Deploy kubernetes via ansible (on cloudstack servers) with logging (efk) & monitoring (prometheus) support
An even better setup with kargo (deploy to aws, azure, baremetal, and hot k8s migration) is available here --> https://github.com/gregbkr/kubernetes-kargo-logging-monitoring
- 1 master node running : k8s for container orchestration, it will pilot and gives work to the minions
- 2(or more) minion/slave/worker nodes : running the actual containers and doing the actual work
- Efk: we will send all k8s container logs to an elasticsearch DB, via fluentd, and visualize dashboards with kibana
- Prometheus will monitoring all this infra, with grafana dashbaord
- Heapster is an alternative for monitoring your k8s cluster
- K8s dashboard addon (not efk dashboard), where you can visualize k8s component in a GUI
- Service-loadbalancer (static haproxy): which is the public gateway to access your internal k8s services (kibana, grafana)
- Dynamic loadbalancer (traefik): an alternative to haproxy, quite powerful with its dynamic service discovery and auto certification
Prerequisit:
- Cloudstack cloud provider (ex: exoscale) / but you can deploy anywhere else with a bit of adaptation in ansible. Deploying logging and monitoring stay the same at the moment you have k8s running
- A vm (ubuntu) with ansible installed, where you will run recipes, and manage k8s with kubeclt
More info: you can find an overview of that setup on my blog: https://greg.satoshi.tech/
git clone https://github.com/gregbkr/kubernetes-ansible-logging-monitoring.git k8s && cd k8s
Kubeclt is your admin local client to pilot the k8s cluster. Please use the same version as server. You will be able to talk and pilot k8s with this tool.
curl -O https://storage.googleapis.com/kubernetes-release/release/v1.4.6/bin/linux/amd64/kubectl
chmod +x kubectl
mv kubectl /usr/local/bin/kubectl
Autocompletion
source <(kubectl completion bash)
kubeclt get nod +[TAB]
If issues, see troubleshooting section.
I will use the nice setup made by Seb: https://www.exoscale.ch/syslog/2016/05/09/kubernetes-ansible/ I just added few lines in file: ansible/roles/k8s/templates/k8s-node.j2 to be able to collect logs with fluentd (# In order to have logs in /var/log/containers to be pickup by fluentd Environment="RKT_OPTS=--volume dns,kind=host,source=/etc/resolv.conf --mount volume=dns,target=/etc/resolv.conf --volume var-log,kind=host,source=/var/log --mount volume=var-log,target=/var/log" )
nano ansible/k8s.yml <-- edit k8s version, num_node, ssh_key if you want to use your own
Next step will create firewall rules k8s, master and minion nodes, and install k8s components Run recipe:
ansible-playbook ansible/k8s.yml
watch kubectl get node <-- wait for the nodes to be up
kubectl get all --all-namespaces <-- should have no error here
kubectl apply -f logging <-- all deployment declarations and configurations are here
kubectl get all --all-namespaces <-- if you see elasticsearch container restarting, please restart all nodes one time only (setting vm.max_map_count, see troubleshooting section)
From here, you should be able to access our services from your laptop, as long as your cloud server ip are public:
- kibana: http://any_minion_node_ip:30601
- ES: http://any_minion_node_ip:30200
To enable that access, we had set Type=NodePort and nodePort:35601/39200 in kibana/elasticsearch-service.yaml, to make it easier to learn at this point. Because we want to control how and from where we should be accessing our public services, we will set in a later section a loadbalancer.
Check logs coming in kibana, you just need to refresh, select Time-field name : @timestamps + create
Load and view your first dashboard: management > Saved Object > Import > logging/dashboards/elk-v1.json
It seems like two schools are gently fighting for container monitoring:
- Heapster: this new player now comes as a kind of k8s addon (you can deploy it via a simple switch in some setup). It seems to be better integrated at the moment, and even more in the future with k8s component depending on it, but still young and few features
- Prometheus: it has been around for some times, lots of nice features (alerting, application metrics) and community resources available (see the public dashboards for example)
More on which one to choose: kubernetes-retired/heapster#645
Create monitoring containers
kubectl apply -f monitoring
kubectl get all --namespace=monitoring
Prometheus
Access the gui: http://any_minion_node_ip:30090
Go to status > target : you should see only some green. We got one false positive error scaping k8s-node with 2 ports 9102 and 80. As long as 9102 is good, we got the data. If you got some "context deadline exceeded" or "getsockopt connection refused", you will have to open firewall rule between the nodes. For exemple in security group k8s, you need to open 9100 and 10255.
Try a query: "node_memory_Active" > Execute > Graph --> you should see 2 lines representing both nodes.
