Compare commits

..

1 Commits

Author SHA1 Message Date
Jérôme Petazzoni
ea3178327a Prepare content for Thoughtworks Infrastructure 2022-08-17 14:51:57 +02:00
51 changed files with 766 additions and 1673 deletions

View File

@@ -17,8 +17,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
---
@@ -30,8 +30,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-certs
namespace: kubernetes-dashboard
type: Opaque
@@ -43,8 +43,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-csrf
namespace: kubernetes-dashboard
type: Opaque
@@ -56,8 +56,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-key-holder
namespace: kubernetes-dashboard
type: Opaque
@@ -71,8 +71,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-settings
namespace: kubernetes-dashboard
---
@@ -84,8 +84,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
rules:
- apiGroups:
@@ -106,8 +106,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
roleRef:
apiGroup: rbac.authorization.k8s.io
@@ -126,8 +126,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
rules:
@@ -182,8 +182,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
roleRef:
@@ -204,8 +204,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
kubernetes.io/cluster-service: "true"
name: kubernetes-dashboard
namespace: kubernetes-dashboard
@@ -229,8 +229,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
spec:
@@ -253,8 +253,8 @@ spec:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
spec:
containers:
- args:
@@ -262,7 +262,7 @@ spec:
- --sidecar-host=http://127.0.0.1:8000
- --enable-skip-login
- --enable-insecure-login
image: kubernetesui/dashboard:v2.7.0
image: kubernetesui/dashboard:v2.5.0
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:
@@ -293,7 +293,7 @@ spec:
name: kubernetes-dashboard-certs
- mountPath: /tmp
name: tmp-volume
- image: kubernetesui/metrics-scraper:v1.0.8
- image: kubernetesui/metrics-scraper:v1.0.7
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:

View File

@@ -17,8 +17,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
---
@@ -30,8 +30,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-certs
namespace: kubernetes-dashboard
type: Opaque
@@ -43,8 +43,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-csrf
namespace: kubernetes-dashboard
type: Opaque
@@ -56,8 +56,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-key-holder
namespace: kubernetes-dashboard
type: Opaque
@@ -71,8 +71,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-settings
namespace: kubernetes-dashboard
---
@@ -84,8 +84,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
rules:
- apiGroups:
@@ -106,8 +106,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
roleRef:
apiGroup: rbac.authorization.k8s.io
@@ -126,8 +126,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
rules:
@@ -182,8 +182,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
roleRef:
@@ -204,8 +204,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
kubernetes.io/cluster-service: "true"
name: kubernetes-dashboard
namespace: kubernetes-dashboard
@@ -229,8 +229,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
spec:
@@ -253,15 +253,15 @@ spec:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
spec:
containers:
- args:
- --namespace=kubernetes-dashboard
- --auto-generate-certificates
- --sidecar-host=http://127.0.0.1:8000
image: kubernetesui/dashboard:v2.7.0
image: kubernetesui/dashboard:v2.5.0
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:
@@ -292,7 +292,7 @@ spec:
name: kubernetes-dashboard-certs
- mountPath: /tmp
name: tmp-volume
- image: kubernetesui/metrics-scraper:v1.0.8
- image: kubernetesui/metrics-scraper:v1.0.7
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:

View File

@@ -17,8 +17,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
---
@@ -30,8 +30,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-certs
namespace: kubernetes-dashboard
type: Opaque
@@ -43,8 +43,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-csrf
namespace: kubernetes-dashboard
type: Opaque
@@ -56,8 +56,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-key-holder
namespace: kubernetes-dashboard
type: Opaque
@@ -71,8 +71,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-settings
namespace: kubernetes-dashboard
---
@@ -84,8 +84,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
rules:
- apiGroups:
@@ -106,8 +106,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard-metrics
roleRef:
apiGroup: rbac.authorization.k8s.io
@@ -126,8 +126,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
rules:
@@ -182,8 +182,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
roleRef:
@@ -204,8 +204,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
kubernetes.io/cluster-service: "true"
name: kubernetes-dashboard
namespace: kubernetes-dashboard
@@ -229,8 +229,8 @@ metadata:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
name: kubernetes-dashboard
namespace: kubernetes-dashboard
spec:
@@ -253,15 +253,15 @@ spec:
app.kubernetes.io/instance: kubernetes-dashboard
app.kubernetes.io/managed-by: Helm
app.kubernetes.io/name: kubernetes-dashboard
app.kubernetes.io/version: 2.7.0
helm.sh/chart: kubernetes-dashboard-6.0.0
app.kubernetes.io/version: 2.5.0
helm.sh/chart: kubernetes-dashboard-5.2.0
spec:
containers:
- args:
- --namespace=kubernetes-dashboard
- --auto-generate-certificates
- --sidecar-host=http://127.0.0.1:8000
image: kubernetesui/dashboard:v2.7.0
image: kubernetesui/dashboard:v2.5.0
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:
@@ -292,7 +292,7 @@ spec:
name: kubernetes-dashboard-certs
- mountPath: /tmp
name: tmp-volume
- image: kubernetesui/metrics-scraper:v1.0.8
- image: kubernetesui/metrics-scraper:v1.0.7
imagePullPolicy: IfNotPresent
livenessProbe:
httpGet:
@@ -344,12 +344,3 @@ metadata:
creationTimestamp: null
name: cluster-admin
namespace: kubernetes-dashboard
---
apiVersion: v1
kind: Secret
type: kubernetes.io/service-account-token
metadata:
name: cluster-admin-token
namespace: kubernetes-dashboard
annotations:
kubernetes.io/service-account.name: cluster-admin

View File

@@ -15,10 +15,10 @@ spec:
- key: "{{ request.operation }}"
operator: Equals
value: UPDATE
- key: "{{ request.oldObject.metadata.labels.color || '' }}"
- key: "{{ request.oldObject.metadata.labels.color }}"
operator: NotEquals
value: ""
- key: "{{ request.object.metadata.labels.color || '' }}"
- key: "{{ request.object.metadata.labels.color }}"
operator: NotEquals
value: ""
validate:

View File

@@ -15,10 +15,10 @@ spec:
- key: "{{ request.operation }}"
operator: Equals
value: UPDATE
- key: "{{ request.oldObject.metadata.labels.color || '' }}"
- key: "{{ request.oldObject.metadata.labels.color }}"
operator: NotEquals
value: ""
- key: "{{ request.object.metadata.labels.color || '' }}"
- key: "{{ request.object.metadata.labels.color }}"
operator: Equals
value: ""
validate:

View File

@@ -70,15 +70,4 @@ add_namespace() {
kubectl create serviceaccount -n kubernetes-dashboard cluster-admin \
-o yaml --dry-run=client \
#
echo ---
cat <<EOF
apiVersion: v1
kind: Secret
type: kubernetes.io/service-account-token
metadata:
name: cluster-admin-token
namespace: kubernetes-dashboard
annotations:
kubernetes.io/service-account.name: cluster-admin
EOF
) > dashboard-with-token.yaml

View File

@@ -1,10 +1,11 @@
---
- hosts: nodes
become: yes
sudo: true
vars_files:
- vagrant.yml
tasks:
- name: clean up the home folder
file:
path: /home/vagrant/{{ item }}
@@ -23,23 +24,25 @@
- name: installing dependencies
apt:
name: apt-transport-https,ca-certificates,python3-pip,tmux
name: apt-transport-https,ca-certificates,python-pip,tmux
state: present
update_cache: true
- name: fetching docker repo key
apt_key:
url: https://download.docker.com/linux/ubuntu/gpg
state: present
keyserver: hkp://p80.pool.sks-keyservers.net:80
id: 58118E89F3A912897C070ADBF76221572C52609D
- name: adding docker repo
- name: adding package repos
apt_repository:
repo: deb https://download.docker.com/linux/ubuntu focal stable
repo: "{{ item }}"
state: present
with_items:
- deb https://apt.dockerproject.org/repo ubuntu-trusty main
- name: installing docker
apt:
name: docker-ce,docker-ce-cli,containerd.io,docker-compose-plugin
name: docker-engine
state: present
update_cache: true
@@ -53,7 +56,7 @@
lineinfile:
dest: /etc/default/docker
line: DOCKER_OPTS="--host=unix:///var/run/docker.sock --host=tcp://0.0.0.0:55555"
regexp: "^#?DOCKER_OPTS=.*$"
regexp: '^#?DOCKER_OPTS=.*$'
state: present
register: docker_opts
@@ -63,14 +66,22 @@
state: restarted
when: docker_opts is defined and docker_opts.changed
- name: install docker-compose from official github repo
get_url:
url: https://github.com/docker/compose/releases/download/1.29.2/docker-compose-Linux-x86_64
dest: /usr/local/bin/docker-compose
mode: "u+x,g+x"
- name: performing pip autoupgrade
pip:
name: pip
state: latest
- name: installing virtualenv
pip:
name: virtualenv
state: latest
- name: Install Docker Compose via PIP
pip: name=docker-compose
- name:
file: path="/usr/local/bin/docker-compose"
file:
path="/usr/local/bin/docker-compose"
state=file
mode=0755
owner=vagrant
@@ -117,3 +128,5 @@
line: "127.0.0.1 localhost {{ inventory_hostname }}"
- regexp: '^127\.0\.1\.1'
line: "127.0.1.1 {{ inventory_hostname }}"

View File

@@ -1,12 +1,13 @@
---
vagrant:
default_box: ubuntu/focal64
default_box: ubuntu/trusty64
default_box_check_update: true
ssh_insert_key: false
min_memory: 256
min_cores: 1
instances:
- hostname: node1
private_ip: 10.10.10.10
memory: 1512
@@ -36,3 +37,6 @@ instances:
private_ip: 10.10.10.50
memory: 512
cores: 1

View File

@@ -53,7 +53,7 @@ The value of the `location` variable is provider-specific. Examples:
| Provider | Example value | How to see possible values
|---------------|-------------------|---------------------------
| Digital Ocean | `ams3` | `doctl compute region list`
| Google Cloud | `europe-north1-a` | `gcloud compute zones list`
| Google Cloud | `europe-north1-a` | `gcloud compute zones list`
| Linode | `eu-central` | `linode-cli regions list`
| Oracle Cloud | `eu-stockholm-1` | `oci iam region list`
@@ -112,7 +112,7 @@ terraform init
See steps above, and add the following extra steps:
- Digital Ocean:
- Digital Coean:
```bash
export DIGITALOCEAN_ACCESS_TOKEN=$(grep ^access-token ~/.config/doctl/config.yaml | cut -d: -f2 | tr -d " ")
```

View File

@@ -3,14 +3,6 @@ set -e
TIME=$(which time)
if [ -f ~/.config/doctl/config.yaml ]; then
export DIGITALOCEAN_ACCESS_TOKEN=$(grep ^access-token ~/.config/doctl/config.yaml | cut -d: -f2 | tr -d " ")
fi
if [ -f ~/.config/linode-cli ]; then
export LINODE_TOKEN=$(grep ^token ~/.config/linode-cli | cut -d= -f2 | tr -d " ")
fi
PROVIDER=$1
[ "$PROVIDER" ] || {
echo "Please specify a provider as first argument, or 'ALL' for parallel mode."

