscpp_tutorials
+
+# might need something different in the bash script
+# COPY bashCheckRoscore.sh /usr/local/bin/bashCheckRoscore.sh
+COPY ros_entrypoint.sh /usr/local/bin/ros_entrypoint.sh
+
+RUN useradd -m -s /bin/bash -p $(openssl passwd -1 password) webssh
+RUN chmod 755 /usr/local/bin/ros_entrypoint.sh && \
+  /usr/local/bin/ros_entrypoint.sh
+
+# make ssh-ing easier
+RUN echo "source /opt/ros/noetic/setup.bash" >> /home/webssh/.bashrc
+RUN echo "source /opt/ros_ws/devel/setup.bash" >> /home/webssh/.bashrc
+
+
+ENTRYPOINT ["/usr/local/bin/ros_entrypoint.sh"]
+CMD ["bash"]
+# CMD ["service ssh start"]
+
+# run ros package launch file
+# CMD ["roslaunch", "roscpp_tutorials", "talker_listener.launch"]

docker/base/ros_entrypoint.sh

+#!/bin/bash
+set -e
+
+echo "==> Executing node image entrypoint ..."
+# setup ros environment
+source "/opt/ros/$ROS_DISTRO/setup.bash"
+# might need to source the devel space as well
+source "$ROS_WS/devel/setup.bash"
+
+service ssh start
+echo "==> Container ready"
+exec "$@"

docker/master/Dockerfile

 ARG from=ros:noetic
 FROM ${from}
 # install build tools
-# RUN apt-get update && apt-get install -y \
-#     python-catkin-tools \
-#     && rm -rf /var/lib/apt/lists/*
 RUN apt-get update && apt-get install -y python3-pip git
 # work around as the apt-get version fails
 RUN pip3 install git+https://github.com/catkin/catkin_tools.git

howtos/ClusterSetup.txt

@@ -62,7 +62,7 @@ sudo kubeadm reset
 
 reset ipv6 tables and such (seems to work fine without but eh):
 
-iptables -F && iptables -t nat -F && iptables -t mangle -F && iptables -X
+sudo iptables -F && sudo iptables -t nat -F && sudo iptables -t mangle -F && sudo iptables -X
 
 another ipvs reset:
 
@@ -108,3 +108,5 @@ crashloop backoff evaluation steps: https://managedkube.com/kubernetes/pod/failu
 docker buildx, building for multiple architectures: https://collabnix.com/building-arm-based-docker-images-on-docker-desktop-made-possible-using-buildx/
 
 buildx platform error thingy: https://github.com/docker/buildx/issues/464
+
+kubeadm join 192.168.1.230:6443 --token mv91uq.jxbm5k6yvog9tuh8 --discovery-token-ca-cert-hash sha256:2cb08cfa15611c1bdf6ab3a387145fa9cf092b59fb1c9134a51a625ab72325e3 --config kubeadm-config.yaml

kubernetes/base-deployment.yaml

+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: base-deployment
+  labels:
+    app: llhcluster
+    node: base
+spec:
+  replicas: 1
+  # The deployment handles all matching templated pods
+  selector:
+    matchLabels:
+      node: base
+  # Template for a replica.
+  # The deployment makes sure that a POD containing the containers
+  # defined below is always running.
+  template:
+    metadata:
+      labels:
+        node: base
+    spec:
+      containers:
+      - name: base
+        image: llh/basestation:v0
+        args:
+        - roscore
+        ports:
+        - containerPort: 11311
+          name: baseport
+      nodeSelector:
+        name: master
+
+---
+
+apiVersion: v1
+kind: Service
+metadata:
+  name: service-master
+  labels:
+    app: llhcluster
+    node: base
+spec:
+  clusterIP: None
+  ports:
+    - port: 11311
+  selector:
+    node: base