Grafana
Login to the interface with login:admin | pass:admin) : http://any_minion_node_ip:30000 Load some dashboards: dashboard > home
Kubernetes pod resources
Prometheus stats
Load other public dashboards
Grafana GUI > Dashboards > Import
Already loaded:
- prometheus stats: https://grafana.net/dashboards/2
- kubernetes cluster : https://grafana.net/dashboards/162
Other good dashboards :
-
node exporter: https://grafana.net/dashboards/704 - https://grafana.net/dashboards/22
-
deployment: pod metrics: https://grafana.net/dashboards/747 - pod resources: https://grafana.net/dashboards/737
kubectl apply -f monitoring2
kubectl get all --namespace=monitoring2
Access services
- Grafana2: http://any_minion_node_ip:30002
You can load Cluster or Pods dashboards. When viewing Pods, type manually "namespace=monitoring2" to view stats for the related containers.
Dashboard addon let you see k8s services and containers via a nice GUI.
kubectl apply -f kubernetes-dashboard
kubectl get all --namespace=kube-system <-- carefull dashboard is running in namespace=kube-system
Access GUI: http://any_minion_node_ip:30999
If you are on aws or google cloud, these provider we automatically set a loadbalancer matching the *-ingress.yaml configuration. For all other cloud provider and baremetal, you will have to take care of that step. Luckyly, I will present you two types of loadlancer below ;-)
- service-loadbalancer (static haproxy) https://github.com/kubernetes/contrib/tree/master/service-loadbalancer
- traefik (dynamic proxy) https://github.com/containous/traefik
Create the load-balancer to be able to connect your service from the internet. Give 1 or more nodes the loadbalancer role:
kubectl label node 185.19.30.121 role=loadbalancer
kubectl apply -f service-loadbalancer-daemonset.yaml
If you change the config, use "kubectl delete -f service-loadbalancer.yaml" to force a delete/create, then the discovery of the newly created service. Add/remove services? please edit service-loadbalancer.yaml
Access services
- kibana (logging): http://lb_node_ip:5601
- grafana (monitoring): http://lb_node_ip:3000 (admin/admin)
- prometheus (monitoring): http://lb_node_ip:3000
- grafana2 (monitoring2): http://lb_node_ip:3002
- kubernetes-dashboard: http://lb_node_ip:9999
Any news services, exposed by *-ingress.yaml, will be caught by traefik and made available without restart.
To experience the full power of traefik, please purchase a domain name (ex: satoshi.tech), and point that record to the node you choose to be the lb. This record will help create the automatic certificate via the acme standard.
- satoshi.tech --> lb_node_ip
Then for each services you will use, create a dns A record:
- kibana.satoshi.tech --> lb_node_ip
- grafana.satoshi.tech --> lb_node_ip
- prometheus.satoshi.tech --> lb_node_ip
- grafana2.satoshi.tech --> lb_node_ip
- kubernetes-dashboard.satoshi.tech --> lb_node_ip
- traefik-console.satoshi.tech --> lb_node_ip
Based on which name you use to access the lb_node, traefik will forward to the right k8s service.
Now you need to edit the configuration:
nano traefik/traefik-daemonset.yaml
[acme] <-- set your data for auto certification
Create the dynamic proxy to be able to connect your service from the internet.
kubectl apply -f traefik <-- if error, probably because you didn't deploy other namespaces, so can ignore
kubectl get all --all-namespaces <-- if traefik pod can't get created, probably issue with port 443 on loadbalancer --> see troubleshooting section
Access services If set in traefik, please use login/pass: test/test You can use http or https
- kibana (logging): http://kibana.satoshi.tech
- grafana (monitoring): http:grafana.satoshi.tech (admin/admin)
- prometheus (monitoring): http://prometheus.satoshi.tech
- grafana2 (monitoring2): http://grafana2.satoshi.tech
- kubernetes-dashboard: http://kubernetes-dashboard.satoshi.tech
- traefik-console: http://traefik-console.satoshi.tech or http://any_minion_node_ip:30080
These lb nodes are some kind of DMZ servers where you could balance later your DNS queries. For production environment, I would recommend that only DMZ services (service-loadbalancer, traefik, nginx, ...) could run in here, because these servers will apply some less restrictive firewall rules (ex: open 80, 433, 5601, 3000) than other minion k8s nodes. So I would create a second security group (sg): k8s-dmz with same rules as k8s, and rules between both zone, so k8s services can talk to k8s and k8s-dmz. Then open 80, 433, 5601, 3000 for k8s-dmz only. Like this, k8s sg still protect more sensitive containers from direct public access/scans.