View File

@@ -1,3 +1,3 @@
INFRACLASS=scaleway
#SCW_INSTANCE_TYPE=DEV1-L
SCW_ZONE=fr-par-2
#SCW_ZONE=fr-par-2

View File

@@ -131,8 +131,6 @@ set nowrap
SQRL
pssh -I "sudo -u $USER_LOGIN tee /home/$USER_LOGIN/.tmux.conf" <<SQRL
set -g status-style bg=yellow,bold
bind h select-pane -L
bind j select-pane -D
bind k select-pane -U
@@ -159,9 +157,6 @@ _cmd_clusterize() {
TAG=$1
need_tag
# Disable unattended upgrades so that they don't mess up with the subsequent steps
pssh sudo rm -f /etc/apt/apt.conf.d/50unattended-upgrades
# Special case for scaleway since it doesn't come with sudo
if [ "$INFRACLASS" = "scaleway" ]; then
pssh -l root "
@@ -187,23 +182,9 @@ _cmd_clusterize() {
pssh "
if [ -f /etc/iptables/rules.v4 ]; then
sudo sed -i 's/-A INPUT -j REJECT --reject-with icmp-host-prohibited//' /etc/iptables/rules.v4
sudo netfilter-persistent flush
sudo netfilter-persistent start
fi"
# oracle-cloud-agent upgrades pacakges in the background.
# This breaks our deployment scripts, because when we invoke apt-get, it complains
# that the lock already exists (symptom: random "Exited with error code 100").
# Workaround: if we detect oracle-cloud-agent, remove it.
# But this agent seems to also take care of installing/upgrading
# the unified-monitoring-agent package, so when we stop the snap,
# it can leave dpkg in a broken state. We "fix" it with the 2nd command.
pssh "
if [ -d /snap/oracle-cloud-agent ]; then
sudo snap remove oracle-cloud-agent
sudo dpkg --remove --force-remove-reinstreq unified-monitoring-agent
fi"
# Copy settings and install Python YAML parser
pssh -I tee /tmp/settings.yaml <tags/$TAG/settings.yaml
pssh "
@@ -281,14 +262,13 @@ EOF
"
##VERSION## https://github.com/docker/compose/releases
COMPOSE_VERSION=v2.11.1
COMPOSE_PLATFORM='linux-$(uname -m)'
# Just in case you need Compose 1.X, you can use the following lines.
# (But it will probably only work for x86_64 machines.)
#COMPOSE_VERSION=1.29.2
#COMPOSE_PLATFORM='Linux-$(uname -m)'
if [ "$ARCHITECTURE" ]; then
COMPOSE_VERSION=v2.2.3
COMPOSE_PLATFORM='linux-$(uname -m)'
else
COMPOSE_VERSION=1.29.2
COMPOSE_PLATFORM='Linux-$(uname -m)'
fi
pssh "
set -e
### Install docker-compose.
@@ -366,8 +346,7 @@ EOF"
pssh --timeout 200 "
sudo apt-get update -q &&
sudo apt-get install -qy kubelet kubeadm kubectl &&
sudo apt-mark hold kubelet kubeadm kubectl &&
kubeadm completion bash | sudo tee /etc/bash_completion.d/kubeadm &&
sudo apt-mark hold kubelet kubeadm kubectl
kubectl completion bash | sudo tee /etc/bash_completion.d/kubectl &&
echo 'alias k=kubectl' | sudo tee /etc/bash_completion.d/k &&
echo 'complete -F __start_kubectl k' | sudo tee -a /etc/bash_completion.d/k"
@@ -440,9 +419,8 @@ EOF
# Install weave as the pod network
pssh "
if i_am_first_node; then
#kubever=\$(kubectl version | base64 | tr -d '\n') &&
#kubectl apply -f https://cloud.weave.works/k8s/net?k8s-version=\$kubever
kubectl apply -f https://github.com/weaveworks/weave/releases/download/v2.8.1/weave-daemonset-k8s-1.11.yaml
kubever=\$(kubectl version | base64 | tr -d '\n') &&
kubectl apply -f https://cloud.weave.works/k8s/net?k8s-version=\$kubever
fi"
# Join the other nodes to the cluster
@@ -500,13 +478,12 @@ _cmd_kubetools() {
# Install kube-ps1
pssh "
set -e
if ! [ -d /opt/kube-ps1 ]; then
if ! [ -f /etc/profile.d/kube-ps1.sh ]; then
cd /tmp
git clone https://github.com/jonmosco/kube-ps1
sudo mv kube-ps1 /opt/kube-ps1
sudo cp kube-ps1/kube-ps1.sh /etc/profile.d/kube-ps1.sh
sudo -u $USER_LOGIN sed -i s/docker-prompt/kube_ps1/ /home/$USER_LOGIN/.bashrc &&
sudo -u $USER_LOGIN tee -a /home/$USER_LOGIN/.bashrc <<EOF
. /opt/kube-ps1/kube-ps1.sh
KUBE_PS1_PREFIX=""
KUBE_PS1_SUFFIX=""
KUBE_PS1_SYMBOL_ENABLE="false"
@@ -517,13 +494,13 @@ EOF
# Install stern
##VERSION## https://github.com/stern/stern/releases
STERN_VERSION=1.22.0
STERN_VERSION=1.20.1
FILENAME=stern_${STERN_VERSION}_linux_${ARCH}
URL=https://github.com/stern/stern/releases/download/v$STERN_VERSION/$FILENAME.tar.gz
pssh "
if [ ! -x /usr/local/bin/stern ]; then
curl -fsSL $URL |
sudo tar -C /usr/local/bin -zx stern
sudo tar -C /usr/local/bin -zx --strip-components=1 $FILENAME/stern
sudo chmod +x /usr/local/bin/stern
stern --completion bash | sudo tee /etc/bash_completion.d/stern
stern --version
@@ -539,7 +516,7 @@ EOF
# Install kustomize
##VERSION## https://github.com/kubernetes-sigs/kustomize/releases
KUSTOMIZE_VERSION=v4.5.7
KUSTOMIZE_VERSION=v4.4.0
URL=https://github.com/kubernetes-sigs/kustomize/releases/download/kustomize/${KUSTOMIZE_VERSION}/kustomize_${KUSTOMIZE_VERSION}_linux_${ARCH}.tar.gz
pssh "
if [ ! -x /usr/local/bin/kustomize ]; then
@@ -558,7 +535,7 @@ EOF
if [ ! -x /usr/local/bin/ship ]; then
##VERSION##
curl -fsSL https://github.com/replicatedhq/ship/releases/download/v0.51.3/ship_0.51.3_linux_$ARCH.tar.gz |
sudo tar -C /usr/local/bin -zx ship
sudo tar -C /usr/local/bin -zx ship
fi"
# Install the AWS IAM authenticator
@@ -566,8 +543,8 @@ EOF
if [ ! -x /usr/local/bin/aws-iam-authenticator ]; then
##VERSION##
sudo curl -fsSLo /usr/local/bin/aws-iam-authenticator https://amazon-eks.s3-us-west-2.amazonaws.com/1.12.7/2019-03-27/bin/linux/$ARCH/aws-iam-authenticator
sudo chmod +x /usr/local/bin/aws-iam-authenticator
aws-iam-authenticator version
sudo chmod +x /usr/local/bin/aws-iam-authenticator
aws-iam-authenticator version
fi"
# Install the krew package manager
@@ -609,7 +586,6 @@ EOF
FILENAME=tilt.\$TILT_VERSION.linux.$TILT_ARCH.tar.gz
curl -fsSL https://github.com/tilt-dev/tilt/releases/download/v\$TILT_VERSION/\$FILENAME |
sudo tar -zxvf- -C /usr/local/bin tilt
tilt completion bash | sudo tee /etc/bash_completion.d/tilt
tilt version
fi"
@@ -618,7 +594,6 @@ EOF
if [ ! -x /usr/local/bin/skaffold ]; then
curl -fsSLo skaffold https://storage.googleapis.com/skaffold/releases/latest/skaffold-linux-$ARCH &&
sudo install skaffold /usr/local/bin/
skaffold completion bash | sudo tee /etc/bash_completion.d/skaffold
skaffold version
fi"
@@ -627,28 +602,9 @@ EOF
if [ ! -x /usr/local/bin/kompose ]; then
curl -fsSLo kompose https://github.com/kubernetes/kompose/releases/latest/download/kompose-linux-$ARCH &&
sudo install kompose /usr/local/bin
kompose completion bash | sudo tee /etc/bash_completion.d/kompose
kompose version
fi"
# Install KinD
pssh "
if [ ! -x /usr/local/bin/kind ]; then
curl -fsSLo kind https://github.com/kubernetes-sigs/kind/releases/latest/download/kind-linux-$ARCH &&
sudo install kind /usr/local/bin
kind completion bash | sudo tee /etc/bash_completion.d/kind
kind version
fi"
# Install YTT
pssh "
if [ ! -x /usr/local/bin/ytt ]; then
curl -fsSLo ytt https://github.com/vmware-tanzu/carvel-ytt/releases/latest/download/ytt-linux-$ARCH &&
sudo install ytt /usr/local/bin
ytt completion bash | sudo tee /etc/bash_completion.d/ytt
ytt version
fi"
##VERSION## https://github.com/bitnami-labs/sealed-secrets/releases
KUBESEAL_VERSION=0.17.4
#case $ARCH in

View File

@@ -36,7 +36,7 @@ if os.path.isfile(domain_or_domain_file):
clusters = [line.split() for line in lines]
else:
ips = open(f"tags/{ips_file_or_tag}/ips.txt").read().split()
settings_file = f"tags/{ips_file_or_tag}/settings.yaml"
settings_file = f"tags/{tag}/settings.yaml"
clustersize = yaml.safe_load(open(settings_file))["clustersize"]
clusters = []
while ips:

View File

@@ -17,17 +17,8 @@
exit 1
}
NETLIFY_CONFIG_FILE=~/.config/netlify/config.json
if ! [ -f "$NETLIFY_CONFIG_FILE" ]; then
echo "Could not find Netlify configuration file ($NETLIFY_CONFIG_FILE)."
echo "Try to run the following command, and try again:"
echo "npx netlify-cli login"
exit 1
fi
NETLIFY_USERID=$(jq .userId < "$NETLIFY_CONFIG_FILE")
NETLIFY_TOKEN=$(jq -r .users[$NETLIFY_USERID].auth.token < "$NETLIFY_CONFIG_FILE")
NETLIFY_USERID=$(jq .userId < ~/.config/netlify/config.json)
NETLIFY_TOKEN=$(jq -r .users[$NETLIFY_USERID].auth.token < ~/.config/netlify/config.json)
netlify() {
URI=$1

View File

@@ -1,71 +0,0 @@
resource "azurerm_resource_group" "_" {
name = var.prefix
location = var.location
}
resource "azurerm_public_ip" "_" {
count = var.how_many_nodes
name = format("%s-%04d", var.prefix, count.index + 1)
location = azurerm_resource_group._.location
resource_group_name = azurerm_resource_group._.name
allocation_method = "Dynamic"
}
resource "azurerm_network_interface" "_" {
count = var.how_many_nodes
name = format("%s-%04d", var.prefix, count.index + 1)
location = azurerm_resource_group._.location
resource_group_name = azurerm_resource_group._.name
ip_configuration {
name = "internal"
subnet_id = azurerm_subnet._.id
private_ip_address_allocation = "Dynamic"
public_ip_address_id = azurerm_public_ip._[count.index].id
}
}
resource "azurerm_linux_virtual_machine" "_" {
count = var.how_many_nodes
name = format("%s-%04d", var.prefix, count.index + 1)
resource_group_name = azurerm_resource_group._.name
location = azurerm_resource_group._.location
size = var.size
admin_username = "ubuntu"
network_interface_ids = [
azurerm_network_interface._[count.index].id,
]
admin_ssh_key {
username = "ubuntu"
public_key = local.authorized_keys
}
os_disk {
caching = "ReadWrite"
storage_account_type = "Standard_LRS"
}
source_image_reference {
publisher = "Canonical"
offer = "UbuntuServer"
sku = "18.04-LTS" # FIXME
version = "latest"
}
}
# The public IP address only gets allocated when the address actually gets
# attached to the virtual machine. So we need to do this extra indrection
# to retrieve the IP addresses. Otherwise the IP addresses show up as blank.
# See: https://github.com/hashicorp/terraform-provider-azurerm/issues/310#issuecomment-335479735
data "azurerm_public_ip" "_" {
count = var.how_many_nodes
name = format("%s-%04d", var.prefix, count.index + 1)
resource_group_name = azurerm_resource_group._.name
depends_on = [azurerm_linux_virtual_machine._]
}
output "ip_addresses" {
value = join("", formatlist("%s\n", data.azurerm_public_ip._.*.ip_address))
}

View File

@@ -1,13 +0,0 @@
resource "azurerm_virtual_network" "_" {
name = "tf-vnet"
address_space = ["10.10.0.0/16"]
location = azurerm_resource_group._.location
resource_group_name = azurerm_resource_group._.name
}
resource "azurerm_subnet" "_" {
name = "tf-subnet"
resource_group_name = azurerm_resource_group._.name
virtual_network_name = azurerm_virtual_network._.name
address_prefixes = ["10.10.0.0/20"]
}

View File

@@ -1,13 +0,0 @@
terraform {
required_version = ">= 1"
required_providers {
azurerm = {
source = "hashicorp/azurerm"
version = "=3.33.0"
}
}
}
provider "azurerm" {
features {}
}

View File

@@ -1,32 +0,0 @@
variable "prefix" {
type = string
default = "provisioned-with-terraform"
}
variable "how_many_nodes" {
type = number
default = 2
}
locals {
authorized_keys = file("~/.ssh/id_rsa.pub")
}
/*
Available sizes:
"Standard_D11_v2" # CPU=2 RAM=14
"Standard_F4s_v2" # CPU=4 RAM=8
"Standard_D1_v2" # CPU=1 RAM=3.5
"Standard_B1ms" # CPU=1 RAM=2
"Standard_B2s" # CPU=2 RAM=4
*/
variable "size" {
type = string
default = "Standard_F4s_v2"
}
variable "location" {
type = string
default = "South Africa North"
}

View File

@@ -1,8 +1,7 @@
# Uncomment and/or edit one of the the following lines if necessary.
#/ /kube-halfday.yml.html 200!
#/ /kube-fullday.yml.html 200!
#/ /kube-twodays.yml.html 200!
/ /live.html 200!
/ /kube.yml.html 200!
# And this allows to do "git clone https://container.training".
/info/refs service=git-upload-pack https://github.com/jpetazzo/container.training/info/refs?service=git-upload-pack

View File

@@ -194,9 +194,9 @@
}
},
"node_modules/engine.io": {
"version": "6.2.1",
"resolved": "https://registry.npmjs.org/engine.io/-/engine.io-6.2.1.tgz",
"integrity": "sha512-ECceEFcAaNRybd3lsGQKas3ZlMVjN3cyWwMP25D2i0zWfyiytVbTpRPa34qrr+FHddtpBVOmq4H/DCv1O0lZRA==",
"version": "6.2.0",
"resolved": "https://registry.npmjs.org/engine.io/-/engine.io-6.2.0.tgz",
"integrity": "sha512-4KzwW3F3bk+KlzSOY57fj/Jx6LyRQ1nbcyIadehl+AnXjKT7gDO0ORdRi/84ixvMKTym6ZKuxvbzN62HDDU1Lg==",
"dependencies": {
"@types/cookie": "^0.4.1",
"@types/cors": "^2.8.12",
@@ -742,9 +742,9 @@
"integrity": "sha512-sGkPx+VjMtmA6MX27oA4FBFELFCZZ4S4XqeGOXCv68tT+jb3vk/RyaKWP0PTKyWtmLSM0b+adUTEvbs1PEaH2w=="
},
"node_modules/socket.io-client/node_modules/socket.io-parser": {
"version": "4.2.1",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.2.1.tgz",
"integrity": "sha512-V4GrkLy+HeF1F/en3SpUaM+7XxYXpuMUWLGde1kSSh5nQMN4hLrbPIkD+otwh6q9R6NOQBN4AMaOZ2zVjui82g==",
"version": "4.2.0",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.2.0.tgz",
"integrity": "sha512-tLfmEwcEwnlQTxFB7jibL/q2+q8dlVQzj4JdRLJ/W/G1+Fu9VSxCx1Lo+n1HvXxKnM//dUuD0xgiA7tQf57Vng==",
"dependencies": {
"@socket.io/component-emitter": "~3.1.0",
"debug": "~4.3.1"
@@ -754,9 +754,9 @@
}
},
"node_modules/socket.io-parser": {
"version": "4.0.5",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.0.5.tgz",
"integrity": "sha512-sNjbT9dX63nqUFIOv95tTVm6elyIU4RvB1m8dOeZt+IgWwcWklFDOdmGcfo3zSiRsnR/3pJkjY5lfoGqEe4Eig==",
"version": "4.0.4",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.0.4.tgz",
"integrity": "sha512-t+b0SS+IxG7Rxzda2EVvyBZbvFPBCjJoyHuE0P//7OAsN23GItzDRdWa6ALxZI/8R5ygK7jAR6t028/z+7295g==",
"dependencies": {
"@types/component-emitter": "^1.2.10",
"component-emitter": "~1.3.0",
@@ -1033,9 +1033,9 @@
"integrity": "sha512-TPJXq8JqFaVYm2CWmPvnP2Iyo4ZSM7/QKcSmuMLDObfpH5fi7RUGmd/rTDf+rut/saiDiQEeVTNgAmJEdAOx0w=="
},
"engine.io": {
"version": "6.2.1",
"resolved": "https://registry.npmjs.org/engine.io/-/engine.io-6.2.1.tgz",
"integrity": "sha512-ECceEFcAaNRybd3lsGQKas3ZlMVjN3cyWwMP25D2i0zWfyiytVbTpRPa34qrr+FHddtpBVOmq4H/DCv1O0lZRA==",
"version": "6.2.0",
"resolved": "https://registry.npmjs.org/engine.io/-/engine.io-6.2.0.tgz",
"integrity": "sha512-4KzwW3F3bk+KlzSOY57fj/Jx6LyRQ1nbcyIadehl+AnXjKT7gDO0ORdRi/84ixvMKTym6ZKuxvbzN62HDDU1Lg==",
"requires": {
"@types/cookie": "^0.4.1",
"@types/cors": "^2.8.12",
@@ -1456,9 +1456,9 @@
"integrity": "sha512-sGkPx+VjMtmA6MX27oA4FBFELFCZZ4S4XqeGOXCv68tT+jb3vk/RyaKWP0PTKyWtmLSM0b+adUTEvbs1PEaH2w=="
},
"socket.io-parser": {
"version": "4.2.1",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.2.1.tgz",
"integrity": "sha512-V4GrkLy+HeF1F/en3SpUaM+7XxYXpuMUWLGde1kSSh5nQMN4hLrbPIkD+otwh6q9R6NOQBN4AMaOZ2zVjui82g==",
"version": "4.2.0",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.2.0.tgz",
"integrity": "sha512-tLfmEwcEwnlQTxFB7jibL/q2+q8dlVQzj4JdRLJ/W/G1+Fu9VSxCx1Lo+n1HvXxKnM//dUuD0xgiA7tQf57Vng==",
"requires": {
"@socket.io/component-emitter": "~3.1.0",
"debug": "~4.3.1"
@@ -1467,9 +1467,9 @@
}
},
"socket.io-parser": {
"version": "4.0.5",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.0.5.tgz",
"integrity": "sha512-sNjbT9dX63nqUFIOv95tTVm6elyIU4RvB1m8dOeZt+IgWwcWklFDOdmGcfo3zSiRsnR/3pJkjY5lfoGqEe4Eig==",
"version": "4.0.4",
"resolved": "https://registry.npmjs.org/socket.io-parser/-/socket.io-parser-4.0.4.tgz",
"integrity": "sha512-t+b0SS+IxG7Rxzda2EVvyBZbvFPBCjJoyHuE0P//7OAsN23GItzDRdWa6ALxZI/8R5ygK7jAR6t028/z+7295g==",
"requires": {
"@types/component-emitter": "^1.2.10",
"component-emitter": "~1.3.0",

View File

@@ -19,7 +19,7 @@ They abstract the connection details for this services, and can help with:
* fail over (how do I know to which instance of a replicated service I should connect?)
* load balancing (how do I spread my requests across multiple instances of a service?)
* load balancing (how to I spread my requests across multiple instances of a service?)
* authentication (what if my service requires credentials, certificates, or otherwise?)

View File

@@ -1,68 +0,0 @@
title: |
Docker Training
chat: "[Mattermost](https://live.container.training/mattermost)"
gitrepo: github.com/jpetazzo/container.training
slides: https://2022-11-live.container.training/
#slidenumberprefix: "#SomeHashTag &mdash; "
exclude:
- self-paced
content:
- shared/title.md
- logistics.md
- containers/intro.md
- shared/about-slides.md
- shared/chat-room-im.md
#- shared/chat-room-zoom-meeting.md
#- shared/chat-room-zoom-webinar.md
- shared/toc.md
- # DAY 1
#- containers/Docker_Overview.md
#- containers/Docker_History.md
- containers/Training_Environment.md
#- containers/Installing_Docker.md
- containers/First_Containers.md
- containers/Background_Containers.md
- containers/Initial_Images.md
- containers/Building_Images_Interactively.md
- containers/Building_Images_With_Dockerfiles.md
- containers/Cmd_And_Entrypoint.md
- containers/Copying_Files_During_Build.md
- containers/Exercise_Dockerfile_Basic.md
- # DAY 2
- containers/Container_Networking_Basics.md
- containers/Local_Development_Workflow.md
- containers/Container_Network_Model.md
- containers/Compose_For_Dev_Stacks.md
- containers/Exercise_Composefile.md
- # DAY 3
- containers/Start_And_Attach.md
- containers/Naming_And_Inspecting.md
- containers/Labels.md
- containers/Getting_Inside.md
- containers/Dockerfile_Tips.md
- containers/Advanced_Dockerfiles.md
- containers/Multi_Stage_Builds.md
- containers/Publishing_To_Docker_Hub.md
- containers/Exercise_Dockerfile_Advanced.md
- # DAY 4
- containers/Buildkit.md
- containers/Network_Drivers.md
- containers/Namespaces_Cgroups.md
#- containers/Copy_On_Write.md
- containers/Orchestration_Overview.md
#- containers/Docker_Machine.md
#- containers/Init_Systems.md
#- containers/Application_Configuration.md
#- containers/Logging.md
#- containers/Containers_From_Scratch.md
#- containers/Container_Engines.md
#- containers/Pods_Anatomy.md
#- containers/Ecosystem.md
- shared/thankyou.md
#- containers/links.md

View File

@@ -4,6 +4,6 @@
(we will use the `rng` service in the dockercoins app)
- See what happens when the load increases
- See what happens when the load increses
(spoiler alert: it involves timeouts!)

View File

@@ -2,7 +2,7 @@
- Add an ingress controller to a Kubernetes cluster
- Create an ingress resource for a couple of web apps on that cluster
- Create an ingress resource for a web app on that cluster
- Challenge: accessing/exposing port 80

View File

@@ -1,131 +1,49 @@
# Exercise — Ingress
- We want to expose a couple of web apps through an ingress controller
- We want to expose a web app through an ingress controller
- This will require:
- the web apps (e.g. two instances of `jpetazzo/color`)
- the web app itself (dockercoins, NGINX, whatever we want)
- an ingress controller
- a domain name (`use \*.nip.io` or `\*.localdev.me`)
- an ingress resource
---
## Different scenarios
## Goal
We will use a different deployment mechanism depending on the cluster that we have:
- We want to be able to access the web app using a URL like:
- Managed cluster with working `LoadBalancer` Services
http://webapp.localdev.me
- Local development cluster
*or*
- Cluster without `LoadBalancer` Services (e.g. deployed with `kubeadm`)
http://webapp.A.B.C.D.nip.io
---
## The apps
- The web apps will be deployed similarly, regardless of the scenario
- Let's start by deploying two web apps, e.g.:
a Deployment called `blue` and another called `green`, using image `jpetazzo/color`
- Expose them with two `ClusterIP` Services
---
## Scenario "classic cloud Kubernetes"
*Difficulty: easy*
For this scenario, we need a cluster with working `LoadBalancer` Services.
(For instance, a managed Kubernetes cluster from a cloud provider.)
We suggest to use "Ingress NGINX" with its default settings.
It can be installed with `kubectl apply` or with `helm`.
Both methods are described in [the documentation][ingress-nginx-deploy].
We want our apps to be available on e.g. http://X.X.X.X/blue and http://X.X.X.X/green
<br/>
(where X.X.X.X is the IP address of the `LoadBalancer` allocated by Ingress NGINX).
[ingress-nginx-deploy]: https://kubernetes.github.io/ingress-nginx/deploy/
---
## Scenario "local development cluster"
*Difficulty: easy-hard (depends on the type of cluster!)*
For this scenario, we want to use a local cluster like KinD, minikube, etc.
We suggest to use "Ingress NGINX" again, like for the previous scenario.
Furthermore, we want to use `localdev.me`.
We want our apps to be available on e.g. `blue.localdev.me` and `green.localdev.me`.
The difficulty is to ensure that `localhost:80` will map to the ingress controller.
(See next slide for hints!)
(where A.B.C.D is the IP address of one of our nodes)
---
## Hints
- With clusters like Docker Desktop, the first `LoadBalancer` service uses `localhost`
- For the ingress controller, we can use:
(if the ingress controller is the first `LoadBalancer` service, we're all set!)
- [ingress-nginx](https://github.com/kubernetes/ingress-nginx/blob/main/docs/deploy/index.md)
- With clusters like K3D and KinD, it is possible to define extra port mappings
- the [Traefik Helm chart](https://doc.traefik.io/traefik/getting-started/install-traefik/#use-the-helm-chart)
(and map e.g. `localhost:80` to port 30080 on the node; then use that as a `NodePort`)
- the container.training [Traefik DaemonSet](https://raw.githubusercontent.com/jpetazzo/container.training/main/k8s/traefik-v2.yaml)
---
- If our cluster supports LoadBalancer Services: easy
## Scenario "on premises cluster", take 1
(nothing special to do)
*Difficulty: easy*
- For local clusters, things can be more difficult; two options:
For this scenario, we need a cluster with nodes that are publicly accessible.
- map localhost:80 to e.g. a NodePort service, and use `\*.localdev.me`
We want to deploy the ingress controller so that it listens on port 80 on all nodes.
This can be done e.g. with the manifests in @@LINK[k8s/traefik.yaml].
We want our apps to be available on e.g. http://X.X.X.X/blue and http://X.X.X.X/green
<br/>
(where X.X.X.X is the IP address of any of our nodes).
---
## Scenario "on premises cluster", take 2
*Difficulty: medium*
We want to deploy the ingress controller so that it listens on port 80 on all nodes.
But this time, we want to use a Helm chart to install the ingress controller.
We can use either the Ingress NGINX Helm chart, or the Traefik Helm chart.
Test with an untainted node first.
Feel free to make it work on tainted nodes (e.g. control plane nodes) later.
---
## Scenario "on premises cluster", take 3
*Difficulty: hard*
This is similar to the previous scenario, but with two significant changes:
1. We only want to run the ingress controller on nodes that have the role `ingress`.
2. We don't want to use `hostNetwork`, but a list of `externalIPs` instead.
- use hostNetwork, or ExternalIP, and use `\*.nip.io`