The same applies for the master node. I would create a new sg for it: k8s-master, so only this group will permit access from kubeclt (port 80, 443).
Then you should remove all NodePort from the services configuration, so no service will be available when scanning a classic minion. For that please comment the section "# type: NodePort" for all *-service.yaml
Add more loadbalancers, by adding more loadbalancers nodes. Because we use Daemonset type of job, all new nodes tagged with loadbalancer will spawn a loadbalancer container.
Use ansible to add a node
nano ansible/k8s.yml <-- edit: k8s_num_nodes: 3
ansible-playbook ansible/k8s.yml
kubeclt get node <-- wait for it!
Label it as a loadbalancer node
kubectl label node your_new_lb_minion_node role=loadbalancer
Then just check the new containers getting created kubectl get all --all-namespaces
For service-loadbalancer, try to access new_lb_minion_ip:5601
For trafik, add a dns A-record kibana.satoshi.tech --> new_lb_minion_ip so we will balance dns resolution to the old and new lb_node. Test some ping, and access kibana.satoshi.tech few times...
kubectl pilot k8s via the api server already on a secured port 443 in https. We will now create a certicate autority, to issue a certificate for the api, and for your admin client, to get even higher level of authentification.
The lines below are based on the work of Kelsey and the cloudflare easy CA. https://github.com/kelseyhightower/docker-kubernetes-tls-guide https://github.com/cloudflare/cfssl
Install your own PKI infra tool: cfssl You need GO 1.6+ and the GOPATH set
apt install golang-go
export GOPATH=$HOME/work
export PATH=$PATH:/usr/local/go/bin
export PATH=$PATH:$GOPATH/bin
go get -u github.com/cloudflare/cfssl/cmd/cfssl
go get -u github.com/cloudflare/cfssl/cmd/...
cd tls
mkdir kubectl master minion
Declare your master Ip (or domain) to the server cert
nano kube-apiserver-server-csr.json <-- add your master_ip in hosts section
Initialize a CA
cfssl gencert -initca ca-csr.json | cfssljson -bare ca/ca
Create an api server cert
cfssl gencert \
-ca=ca/ca.pem \
-ca-key=ca/ca-key.pem \
-config=ca-config.json \
-profile=server \
kube-apiserver-server-csr.json | cfssljson -bare master/kube-apiserver-server
cp ca/ca.pem master
Create kubeclt client cert
cfssl gencert \
-ca=ca/ca.pem \
-ca-key=ca/ca-key.pem \
-config=ca-config.json \
-profile=client \
kubernetes-admin-user.csr.json | cfssljson -bare kubectl/kubernetes-admin-user
kubectl config set-cluster secure --server=https://185.19.30.189:443 --certificate-authority=master/ca.pem --embed-certs=true
kubectl config set-credentials admin \
--client-key=kubectl/kubernetes-admin-user-key.pem \
--client-certificate=kubectl/kubernetes-admin-user.pem \
--embed-certs=true
kubectl config set-context secure --cluster=secure --user=admin
kubectl config use-context secure
Copy tls/master folder to node master
scp -r -i ~/.ssh/id_rsa_foobar master core@185.19.30.189:/home/core
Edit master
ssh -i ~/.ssh/id_rsa_foobar core@185.19.30.189
mkdir /etc/kubernetes
mv master/* /etc/kubernetes/.
sudo vim /etc/systemd/system/kube-apiserver.service
--client-ca-file=/etc/kubernetes/ca.pem \
--tls-cert-file=/etc/kubernetes/kube-api-server-server.pem \
--tls-private-key-file=/etc/kubernetes/kube-apiserver-server-key.pem \
sudo systemctl daemon-reload
sudo systemctl restart kube-apiserver.service
Note: this systemclt config will be erased at the next reboot. If you want to use cert, edit ansible master node template with the right config, then deploy a new master. Then generate certs...
Test
curl --cert kubectl/kubernetes-admin-user.pem --key kubectl/kubernetes-admin-user-key.pem --cacert master/ca.pem https://185.19.30.189/api -v
kubectl get node
Create kubelet client cert
cfssl gencert \
-ca=ca/ca.pem \
-ca-key=ca/ca-key.pem \
-config=ca-config.json \
-profile=client \
kubelet-client-csr.json | cfssljson -bare minion/kubelet/kubelet-client
Edit minion node
Copy and mv the file:
scp -r -i ~/.ssh/id_rsa_foobar minion core@185.19.30.189:/home/core
ssh -i ~/.ssh/id_rsa_foobar core@185.19.3.31
mkdir /etc/kubernetes
mv minion/* /etc/kubernetes/.
sudo vim /etc/systemd/system/kube-kubelet.service
--api-servers=https://185.19.30.189:443 \
--kubeconfig=/etc/kubernetes/kubelet/kubelet.kubeconfig \
sudo systemctl daemon-reload
sudo systemctl restart kube-apiserver.service
Check logs:
journalctl --since "10 minutes ago" -u kube-kubelet --no-pager
Try
kubectl get node
All services (kube-proxy, kube-client, kube-controller) can be set to use certificate. But this is a subject for another setup.