View File

@@ -13,4 +13,3 @@ https://gallant-turing-d0d520.netlify.com/containers/train-of-containers-1.jpg
https://gallant-turing-d0d520.netlify.com/containers/train-of-containers-2.jpg
https://gallant-turing-d0d520.netlify.com/containers/two-containers-on-a-truck.jpg
https://gallant-turing-d0d520.netlify.com/containers/wall-of-containers.jpeg
https://gallant-turing-d0d520.netlify.com/containers/catene-de-conteneurs.jpg

View File

@@ -274,7 +274,7 @@ class: extra-details
- ...or with a Secret with the right [type and annotation][create-token]
[create-token]: https://kubernetes.io/docs/reference/access-authn-authz/service-accounts-admin/#create-token
[create-token]: https://kubernetes.io/docs/reference/access-authn-authz/service-accounts-admin/#to-create-additional-api-tokens
---

View File

@@ -202,9 +202,7 @@ class: extra-details
- These are JWS signatures using HMAC-SHA256
(see [the reference documentation][configmap-signing] for more details)
[configmap-signing]: https://kubernetes.io/docs/reference/access-authn-authz/bootstrap-tokens/#configmap-signing
(see [here](https://kubernetes.io/docs/reference/access-authn-authz/bootstrap-tokens/#configmap-signing) for more details)
---

View File

@@ -48,7 +48,7 @@
- We must run nodes on a supported infrastructure
- Check the [GitHub repo][autoscaler-providers] for a non-exhaustive list of supported providers
- See [here] for a non-exhaustive list of supported providers
- Sometimes, the cluster autoscaler is installed automatically
@@ -58,7 +58,7 @@
(which is often non-trivial and highly provider-specific)
[autoscaler-providers]: https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler/cloudprovider
[here]: https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler/cloudprovider
---

View File

@@ -138,7 +138,7 @@ class: extra-details
- The Cluster Autoscaler only supports a few cloud infrastructures
(see the [kubernetes/autoscaler repo][kubernetes-autoscaler-repo] for a list)
(see [here](https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler/cloudprovider) for a list)
- The Cluster Autoscaler cannot scale down nodes that have pods using:
@@ -148,8 +148,6 @@ class: extra-details
- a restrictive PodDisruptionBudget
[kubernetes-autoscaler-repo]: https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler/cloudprovider
---
## Other way to do capacity planning

View File

@@ -24,11 +24,11 @@
- Interface parameters (MTU, sysctls) could be tweaked by the `tuning` plugin
The reference plugins are available [here][cni-reference-plugins].
The reference plugins are available [here].
Look in each plugin's directory for its documentation.
[cni-reference-plugins]: https://github.com/containernetworking/plugins/tree/master/plugins
[here]: https://github.com/containernetworking/plugins/tree/master/plugins
---

View File

@@ -287,9 +287,7 @@ No!
--
- The Docker Engine used to be the default option to run containers with Kubernetes
- Support for Docker (specifically: dockershim) was removed in Kubernetes 1.24
- By default, Kubernetes uses the Docker Engine to run containers
- We can leverage other pluggable runtimes through the *Container Runtime Interface*
@@ -331,26 +329,32 @@ Yes!
- We can do these things without Docker
<br/>
(but with some languages/frameworks, it might be much harder)
(and get diagnosed with NIH¹ syndrome)
- Docker is still the most stable container engine today
<br/>
(but other options are maturing very quickly)
.footnote[¹[Not Invented Here](https://en.wikipedia.org/wiki/Not_invented_here)]
---
class: extra-details
## Do we need to run Docker at all?
- On our Kubernetes clusters:
*Not anymore*
- On our development environments, CI pipelines ... :
*Yes, almost certainly*
- On our production servers:
*Yes (today)*
*Probably not (in the future)*
.footnote[More information about CRI [on the Kubernetes blog](https://kubernetes.io/blog/2016/12/container-runtime-interface-cri-in-kubernetes)]
---
## Interacting with Kubernetes

View File

@@ -1,62 +1,46 @@
# Healthchecks
- Containers can have *healthchecks* (also called "probes")
- Containers can have *healthchecks*
- There are three kinds of healthchecks, corresponding to different use-cases:
- There are three kinds of healthchecks, corresponding to very different use-cases:
`startupProbe`, `readinessProbe`, `livenessProbe`
- liveness = detect when a container is "dead" and needs to be restarted
- readiness = detect when a container is ready to serve traffic
- startup = detect if a container has finished to boot
- These healthchecks are optional (we can use none, all, or some of them)
- Different probes are available:
- Different probes are available (HTTP request, TCP connection, program execution)
HTTP GET, TCP connection, arbitrary program execution, GRPC
- All these probes have a binary result (success/failure)
- Probes that aren't defined will default to a "success" result
- Let's see the difference and how to use them!
---
## Use-cases in brief
*My container takes a long time to boot before being able to serve traffic.*
→ use a `startupProbe` (but often a `readinessProbe` can also do the job)
*Sometimes, my container is unavailable or overloaded, and needs to e.g. be taken temporarily out of load balancer rotation.*
→ use a `readinessProbe`
*Sometimes, my container enters a broken state which can only be fixed by a restart.*
→ use a `livenessProbe`
---
## Liveness probes
## Liveness probe
*This container is dead, we don't know how to fix it, other than restarting it.*
- Check if the container is dead or alive
- Indicates if the container is dead or alive
- If Kubernetes determines that the container is dead:
- A dead container cannot come back to life
- it terminates the container gracefully
- If the liveness probe fails, the container is killed (destroyed)
- it restarts the container (unless the Pod's `restartPolicy` is `Never`)
(to make really sure that it's really dead; no zombies or undeads!)
- With the default parameters, it takes:
- What happens next depends on the pod's `restartPolicy`:
- up to 30 seconds to determine that the container is dead
- `Never`: the container is not restarted
- up to 30 seconds to terminate it
- `OnFailure` or `Always`: the container is restarted
---
## When to use a liveness probe
- To detect failures that can't be recovered
- To indicate failures that can't be recovered
- deadlocks (causing all requests to time out)
@@ -64,45 +48,47 @@
- Anything where our incident response would be "just restart/reboot it"
---
## Liveness probes gotchas
.warning[**Do not** use liveness probes for problems that can't be fixed by a restart]
- Otherwise we just restart our pods for no reason, creating useless load
.warning[**Do not** depend on other services within a liveness probe]
---
- Otherwise we can experience cascading failures
## Readiness probe (1)
(example: web server liveness probe that makes a requests to a database)
*Make sure that a container is ready before continuing a rolling update.*
.warning[**Make sure** that liveness probes respond quickly]
- Indicates if the container is ready to handle traffic
- The default probe timeout is 1 second (this can be tuned!)
- When doing a rolling update, the Deployment controller waits for Pods to be ready
- If the probe takes longer than that, it will eventually cause a restart
(a Pod is ready when all the containers in the Pod are ready)
- Improves reliability and safety of rolling updates:
- don't roll out a broken version (that doesn't pass readiness checks)
- don't lose processing capacity during a rolling update
---
## Readiness probes
## Readiness probe (2)
*Sometimes, my container "needs a break".*
*Temporarily remove a container (overloaded or otherwise) from a Service load balancer.*
- Check if the container is ready or not
- A container can mark itself "not ready" temporarily
- If the container is not ready, its Pod is not ready
(e.g. if it's overloaded or needs to reload/restart/garbage collect...)
- If the Pod belongs to a Service, it is removed from its Endpoints
- If a container becomes "unready" it might be ready again soon
(it stops receiving new connections but existing ones are not affected)
- If the readiness probe fails:
- If there is a rolling update in progress, it might pause
- the container is *not* killed
(Kubernetes will try to respect the MaxUnavailable parameter)
- if the pod is a member of a service, it is temporarily removed
- As soon as the readiness probe suceeds again, everything goes back to normal
- it is re-added as soon as the readiness probe passes again
---
@@ -116,31 +102,67 @@
- To indicate temporary failure or unavailability
- runtime is busy doing garbage collection or (re)loading data
- application can only service *N* parallel connections
- new connections will be directed to other Pods
- runtime is busy doing garbage collection or initial data load
- To redirect new connections to other Pods
(e.g. fail the readiness probe when the Pod's load is too high)
---
## Startup probes
## Dependencies
*My container takes a long time to boot before being able to serve traffic.*
- If a web server depends on a database to function, and the database is down:
- After creating a container, Kubernetes runs its startup probe
- the web server's liveness probe should succeed
- The container will be considered "unhealthy" until the probe succeeds
- the web server's readiness probe should fail
- As long as the container is "unhealthy", its Pod...:
- Same thing for any hard dependency (without which the container can't work)
- is not added to Services' endpoints
.warning[**Do not** fail liveness probes for problems that are external to the container]
- is not considered as "available" for rolling update purposes
---
- Readiness and liveness probes are enabled *after* startup probe reports success
## Timing and thresholds
(if there is no startup probe, readiness and liveness probes are enabled right away)
- Probes are executed at intervals of `periodSeconds` (default: 10)
- The timeout for a probe is set with `timeoutSeconds` (default: 1)
.warning[If a probe takes longer than that, it is considered as a FAIL]
- A probe is considered successful after `successThreshold` successes (default: 1)
- A probe is considered failing after `failureThreshold` failures (default: 3)
- A probe can have an `initialDelaySeconds` parameter (default: 0)
- Kubernetes will wait that amount of time before running the probe for the first time
(this is important to avoid killing services that take a long time to start)
---
## Startup probe
*The container takes too long to start, and is killed by the liveness probe!*
- By default, probes (including liveness) start immediately
- With the default probe interval and failure threshold:
*a container must respond in less than 30 seconds, or it will be killed!*
- There are two ways to avoid that:
- set `initialDelaySeconds` (a fixed, rigid delay)
- use a `startupProbe`
- Kubernetes will run only the startup probe, and when it succeeds, run the other probes
---
@@ -156,296 +178,121 @@
---
## Startup probes gotchas
- When defining a `startupProbe`, we almost always want to adjust its parameters
(specifically, its `failureThreshold` - this is explained in next slide)
- Otherwise, if the container fails to start within 30 seconds...
*Kubernetes terminates the container and restarts it!*
- Sometimes, it's easier/simpler to use a `readinessProbe` instead
(except when also using a `livenessProbe`)
---
## Timing and thresholds
- Probes are executed at intervals of `periodSeconds` (default: 10)
- The timeout for a probe is set with `timeoutSeconds` (default: 1)
.warning[If a probe takes longer than that, it is considered as a FAIL]
.warning[For liveness probes **and startup probes** this terminates and restarts the container]
- A probe is considered successful after `successThreshold` successes (default: 1)
- A probe is considered failing after `failureThreshold` failures (default: 3)
- All these parameters can be set independently for each probe
---
class: extra-details
## `initialDelaySeconds`
- A probe can have an `initialDelaySeconds` parameter (default: 0)
- Kubernetes will wait that amount of time before running the probe for the first time
- It is generally better to use a `startupProbe` instead
(but this parameter did exist before startup probes were implemented)
---
class: extra-details
## `readinessProbe` vs `startupProbe`
- A lot of blog posts / documentations / tutorials recommend readiness probes...
- ...even in scenarios where a startup probe would seem more appropriate!
- This is because startup probes are relatively recent
(they reached GA status in Kubernetes 1.20)
- When there is no `livenessProbe`, using a `readinessProbe` is simpler:
- a `startupProbe` generally requires to change the `failureThreshold`
- a `startupProbe` generally also requires a `readinessProbe`
- a single `readinessProbe` can fulfill both roles
---
## Different types of probes
- Kubernetes supports the following mechanisms:
- HTTP request
- `exec` (arbitrary program execution)
- specify URL of the request (and optional headers)
- `httpGet` (HTTP GET request)
- any status code between 200 and 399 indicates success
- `tcpSocket` (check if a TCP port is accepting connections)
- TCP connection
- `grpc` (standard [GRPC Health Checking Protocol][grpc])
- the probe succeeds if the TCP port is open
- All probes give binary results ("it works" or "it doesn't")
- arbitrary exec
- Let's see the specific details for each of them!
- a command is executed in the container
[grpc]: https://grpc.github.io/grpc/core/md_doc_health-checking.html
- exit status of zero indicates success
---
## `exec`
## Benefits of using probes
- Runs an arbitrary program *inside* the container
- Rolling updates proceed when containers are *actually ready*
(like with `kubectl exec` or `docker exec`)
(as opposed to merely started)
- The program must be available in the container image
- Containers in a broken state get killed and restarted
- Kubernetes uses the exit status of the program
(instead of serving errors or timeouts)
(standard UNIX convention: 0 = success, anything else = failure)
- Unavailable backends get removed from load balancer rotation
(thus improving response times across the board)
- If a probe is not defined, it's as if there was an "always successful" probe
---
## `exec` example
## Example: HTTP probe
When the worker is ready, it should create `/tmp/ready`.
<br/>
The following probe will give it 5 minutes to do so.
Here is a pod template for the `rng` web service of the DockerCoins app:
```yaml
apiVersion: v1
kind: Pod
metadata:
name: queueworker
name: healthy-app
spec:
containers:
- name: worker
image: myregistry.../worker:v1.0
startupProbe:
exec:
command:
- test
- -f
- /tmp/ready
failureThreshold: 30
```
---
## Using shell constructs
- If we want to use pipes, conditionals, etc. we should invoke a shell
- Example:
```yaml
exec:
command:
- sh
- -c
- "curl http://localhost:5000/status | jq .ready | grep true"
```
---
## `httpGet`
- Make an HTTP GET request to the container
- The request will be made by Kubelet
(doesn't require extra binaries in the container image)
- `port` must be specified
- `path` and extra `httpHeaders` can be specified optionally
- Kubernetes uses HTTP status code of the response:
- 200-399 = success
- anything else = failure
---
## `httpGet` example
The following liveness probe restarts the container if it stops responding on `/healthz`:
```yaml
apiVersion: v1
kind: Pod
metadata:
name: frontend
spec:
containers:
- name: frontend
image: myregistry.../frontend:v1.0
- name: myapp
image: myregistry.io/myapp:v1.0
livenessProbe:
httpGet:
path: /health
port: 80
path: /healthz
periodSeconds: 5
```
---
## `tcpSocket`
- Kubernetes checks if the indicated TCP port accepts connections
- There is no additional check
.warning[It's quite possible for a process to be broken, but still accept TCP connections!]
If the backend serves an error, or takes longer than 1s, 3 times in a row, it gets killed.
---
## `grpc`
## Example: exec probe
<!-- ##VERSION## -->
Here is a pod template for a Redis server:
- Available in beta since Kubernetes 1.24
```yaml
apiVersion: v1
kind: Pod
metadata:
name: redis-with-liveness
spec:
containers:
- name: redis
image: redis
livenessProbe:
exec:
command: ["redis-cli", "ping"]
```
- Leverages standard [GRPC Health Checking Protocol][grpc]
[grpc]: https://grpc.github.io/grpc/core/md_doc_health-checking.html
If the Redis process becomes unresponsive, it will be killed.
---
## Best practices for healthchecks
## Questions to ask before adding healthchecks
- Readiness probes are almost always beneficial
- Do we want liveness, readiness, both?
- don't hesitate to add them early!
(sometimes, we can use the same check, but with different failure thresholds)
- we can even make them *mandatory*
- Do we have existing HTTP endpoints that we can use?
- Be more careful with liveness and startup probes
- Do we need to add new endpoints, or perhaps use something else?
- they aren't always necessary
- Are our healthchecks likely to use resources and/or slow down the app?
- they can even cause harm
- Do they depend on additional services?
(this can be particularly tricky, see next slide)
---
## Readiness probes
## Healthchecks and dependencies
- Almost always beneficial
- Liveness checks should not be influenced by the state of external services
- Exceptions:
- All checks should reply quickly (by default, less than 1 second)
- web service that doesn't have a dedicated "health" or "ping" route
- Otherwise, they are considered to fail
- ...and all requests are "expensive" (e.g. lots of external calls)
- This might require to check the health of dependencies asynchronously
---
## Liveness probes
- If we're not careful, we end up restarting containers for no reason
(which can cause additional load on the cluster, cascading failures, data loss, etc.)
- Suggestion:
- don't add liveness probes immediately
- wait until you have a bit of production experience with that code
- then add narrow-scoped healthchecks to detect specific failure modes
- Readiness and liveness probes should be different
(different check *or* different timeouts *or* different thresholds)
---
## Startup probes
- Only beneficial for containers that need a long time to start
(more than 30 seconds)
- If there is no liveness probe, it's simpler to just use a readiness probe
(since we probably want to have a readiness probe anyway)
- In other words, startup probes are useful in one situation:
*we have a liveness probe, AND the container needs a lot of time to start*
- Don't forget to change the `failureThreshold`
(otherwise the container will fail to start and be killed)
---
## Recap of the gotchas
- The default timeout is 1 second
- if a probe takes longer than 1 second to reply, Kubernetes considers that it fails
- this can be changed by setting the `timeoutSeconds` parameter
<br/>(or refactoring the probe)
- Liveness probes should not be influenced by the state of external services
- Liveness probes and readiness probes should have different paramters
- For startup probes, remember to increase the `failureThreshold`
(e.g. if a database or API might be healthy but still take more than
1 second to reply, we should check the status asynchronously and report
a cached status)
---
@@ -453,21 +300,21 @@ spec:
(In that context, worker = process that doesn't accept connections)
- A relatively easy solution is to use files
- Readiness is useful mostly for rolling updates
- For a startup or readiness probe:
(because workers aren't backends for a service)
- worker creates `/tmp/ready` when it's ready
- probe checks the existence of `/tmp/ready`
- Liveness may help us restart a broken worker, but how can we check it?
- For a liveness probe:
- Embedding an HTTP server is a (potentially expensive) option
- worker touches `/tmp/alive` regularly
<br/>(e.g. just before starting to work on a job)
- probe checks that the timestamp on `/tmp/alive` is recent
- if the timestamp is old, it means that the worker is stuck
- Using a "lease" file can be relatively easy:
- Sometimes it can also make sense to embed a web server in the worker
- touch a file during each iteration of the main loop
- check the timestamp of that file from an exec probe
- Writing logs (and checking them from the probe) also works
???

View File

@@ -317,22 +317,6 @@ class: extra-details
class: extra-details
## Determining if we're in a subchart
- `.Chart.IsRoot` indicates if we're in the top-level chart or in a sub-chart
- Useful in charts that are designed to be used standalone or as dependencies
- Example: generic chart
- when used standalone (`.Chart.IsRoot` is `true`), use `.Release.Name`
- when used as a subchart e.g. with multiple aliases, use `.Chart.Name`
---
class: extra-details
## Compatibility with Helm 2
- Chart `apiVersion: v1` is the only version supported by Helm 2

View File

@@ -1,148 +0,0 @@
## Ingress and canary releases
- Let's see how to implement *canary releases*
- The example here will use Traefik v1
(which is obsolete)
- It won't work on your Kubernetes cluster!
(unless you're running an oooooold version of Kubernetes)
(and an equally oooooooold version of Traefik)
- We've left it here just as an example!
---
## Canary releases
- A *canary release* (or canary launch or canary deployment) is a release that will process only a small fraction of the workload
- After deploying the canary, we compare its metrics to the normal release
- If the metrics look good, the canary will progressively receive more traffic
(until it gets 100% and becomes the new normal release)
- If the metrics aren't good, the canary is automatically removed
- When we deploy a bad release, only a tiny fraction of traffic is affected
---
## Various ways to implement canary
- Example 1: canary for a microservice
- 1% of all requests (sampled randomly) are sent to the canary
- the remaining 99% are sent to the normal release
- Example 2: canary for a web app
- 1% of users are sent to the canary web site
- the remaining 99% are sent to the normal release
- Example 3: canary for shipping physical goods
- 1% of orders are shipped with the canary process
- the remaining 99% are shipped with the normal process
- We're going to implement example 1 (per-request routing)
---
## Canary releases with Traefik v1
- We need to deploy the canary and expose it with a separate service
- Then, in the Ingress resource, we need:
- multiple `paths` entries (one for each service, canary and normal)
- an extra annotation indicating the weight of each service
- If we want, we can send requests to more than 2 services
---
## The Ingress resource
.small[
```yaml
apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
name: rgb
annotations:
traefik.ingress.kubernetes.io/service-weights: |
red: 50%
green: 25%
blue: 25%
spec:
rules:
- host: rgb.`A.B.C.D`.nip.io
http:
paths:
- path: /
backend:
serviceName: red
servicePort: 80
- path: /
backend:
serviceName: green
servicePort: 80
- path: /
backend:
serviceName: blue
servicePort: 80
```
]
---
class: extra-details
## Other ingress controllers
*Just to illustrate how different things are ...*
- With the NGINX ingress controller:
- define two ingress ressources
<br/>
(specifying rules with the same host+path)
- add `nginx.ingress.kubernetes.io/canary` annotations on each
- With Linkerd2:
- define two services
- define an extra service for the weighted aggregate of the two
- define a TrafficSplit (this is a CRD introduced by the SMI spec)
---
class: extra-details
## We need more than that
What we saw is just one of the multiple building blocks that we need to achieve a canary release.
We also need:
- metrics (latency, performance ...) for our releases
- automation to alter canary weights
(increase canary weight if metrics look good; decrease otherwise)
- a mechanism to manage the lifecycle of the canary releases
(create them, promote them, delete them ...)
For inspiration, check [flagger by Weave](https://github.com/weaveworks/flagger).