Probably your custom bash profile is interfering with the sourcing.
- Use kubectl completion bash > /etc/bash_completion.d/kubectl
- And make sure that lines are uncomment in
nano ~/.bashrc
if [ -f /etc/bash_completion ] && ! shopt -oq posix; then
. /etc/bash_completion
fi
Then logon and try again.
- manually on all node: fix an issue with hungry es v5
ssh -i ~/.ssh/id_rsa_foobar core@185.19.29.212
sudo sysctl -w vm.max_map_count=262144
- make it persistent:
sudo vi /etc/sysctl.d/elasticsearch.conf
vm.max_map_count=262144
sudo sysctl --system
- First see if ubuntu will be in the same namespace as the service you want to check:
nano utils/ubuntu.yaml
kubectl apply -f utils/ubuntu.yaml
- Depending in which namespace ubuntu runs, you can check services with one of these commands:
kubectl exec ubuntu -- curl elasticseach:9200 <-- should returns ... "cluster_name" : "elasticsearch"...
kubectl exec ubuntu -- curl kibana:5601 <-- should returns ... var defaultRoute = '/app/kibana'...
kubectl exec ubuntu -- curl elasticsearch.logging.svc.cluster.local:9200 <-- ubuntu in default namespace
kubectl exec ubuntu --namespace=logging -- nslookup elasticsearch <-- ubuntu in logging namespace
kubectl exec ubuntu --namespace=logging -- nslookup kubernetes.default.svc.cluster.local <-- ubuntu in logging namespace
- Check port 9200 on the node running elasticsearch container: ssh -i ~/.ssh/id_rsa_foobar core@185.19.29.212 netstat -plunt
- Uncomment type: NodePort and nodePort: 39200 if you want to access elasticsearch from any node_ip
- Check data in elasticsearch kubectl exec ubuntu -- curl es:9200/_search?q=* curl node_ip:39200/_search?q=* <-- if type: NodePort set in es.yaml
- check that there are file in node: ssh -i ~/.ssh/id_rsa_foobar core@185.19.29.212 ls /var/log/containers
Can't connect to k8s-dashboard addon:
- Carefull, addon are in kube-system namespace! If stuck use type: NodePort and
- Find the node public port: kubectl describe service kubernetes-dashboard --namespace=kube-system
- Access it from nodes : http://185.19.30.220:31224/
kubectl exec ubuntu -- nslookup google.com
kubectl exec ubuntu -- nslookup kubernetes
kubectl exec ubuntu -- nslookup kubernetes.default
kubectl exec ubuntu -- nslookup kubernetes-dashboard.kube-system
kubectl describe po/elastcisearch
kubectl logs -f elasticsearch-ret5zg
Possibly firewall issues! You need to open firewall internal rules between all nodes port 9100 (endpoint) and 10255 (node)
kubectl exec ubuntu --namespace=monitoring2 -- curl -sl -I influxdb:8086/ping
Not very sure yet what the issue on lb_node, but try to start traefik on another minion:
kubectl label node 185.19.29.192 role=minion <-- give another node the minion role
nano traefik/traefik-deployment.yaml <-- and edit section spec > nodeSelector: > role: minion
Then delete and recreate traefik, should be all good.
apt install httpie
http --verify=no --auth test:test https://kibana.satoshi.tech -v
alias k='kubectl'
alias kk='kubectl get all'
alias wk='watch kubectl get all'
alias ka='kubectl get all --all-namespaces'
alias kc='kubectl create -f'
alias kdel='kubectl delete -f'
alias kcdel='kubectl delete configmap'
alias kd='kubectl describe'
alias kl='kubectl logs'
Edit ansible-cloudstack/k8s.yml and run again the deploy
Delete the corresponding namespace, all related containers/services will be destroyed.
kubectl delete namespace monitoring
kubectl delete namespace logging
- Use different firewalls security group: k8s, k8s-dmz, k8s-master, to be ready for production
- Use persistent data for Elasticsearch and prometheus
- Fix prometheus k8s_pod scraping both port 80 and 9102...