View File

@@ -1,36 +1,34 @@
# Exposing HTTP services with Ingress resources
- Service = layer 4 (TCP, UDP, SCTP)
- HTTP services are typically exposed on port 80
- works with every TCP/UDP/SCTP protocol
(and 443 for HTTPS)
- doesn't "see" or interpret HTTP
- `NodePort` services are great, but they are *not* on port 80
- Ingress = layer 7 (HTTP)
(by default, they use port range 30000-32767)
- only for HTTP
- can route requests depending on URI or host header
- can handle TLS
- How can we get *many* HTTP services on port 80? 🤔
---
## Why should we use Ingress resources?
## Various ways to expose something on port 80
A few use-cases:
- Service with `type: LoadBalancer`
- URI routing (e.g. for single page apps)
*costs a little bit of money; not always available*
`/api` → service `api:5000`
- Service with one (or multiple) `ExternalIP`
everything else → service `static:80`
*requires public nodes; limited by number of nodes*
- Cost optimization
- Service with `hostPort` or `hostNetwork`
(because individual `LoadBalancer` services typically cost money)
*same limitations as `ExternalIP`; even harder to manage*
- Automatic handling of TLS certificates
- Ingress resources
*addresses all these limitations, yay!*
---
@@ -183,70 +181,20 @@ class: extra-details
---
## Accepting connections on port 80 (and 443)
- Web site users don't want to specify port numbers
(e.g. "connect to https://blahblah.whatever:31550")
- Our ingress controller needs to actually be exposed on port 80
(and 443 if we want to handle HTTPS)
- Let's see how we can achieve that!
---
## Various ways to expose something on port 80
- Service with `type: LoadBalancer`
*costs a little bit of money; not always available*
- Service with one (or multiple) `ExternalIP`
*requires public nodes; limited by number of nodes*
- Service with `hostPort` or `hostNetwork`
*same limitations as `ExternalIP`; even harder to manage*
---
## Deploying pods listening on port 80
- We are going to run Traefik in Pods with `hostNetwork: true`
- We want our ingress load balancer to be available on port 80
(so that our load balancer can use the "real" port 80 of our nodes)
- The best way to do that would be with a `LoadBalancer` service
- Traefik Pods will be created by a DaemonSet
... but it requires support from the underlying infrastructure
(so that we get one instance of Traefik on every node of the cluster)
- Instead, we are going to use the `hostNetwork` mode on the Traefik pods
- This means that we will be able to connect to any node of the cluster on port 80
.warning[This is not typical of a production setup!]
- Let's see what this `hostNetwork` mode is about ...
---
## Doing it in production
- When running "on cloud", the easiest option is a `LoadBalancer` service
- When running "on prem", it depends:
- [MetalLB] is a good option if a pool of public IP addresses is available
- otherwise, using `externalIPs` on a few nodes (2-3 for redundancy)
- Many variations/optimizations are possible depending on our exact scenario!
[MetalLB]: https://metallb.org/
---
class: extra-details
## Without `hostNetwork`
- Normally, each pod gets its own *network namespace*
@@ -263,8 +211,6 @@ class: extra-details
---
class: extra-details
## With `hostNetwork: true`
- No network namespace gets created
@@ -283,6 +229,26 @@ class: extra-details
---
class: extra-details
## Other techniques to expose port 80
- We could use pods specifying `hostPort: 80`
... but with most CNI plugins, this [doesn't work or requires additional setup](https://github.com/kubernetes/kubernetes/issues/23920)
- We could use a `NodePort` service
... but that requires [changing the `--service-node-port-range` flag in the API server](https://kubernetes.io/docs/reference/command-line-tools-reference/kube-apiserver/)
- We could create a service with an external IP
... this would work, but would require a few extra steps
(figuring out the IP address and adding it to the service)
---
## Running Traefik
- The [Traefik documentation][traefikdoc] recommends to use a Helm chart
@@ -304,8 +270,6 @@ class: extra-details
---
class: extra-details
## Taints and tolerations
- A *taint* is an attribute added to a node
@@ -532,6 +496,10 @@ This is normal: we haven't provided any ingress rule yet.
## Creating ingress resources
- Before Kubernetes 1.19, we must use YAML manifests
(see example on next slide)
- Since Kubernetes 1.19, we can use `kubectl create ingress`
```bash
@@ -566,21 +534,7 @@ This is normal: we haven't provided any ingress rule yet.
---
## Before Kubernetes 1.19
- Before Kubernetes 1.19:
- `kubectl create ingress` wasn't available
- `apiVersion: networking.k8s.io/v1` wasn't supported
- It was necessary to use YAML, and `apiVersion: networking.k8s.io/v1beta1`
(see example on next slide)
---
## YAML for old ingress resources
## Ingress resources in YAML
Here is a minimal host-based ingress resource:
@@ -601,15 +555,23 @@ spec:
```
(It is in `k8s/ingress.yaml`.)
---
## YAML for new ingress resources
class: extra-details
## Ingress API version
- The YAML on the previous slide uses `apiVersion: networking.k8s.io/v1beta1`
- Starting with Kubernetes 1.19, `networking.k8s.io/v1` is available
- And we can use `kubectl create ingress` 🎉
- However, with Kubernetes 1.19 (and later), we can use `kubectl create ingress`
- We can see "modern" YAML with `-o yaml --dry-run=client`:
- We chose to keep an "old" (deprecated!) YAML example for folks still using older versions of Kubernetes
- If we want to see "modern" YAML, we can use `-o yaml --dry-run=client`:
```bash
kubectl create ingress red -o yaml --dry-run=client \
@@ -679,6 +641,157 @@ class: extra-details
- It is still in alpha stage
---
## Vendor-specific example
- Let's see how to implement *canary releases*
- The example here will use Traefik v1
(which is obsolete)
- It won't work on your Kubernetes cluster!
(unless you're running an oooooold version of Kubernetes)
(and an equally oooooooold version of Traefik)
- We've left it here just as an example!
---
## Canary releases
- A *canary release* (or canary launch or canary deployment) is a release that will process only a small fraction of the workload
- After deploying the canary, we compare its metrics to the normal release
- If the metrics look good, the canary will progressively receive more traffic
(until it gets 100% and becomes the new normal release)
- If the metrics aren't good, the canary is automatically removed
- When we deploy a bad release, only a tiny fraction of traffic is affected
---
## Various ways to implement canary
- Example 1: canary for a microservice
- 1% of all requests (sampled randomly) are sent to the canary
- the remaining 99% are sent to the normal release
- Example 2: canary for a web app
- 1% of users are sent to the canary web site
- the remaining 99% are sent to the normal release
- Example 3: canary for shipping physical goods
- 1% of orders are shipped with the canary process
- the remaining 99% are shipped with the normal process
- We're going to implement example 1 (per-request routing)
---
## Canary releases with Traefik v1
- We need to deploy the canary and expose it with a separate service
- Then, in the Ingress resource, we need:
- multiple `paths` entries (one for each service, canary and normal)
- an extra annotation indicating the weight of each service
- If we want, we can send requests to more than 2 services
---
## The Ingress resource
.small[
```yaml
apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
name: rgb
annotations:
traefik.ingress.kubernetes.io/service-weights: |
red: 50%
green: 25%
blue: 25%
spec:
rules:
- host: rgb.`A.B.C.D`.nip.io
http:
paths:
- path: /
backend:
serviceName: red
servicePort: 80
- path: /
backend:
serviceName: green
servicePort: 80
- path: /
backend:
serviceName: blue
servicePort: 80
```
]
---
class: extra-details
## Other ingress controllers
*Just to illustrate how different things are ...*
- With the NGINX ingress controller:
- define two ingress ressources
<br/>
(specifying rules with the same host+path)
- add `nginx.ingress.kubernetes.io/canary` annotations on each
- With Linkerd2:
- define two services
- define an extra service for the weighted aggregate of the two
- define a TrafficSplit (this is a CRD introduced by the SMI spec)
---
class: extra-details
## We need more than that
What we saw is just one of the multiple building blocks that we need to achieve a canary release.
We also need:
- metrics (latency, performance ...) for our releases
- automation to alter canary weights
(increase canary weight if metrics look good; decrease otherwise)
- a mechanism to manage the lifecycle of the canary releases
(create them, promote them, delete them ...)
For inspiration, check [flagger by Weave](https://github.com/weaveworks/flagger).
???
:EN:- The Ingress resource

View File

@@ -86,8 +86,8 @@
(This is inspired by the
[uselessoperator](https://github.com/tilt-dev/uselessoperator)
written by
[V Körbes](https://twitter.com/veekorbes).
written by
[L Körbes](https://twitter.com/ellenkorbes).
Highly recommend!💯)
---
@@ -160,31 +160,34 @@ type MachineSpec struct {
We can use Go *marker comments* to give `controller-gen` extra details about how to handle our type, for instance:
```go
//+kubebuilder:object:root=true
```
→ top-level type exposed through API (as opposed to "member field of another type")
```go
//+kubebuilder:subresource:status
```
→ automatically generate a `status` subresource (very common with many types)
```go
//+kubebuilder:printcolumn:JSONPath=".spec.switchPosition",name=Position,type=string
// +kubebuilder:object:root=true
// +kubebuilder:subresource:status
// +kubebuilder:printcolumn:JSONPath=".spec.switchPosition",name=Position,type=string
```
(See
[marker syntax](https://book.kubebuilder.io/reference/markers.html),
[CRD generation](https://book.kubebuilder.io/reference/markers/crd.html),
[CRD validation](https://book.kubebuilder.io/reference/markers/crd-validation.html),
[Object/DeepCopy](https://master.book.kubebuilder.io/reference/markers/object.html)
[CRD validation](https://book.kubebuilder.io/reference/markers/crd-validation.html)
)
---
class: extra-details
## Using CRD v1
- By default, kubebuilder generates v1alpha1 CRDs
- If we want to generate v1 CRDs:
- edit `Makefile`
- update `crd:crdVersions=v1`
---
## Installing the CRD
After making these changes, we can run `make install`.
@@ -205,7 +208,6 @@ Edit `config/samples/useless_v1alpha1_machine.yaml`:
kind: Machine
apiVersion: useless.container.training/v1alpha1
metadata:
labels: # ...
name: machine-1
spec:
# Our useless operator will change that to "down"
@@ -250,23 +252,20 @@ spec:
## Loading an object
Open `controllers/machine_controller.go`.
Add that code in the `Reconcile` method, at the `TODO(user)` location:
Open `controllers/machine_controller.go` and add that code in the `Reconcile` method:
```go
var machine uselessv1alpha1.Machine
logger := log.FromContext(ctx)
if err := r.Get(ctx, req.NamespacedName, &machine); err != nil {
logger.Info("error getting object")
return ctrl.Result{}, err
log.Info("error getting object")
return ctrl.Result{}, err
}
logger.Info(
"reconciling",
"machine", req.NamespacedName,
"switchPosition", machine.Spec.SwitchPosition,
r.Log.Info(
"reconciling",
"machine", req.NamespaceName,
"switchPosition", machine.Spec.SwitchPosition,
)
```
@@ -289,7 +288,7 @@ Then:
--
We get a bunch of errors and go stack traces! 🤔
🤔
---
@@ -325,7 +324,7 @@ Let's try to update the machine like this:
if machine.Spec.SwitchPosition != "down" {
machine.Spec.SwitchPosition = "down"
if err := r.Update(ctx, &machine); err != nil {
logger.Info("error updating switch position")
log.Info("error updating switch position")
return ctrl.Result{}, client.IgnoreNotFound(err)
}
}
@@ -345,9 +344,9 @@ Again - update, `make run`, test.
(maybe with degraded behavior in the meantime)
- Status will almost always be a sub-resource, so that it can be updated separately
- Status will almost always be a sub-resource
(and potentially with different permissions)
(so that it can be updated separately "cheaply")
---
@@ -400,8 +399,8 @@ class: extra-details
## To requeue ...
`return ctrl.Result{RequeueAfter: 1 * time.Second}, nil`
`return ctrl.Result{RequeueAfter: 1 * time.Second}`
- That means: "try again in 1 second, and I will check if progress was made"
- This *does not* guarantee that we will be called exactly 1 second later:
@@ -410,9 +409,7 @@ class: extra-details
- we might be called after (if the controller is busy with other objects)
- If we are waiting for another Kubernetes resource to change, there is a better way
(explained on next slide)
- If we are waiting for another resource to change, there is an even better way!
---
@@ -420,41 +417,23 @@ class: extra-details
`return ctrl.Result{}, nil`
- That means: "we're done here!"
- That means: "no need to set an alarm; we'll be notified some other way"
- This is also what we should use if we are waiting for another resource
- Use this if we are waiting for another resource to update
(e.g. a LoadBalancer to be provisioned, a Pod to be ready...)
- In that case, we will need to set a *watch* (more on that later)
---
## Keeping track of state
- If we simply requeue the object to examine it 1 second later...
- ...We'll keep examining/requeuing it forever!
- We need to "remember" that we saw it (and when)
- Option 1: keep state in controller
(e.g. an internal `map`)
- Option 2: keep state in the object
(typically in its status field)
- Tradeoffs: concurrency / failover / control plane overhead...
- For this to work, we need to set a *watch* (more on that later)
---
## "Improving" our controller, take 2
Let's store in the machine status the moment when we saw it:
- Let's store in the machine status the moment when we saw it
```go
// +kubebuilder:printcolumn:JSONPath=".status.seenAt",name=Seen,type=date
type MachineStatus struct {
// Time at which the machine was noticed by our controller.
SeenAt *metav1.Time ``json:"seenAt,omitempty"``
@@ -467,12 +446,6 @@ Note: `date` fields don't display timestamps in the future.
(That's why for this example it's simpler to use `seenAt` rather than `changeAt`.)
And for better visibility, add this along with the other `printcolumn` comments:
```go
//+kubebuilder:printcolumn:JSONPath=".status.seenAt",name=Seen,type=date
```
---
## Set `seenAt`
@@ -484,7 +457,7 @@ if machine.Status.SeenAt == nil {
now := metav1.Now()
machine.Status.SeenAt = &now
if err := r.Status().Update(ctx, &machine); err != nil {
logger.Info("error updating status.seenAt")
log.Info("error updating status.seenAt")
return ctrl.Result{}, client.IgnoreNotFound(err)
}
return ctrl.Result{RequeueAfter: 5 * time.Second}, nil
@@ -505,9 +478,8 @@ if machine.Spec.SwitchPosition != "down" {
changeAt := machine.Status.SeenAt.Time.Add(5 * time.Second)
if now.Time.After(changeAt) {
machine.Spec.SwitchPosition = "down"
machine.Status.SeenAt = nil
if err := r.Update(ctx, &machine); err != nil {
logger.Info("error updating switch position")
log.Info("error updating switch position")
return ctrl.Result{}, client.IgnoreNotFound(err)
}
}
@@ -524,33 +496,15 @@ if machine.Spec.SwitchPosition != "down" {
- We will now have two kinds of objects: machines, and switches
- Machines will store the number of switches in their spec
- Machines should have *at least* one switch, possibly *multiple ones*
- Our controller will automatically create switches if needed
(a bit like the ReplicaSet controller automatically creates Pods)
- The switches will be tied to their machine through a label
(let's pick `machine=name-of-the-machine`)
---
## Switch state
- The position of a switch will now be stored in the switch
(not in the machine like in the first scenario)
- The position will now be stored in the switch, not the machine
- The machine will also expose the combined state of the switches
(through its status)
- The switches will be tied to their machine through a label
- The machine's status will be automatically updated by the controller
(each time a switch is added/changed/removed)
(See next slide for an example)
---
@@ -562,7 +516,7 @@ NAME SWITCHES POSITIONS
machine-cz2vl 3 ddd
machine-vf4xk 1 d
[jp@hex ~]$ kubectl get switches --show-labels
[jp@hex ~]$ kubectl get switches --show-labels
NAME POSITION SEEN LABELS
switch-6wmjw down machine=machine-cz2vl
switch-b8csg down machine=machine-cz2vl
@@ -576,95 +530,39 @@ switch-rc59l down machine=machine-vf4xk
## Tasks
1. Create the new resource type (but don't create a controller)
2. Update `machine_types.go` and `switch_types.go`
3. Implement logic to display machine status (status of its switches)
4. Implement logic to automatically create switches
5. Implement logic to flip all switches down immediately
6. Then tweak it so that a given machine doesn't flip more than one switch every 5 seconds
*See next slides for detailed steps!*
---
## Creating the new type
Create the new resource type (but don't create a controller):
```bash
kubebuilder create api --group useless --version v1alpha1 --kind Switch
```
Note: this time, only create a new custom resource; not a new controller.
Update `machine_types.go` and `switch_types.go`.
Implement the logic so that the controller flips all switches down immediately.
Then change it so that a given machine doesn't flip more than one switch every 5 seconds.
See next slides for hints!
---
## Updating our types
## Listing objects
- Move the "switch position" and "seen at" to the new `Switch` type
We can use the `List` method with filters:
- Update the `Machine` type to have:
```go
var switches uselessv1alpha1.SwitchList
- `spec.switches` (Go type: `int`, JSON type: `integer`)
if err := r.List(ctx, &switches,
client.InNamespace(req.Namespace),
client.MatchingLabels{"machine": req.Name},
); err != nil {
log.Error(err, "unable to list switches of the machine")
return ctrl.Result{}, client.IgnoreNotFound(err)
}
- `status.positions` of type `string`
- Bonus points for adding [CRD Validation](https://book.kubebuilder.io/reference/markers/crd-validation.html) to the numbers of switches!
- Then install the new CRDs with `make install`
- Create a Machine, and a Switch linked to the Machine (by setting the `machine` label)
---
## Listing switches
- Switches are associated to Machines with a label
(`kubectl label switch switch-xyz machine=machine-xyz`)
- We can retrieve associated switches like this:
```go
var switches uselessv1alpha1.SwitchList
if err := r.List(ctx, &switches,
client.InNamespace(req.Namespace),
client.MatchingLabels{"machine": req.Name},
); err != nil {
logger.Error(err, "unable to list switches of the machine")
return ctrl.Result{}, client.IgnoreNotFound(err)
}
logger.Info("Found switches", "switches", switches)
```
---
## Updating status
- Each time we reconcile a Machine, let's update its status:
```go
status := ""
for _, sw := range switches.Items {
status += string(sw.Spec.Position[0])
}
machine.Status.Positions = status
if err := r.Status().Update(ctx, &machine); err != nil {
...
```
- Run the controller and check that POSITIONS gets updated
- Add more switches linked to the same machine
- ...The POSITIONS don't get updated, unless we restart the controller
- We'll see later how to fix that!
log.Info("Found switches", "switches", switches)
```
---
@@ -692,28 +590,20 @@ if err := r.Create(ctx, &sw); err != nil { ...
---
## Create missing switches
- In our reconciler, if a machine doesn't have enough switches, create them!
- Option 1: directly create the number of missing switches
- Option 2: create only one switch (and rely on later requeuing)
- Note: option 2 won't quite work yet, since we haven't set up *watches* yet
---
## Watches
- Our controller doesn't react when switches are created/updated/deleted
- Our controller will correctly flip switches when it starts
- It will also react to machine updates
- But it won't react if we directly touch the switches!
- By default, it only monitors machines, not switches
- We need to tell it to watch switches
- We also need to tell it how to map a switch to its machine
(so that the correct machine gets queued and reconciled when a switch is updated)
---
## Mapping a switch to its machine
@@ -721,15 +611,16 @@ if err := r.Create(ctx, &sw); err != nil { ...
Define the following helper function:
```go
func (r *MachineReconciler) machineOfSwitch(obj client.Object) []ctrl.Request {
return []ctrl.Request{
ctrl.Request{
NamespacedName: types.NamespacedName{
Name: obj.GetLabels()["machine"],
Namespace: obj.GetNamespace(),
},
},
}
func (r *MachineReconciler) machineOfSwitch(obj handler.MapObject) []ctrl.Request {
r.Log.Debug("mos", "obj", obj)
return []ctrl.Request{
ctrl.Request{
NamespacedName: types.NamespacedName{
Name: obj.Meta.GetLabels()["machine"],
Namespace: obj.Meta.GetNamespace(),
},
},
}
}
```
@@ -740,46 +631,24 @@ func (r *MachineReconciler) machineOfSwitch(obj client.Object) []ctrl.Request {
Update the `SetupWithManager` method in the controller:
```go
// SetupWithManager sets up the controller with the Manager.
func (r *MachineReconciler) SetupWithManager(mgr ctrl.Manager) error {
return ctrl.NewControllerManagedBy(mgr).
For(&uselessv1alpha1.Machine{}).
Owns(&uselessv1alpha1.Switch{}).
Watches(
&source.Kind{Type: &uselessv1alpha1.Switch{}},
handler.EnqueueRequestsFromMapFunc(r.machineOfSwitch),
).
Complete(r)
return ctrl.NewControllerManagedBy(mgr).
For(&uselessv1alpha1.Machine{}).
Owns(&uselessv1alpha1.Switch{}).
Watches(
&source.Kind{Type: &uselessv1alpha1.Switch{}},
&handler.EnqueueRequestsFromMapFunc{
ToRequests: handler.ToRequestsFunc(r.machineOfSwitch),
}).
Complete(r)
}
```
---
## ...And a few extra imports
Import the following packages referenced by the previous code:
```go
"sigs.k8s.io/controller-runtime/pkg/handler"
"sigs.k8s.io/controller-runtime/pkg/source"
"k8s.io/apimachinery/pkg/types"
```
After this, when we update a switch, it should reflect on the machine.
(Try to change switch positions and see the machine status update!)
After this, our controller should now react to switch changes.
---
## Flipping switches
- Now re-add logic to flip switches that are not in "down" position
- Re-add logic to wait a few seconds before flipping a switch
- Change the logic to toggle one switch per machine every few seconds
(i.e. don't change all the switches for a machine; move them one at a time)
## Bonus points
- Handle "scale down" of a machine (by deleting extraneous switches)
@@ -791,25 +660,9 @@ After this, when we update a switch, it should reflect on the machine.
---
## Other possible improvements
- Formalize resource ownership
(by setting `ownerReferences` in the switches)
- This can simplify the watch mechanism a bit
- Allow to define a selector
(instead of using the hard-coded `machine` label)
- And much more!
---
## Acknowledgements
- Useless Operator, by [V Körbes](https://twitter.com/veekorbes)
- Useless Operator, by [L Körbes](https://twitter.com/ellenkorbes)
[code](https://github.com/tilt-dev/uselessoperator)
|

View File

@@ -141,7 +141,7 @@ class: extra-details
- There are external tools to address these shortcomings
(e.g.: [Stern](https://github.com/stern/stern))
(e.g.: [Stern](https://github.com/wercker/stern))
---

View File

@@ -170,9 +170,9 @@ def hash_bytes(data):
headers={"Content-Type": "application/octet-stream"})
```
(Feel free to check the [full source code][dockercoins-worker-code] of the worker!)
[dockercoins-worker-code]: https://@@GITREPO@@/blob/8279a3bce9398f7c1a53bdd95187c53eda4e6435/dockercoins/worker/worker.py#L17
(Full source code available [here](
https://@@GITREPO@@/blob/8279a3bce9398f7c1a53bdd95187c53eda4e6435/dockercoins/worker/worker.py#L17
))
---

View File

@@ -142,7 +142,7 @@ configMapGenerator:
- overlays can only *add* resources, not *remove* them
- See the full list of [eschewed features](https://kubectl.docs.kubernetes.io/faq/kustomize/eschewedfeatures/) for more details
- See the full list of [eschewed features](https://github.com/kubernetes-sigs/kustomize/blob/master/docs/eschewedFeatures.md) for more details
---

View File

@@ -156,7 +156,7 @@
- Install Kyverno:
```bash
kubectl create -f https://raw.githubusercontent.com/kyverno/kyverno/release-1.7/config/release/install.yaml
kubectl create -f https://raw.githubusercontent.com/kyverno/kyverno/release-1.5/definitions/release/install.yaml
```
]
@@ -302,35 +302,23 @@
---
## Comparing "old" and "new"
- The fields of the webhook payload are available through `{{ request }}`
- For UPDATE requests, we can access:
`{{ request.oldObject }}` → the object as it is right now (before the request)
`{{ request.object }}` → the object with the changes made by the request
---
## Missing labels
## Invalid references
- We can access the `color` label through `{{ request.object.metadata.labels.color }}`
- If we reference a label (or any field) that doesn't exist, the policy fails
(with an error similar to `JMESPAth query failed: Unknown key ... in path`)
- Except in *preconditions*: it then evaluates to an empty string
- To work around that, [use an OR expression][non-existence-checks]:
- We use a *precondition* to makes sure the label exists in both "old" and "new" objects
`{{ requests.object.metadata.labels.color || '' }}`
- Then in the *deny* block we can compare the old and new values
- Note that in older versions of Kyverno, this wasn't always necessary
(and reject changes)
(e.g. in *preconditions*, a missing label would evalute to an empty string)
- "Old" and "new" versions of the pod can be referenced through
[non-existence-checks]: https://kyverno.io/docs/writing-policies/jmespath/#non-existence-checks
`{{ request.oldObject }}` and `{{ request.object }}`
---
@@ -566,28 +554,6 @@ Note: the `apiVersion` field appears to be optional.
---
class: extra-details
## Managing `ownerReferences`
- By default, the generated object and triggering object have independent lifecycles
(deleting the triggering object doesn't affect the generated object)
- It is possible to associate the generated object with the triggering object
(so that deleting the triggering object also deletes the generated object)
- This is done by adding the triggering object information to `ownerReferences`
(in the generated object `metadata`)
- See [Linking resources with ownerReferences][ownerref] for an example
[ownerref]: https://kyverno.io/docs/writing-policies/generate/#linking-resources-with-ownerreferences
---
## Asynchronous creation
- Kyverno creates resources asynchronously
@@ -606,7 +572,7 @@ class: extra-details
## Footprint
- 8 CRDs
- 7 CRDs
- 5 webhooks

View File

@@ -69,14 +69,12 @@ Exactly what we need!
(no dependencies, extra libraries to install, etc)
- Binary releases are available [on GitHub][stern-releases]
- Binary releases are available [here](https://github.com/stern/stern/releases) on GitHub
- Stern is also available through most package managers
(e.g. on macOS, we can `brew install stern` or `sudo port install stern`)
[stern-releases]: https://github.com/stern/stern/releases
---
## Using Stern

View File

@@ -256,9 +256,9 @@ class: extra-details
- or stored in the node's `spec.podCIDR` field
.footnote[See [here][kubenet-plugin] for more details about this `kubenet` plugin.]
.footnote[See [here] for more details about this `kubenet` plugin.]
[kubenet-plugin]: https://kubernetes.io/docs/concepts/extend-kubernetes/compute-storage-net/network-plugins/#kubenet
[here]: https://kubernetes.io/docs/concepts/extend-kubernetes/compute-storage-net/network-plugins/#kubenet
---

View File

@@ -1,4 +1,4 @@
# Writing a tiny operator
# Writing an tiny operator
- Let's look at a simple operator

View File

@@ -14,20 +14,32 @@
- CPU is a *compressible resource*
- it can be preempted immediately without adverse effect
- if we have N CPU and need 2N, we run at 50% speed
(it can be preempted immediately without adverse effect)
- Memory is an *incompressible resource*
- it needs to be swapped out to be reclaimed; and this is costly
- if we have N GB RAM and need 2N, we might run at... 0.1% speed!
(it needs to be swapped out to be reclaimed; and this is costly)
- As a result, exceeding limits will have different consequences for CPU and memory
---
## Exceeding CPU limits
- CPU can be reclaimed instantaneously
(in fact, it is preempted hundreds of times per second, at each context switch)
- If a container uses too much CPU, it can be throttled
(it will be scheduled less often)
- The processes in that container will run slower
(or rather: they will not run faster)
---
class: extra-details
## CPU limits implementation details
@@ -134,59 +146,39 @@ For more details, check [this blog post](https://erickhun.com/posts/kubernetes-f
---
## Running low on memory
## Exceeding memory limits
- When the system runs low on memory, it starts to reclaim used memory
- Memory needs to be swapped out before being reclaimed
(we talk about "memory pressure")
- "Swapping" means writing memory pages to disk, which is very slow
- Option 1: free up some buffers and caches
- On a classic system, a process that swaps can get 1000x slower
(fastest option; might affect performance if cache memory runs very low)
(because disk I/O is 1000x slower than memory I/O)
- Option 2: swap, i.e. write to disk some memory of one process to give it to another
- Exceeding the memory limit (even by a small amount) can reduce performance *a lot*
(can have a huge negative impact on performance because disks are slow)
- Kubernetes *does not support swap* (more on that later!)
- Option 3: terminate a process and reclaim all its memory
(OOM or Out Of Memory Killer on Linux)
- Exceeding the memory limit will cause the container to be killed
---
## Memory limits on Kubernetes
## Limits vs requests
- Kubernetes *does not support swap*
(but it may support it in the future, thanks to [KEP 2400])
- If a container exceeds its memory *limit*, it gets killed immediately
- If a node is overcommitted and under memory pressure, it will terminate some pods
(see next slide for some details about what "overcommit" means here!)
[KEP 2400]: https://github.com/kubernetes/enhancements/blob/master/keps/sig-node/2400-node-swap/README.md#implementation-history
---
## Overcommitting resources
- *Limits* are "hard limits" (a container *cannot* exceed its limits)
- Limits are "hard limits" (they can't be exceeded)
- a container exceeding its memory limit is killed
- a container exceeding its CPU limit is throttled
- On a given node, the sum of pod *limits* can be higher than the node size
- Requests are used for scheduling purposes
- *Requests* are used for scheduling purposes
- a container using *less* than what it requested will never be killed or throttled
- a container can use more than its requested CPU or RAM amounts
- the scheduler uses the requested sizes to determine placement
- a container using *less* than what it requested should never be killed or throttled
- On a given node, the sum of pod *requests* cannot be higher than the node size
- the resources requested by all pods on a node will never exceed the node size
---
@@ -230,31 +222,9 @@ Each pod is assigned a QoS class (visible in `status.qosClass`).
---
class: extra-details
## Where is my swap?
## CPU and RAM reservation
- Kubernetes passes resources requests and limits to the container engine
- The container engine applies these requests and limits with specific mechanisms
- Example: on Linux, this is typically done with control groups aka cgroups
- Most systems use cgroups v1, but cgroups v2 are slowly being rolled out
(e.g. available in Ubuntu 22.04 LTS)
- Cgroups v2 have new, interesting features for memory control:
- ability to set "minimum" memory amounts (to effectively reserve memory)
- better control on the amount of swap used by a container
---
class: extra-details
## What's the deal with swap?
- The semantics of memory and swap limits on Linux cgroups are complex
- With cgroups v1, it's not possible to disable swap for a cgroup
@@ -268,8 +238,6 @@ class: extra-details
- The simplest solution was to disable swap entirely
- Kubelet will refuse to start if it detects that swap is enabled!
---
## Alternative point of view
@@ -300,7 +268,7 @@ class: extra-details
- You will need to add the flag `--fail-swap-on=false` to kubelet
(remember: it won't otherwise start if it detects that swap is enabled)
(otherwise, it won't start!)
---
@@ -698,18 +666,6 @@ class: extra-details
---
## Underutilization
- Remember: when assigning a pod to a node, the scheduler looks at *requests*
(not at current utilization on the node)
- If pods request resources but don't use them, this can lead to underutilization
(because the scheduler will consider that the node is full and can't fit new pods)
---
## Viewing a namespace limits and quotas
- `kubectl describe namespace` will display resource limits and quotas

View File

@@ -18,37 +18,15 @@
---
### CoLiMa
- Container runtimes for LiMa
(LiMa = Linux on macOS)
- For macOS only (Intel and ARM architectures)
- CLI-driven (no GUI like Docker/Rancher Desktop)
- Supports containerd, Docker, Kubernetes
- Installable with brew, nix, or ports
- More info: https://github.com/abiosoft/colima
---
## Docker Desktop
- Available on Linux, Mac, and Windows
- Free for personal use and small businesses
(less than 250 employees and less than $10 millions in annual revenue)
- Available on Mac and Windows
- Gives you one cluster with one node
- Streamlined installation and user experience
- Very easy to use if you are already using Docker Desktop:
- Great integration with various network stacks and e.g. corporate VPNs
go to Docker Desktop preferences and enable Kubernetes
- Ideal for Docker users who need good integration between both platforms
@@ -62,11 +40,13 @@
- Runs Kubernetes nodes in Docker containers
- Can deploy multiple clusters, with multiple nodes
- Can deploy multiple clusters, with multiple nodes, and multiple master nodes
- Runs the control plane on Kubernetes nodes
- As of June 2020, two versions co-exist: stable (1.7) and beta (3.0)
- Control plane can also run on multiple nodes
- They have different syntax and options, this can be confusing
(but don't let that stop you!)
---
@@ -97,8 +77,6 @@
- Requires Docker (obviously!)
- Should also work with Podman and Rootless Docker
- Deploying a single node cluster using the latest version is simple:
```bash
kind create cluster
@@ -106,26 +84,12 @@
- More advanced scenarios require writing a short [config file](https://kind.sigs.k8s.io/docs/user/quick-start#configuring-your-kind-cluster)
(to define multiple nodes, multiple control plane nodes, set Kubernetes versions ...)
(to define multiple nodes, multiple master nodes, set Kubernetes versions ...)
- Can deploy multiple clusters
---
## [MicroK8s](https://microk8s.io/)
- Available on Linux, and since recently, on Mac and Windows as well
- The Linux version is installed through Snap
(which is pre-installed on all recent versions of Ubuntu)
- Also supports clustering (as in, multiple machines running MicroK8s)
- DNS is not enabled by default; enable it with `microk8s enable dns`
---
## [Minikube](https://minikube.sigs.k8s.io/docs/)
- The "legacy" option!
@@ -144,11 +108,23 @@
---
## [MicroK8s](https://microk8s.io/)
- Available on Linux, and since recently, on Mac and Windows as well
- The Linux version is installed through Snap
(which is pre-installed on all recent versions of Ubuntu)
- Also supports clustering (as in, multiple machines running MicroK8s)
- DNS is not enabled by default; enable it with `microk8s enable dns`
---
## [Rancher Desktop](https://rancherdesktop.io/)
- Available on Linux, Mac, and Windows
- Free and open-source
- Available on Mac and Windows
- Runs a single cluster with a single node
@@ -158,7 +134,7 @@
- Emphasis on ease of use (like Docker Desktop)
- Relatively young product (first release in May 2021)
- Very young product (first release in May 2021)
- Based on k3s and other proven components

View File

@@ -389,7 +389,7 @@ class: extra-details
- A replacement Pod is created on another Node
- ... But it doesn't start yet!
- ... But it doens't start yet!
- Why? 🤔

View File

@@ -1,11 +1,12 @@
title: |
Kubernetes Training
Thoughtworks Infrastructure
(Starring: Kubernetes!)
chat: "[Mattermost](https://live.container.training/mattermost)"
chat: "[thoughtworks-infrastructure Slack](https://skillsmatter.slack.com/archives/C03E90W6Z6U)"
gitrepo: github.com/jpetazzo/container.training
slides: https://2022-11-live.container.training/
slides: https://2022-08-thoughtworks.container.training/
#slidenumberprefix: "#SomeHashTag &mdash; "
@@ -17,79 +18,28 @@ content:
- logistics.md
- k8s/intro.md
- shared/about-slides.md
- shared/chat-room-im.md
#- shared/chat-room-zoom-meeting.md
#- shared/chat-room-zoom-webinar.md
#- shared/chat-room-im.md
#- shared/chat-room-zoom.md
- shared/prereqs.md
#- shared/webssh.md
- shared/connecting.md
- exercises/k8sfundamentals-brief.md
- exercises/localcluster-brief.md
- exercises/healthchecks-brief.md
- shared/toc.md
- # 1
#- k8s/versions-k8s.md
- shared/sampleapp.md
#- shared/composescale.md
#- shared/hastyconclusions.md
- shared/composedown.md
- k8s/concepts-k8s.md
- k8s/kubectlget.md
- k8s/kubectl-run.md
- k8s/kubenet.md
- k8s/kubectlexpose.md
- k8s/shippingimages.md
#- k8s/buildshiprun-selfhosted.md
- k8s/buildshiprun-dockerhub.md
- exercises/k8sfundamentals-details.md
- k8s/ourapponkube.md
#- k8s/exercise-wordsmith.md
- # 2
- k8s/labels-annotations.md
- k8s/kubectl-logs.md
- k8s/logs-cli.md
- k8s/namespaces.md
- k8s/yamldeploy.md
- shared/declarative.md
- k8s/declarative.md
- k8s/deploymentslideshow.md
- k8s/authoring-yaml.md
- k8s/setup-overview.md
- k8s/setup-devel.md
#- k8s/setup-managed.md
#- k8s/setup-selfhosted.md
- #1
- k8s/demo-apps.md
- k8s/netpol.md
- k8s/authn-authz.md
- exercises/netpol-details.md
- exercises/rbac-details.md
- #2
- k8s/rollout.md
- k8s/healthchecks.md
- k8s/localkubeconfig.md
- k8s/accessinternal.md
- k8s/kubectlproxy.md
- k8s/setup-devel.md
- exercises/localcluster-details.md
- # 3
#- k8s/kubectlscale.md
- k8s/scalingdockercoins.md
- shared/hastyconclusions.md
- k8s/daemonset.md
- k8s/rollout.md
- k8s/healthchecks.md
#- k8s/healthchecks-more.md
- k8s/dashboard.md
- k8s/k9s.md
- k8s/tilt.md
- exercises/healthchecks-details.md
- # 4
- k8s/ingress.md
#- k8s/ingress-tls.md
#- k8s/ingress-advanced.md
- k8s/volumes.md
#- k8s/exercise-configmap.md
#- k8s/build-with-docker.md
#- k8s/build-with-kaniko.md
- k8s/configuration.md
- k8s/secrets.md
- k8s/helm-intro.md
- exercises/ingress-details.md
- # 5
- k8s/authn-authz.md
- k8s/resource-limits.md
- k8s/metrics-server.md
- k8s/cluster-sizing.md
- k8s/batch-jobs.md
- shared/thankyou.md
- #3
- |
# (Extra material)
- k8s/deploymentslideshow.md

View File

@@ -1,27 +0,0 @@
<?xml version="1.0"?>
<html>
<head>
<style>
td {
background: #ccc;
padding: 1em;
}
</style>
</head>
<body>
<table>
<tr>
<td>Tuesday November 29th - Friday December 2nd 2022</td>
<td>
<a href="docker.yml.html">Docker</a>
</td>
</tr>
<tr>
<td>Monday December 5th - Friday December 9th 2022</td>
<td>
<a href="kube.yml.html">Kubernetes</a>
</td>
</tr>
</table>
</body>
</html>

View File

@@ -1,54 +1,25 @@
## Introductions
- Hello! We are:
- Hello! I'm Jérôme Petazzoni ([@jpetazzo])
- 👷🏻‍♀️ AJ ([@s0ulshake], [EphemeraSearch], [Quantgene])
- We'll have two 2-hour workshops
- 👩🏼‍🔬 Dana ([@bigdana])
(August 17th and 24th)
- 🐳 Jérôme ([@jpetazzo], Tiny Shell Script LLC)
- We'll do a short 5-minute break in the middle of each workshop
- Live feedback, questions, help: @@CHAT@@
- Feel free to interrupt for questions at any time!
(ask questions on the public channel so that AJ & Dana can see them;
<br/>they have more production experience than me, especially on k8s :))
- Live feedback, questions, help, useful links:
- Feel free to send questions at any time!
---
## Schedule
| Pacific | Eastern | |
|-----------|-----------|-------------------|
| 8am-10am | 11am-1pm | First period |
| 10am-11am | 1pm-2pm | Long break |
| 11am-1pm | 2pm-4pm | Second period |
| After 1pm | After 4pm | Homework, labs... |
We'll also have short breaks in the middle of each period.
@@CHAT@@
- I'll be available on that Slack channel after the workshop, too!
<!-- -->
[@alexbuisine]: https://twitter.com/alexbuisine
[EphemeraSearch]: https://ephemerasearch.com/
[@jpetazzo]: https://mastodon.social/@jpetazzo
[@jpetazzo]: https://twitter.com/jpetazzo
[@s0ulshake]: https://twitter.com/s0ulshake
[Quantgene]: https://www.quantgene.com/
[@bigdana]: https://twitter.com/bigdana
---
## Exercises
- At the end of each day, there is a series of exercises
- To make the most out of the training, please try the exercises!
(it will help to practice and memorize the content of the day)
- We recommend to take at least one hour to work on the exercises
(if you understood the content of the day, it will be much faster)
- Each day will start with a quick review of the exercises of the previous day

View File

@@ -167,6 +167,5 @@ It has been preinstalled on your workshop nodes.*
- Ctrl-b Alt-1 → rearrange windows in columns
- Ctrl-b Alt-2 → rearrange windows in rows
- Ctrl-b arrows → navigate to other windows
- Ctrl-b , → rename window
- Ctrl-b d → detach session
- tmux attach → re-attach to session