[chaosblade: node CPU load, node network delay, node network packet loss, node domain name access exception]

node CPU load

The goal is ： Specify a node , do CPU load 80% experiment

node_cpu_load.yaml Content ：

apiVersion: chaosblade.io/v1alpha1
kind: ChaosBlade
metadata:
  name: cpu-load
spec:
  experiments:
  - scope: node
    target: cpu
    action: fullload
    desc: "increase node cpu load by names"
    matchers:
    - name: names
      value:
      - "docker20"
    - name: cpu-percent
      value:
      - "80"

 Select a node , modify  node_cpu_load.yaml  Medium  names  value .
 Carry out orders ：
$ kubectl apply -f node_cpu_load.yaml

 Check the status 
 perform  kubectl get blade cpu-load -o json  command , Check the experiment status .

 View results 
 Access to the  Node  node , Use  top  Command can see this node  CPU  Achieve the desired effect .

 stop it 
 Carry out orders ：kubectl delete -f node_cpu_load.yaml
 Or delete  blade  resources ：kubectl delete blade cpu-load

 Node network related scenarios 
 Before operation , Please log in first  node  node , Use  ifconfig  Command to view network card information , Not all system default network card names are  eth0

Node network delay scenario

The goal is ： Specify the local of the node 32436 Port add 3000 Millisecond access delay , The delay time fluctuates up and down 1000 millisecond

 Select a node , modify  delay_node_network_by_names.yaml  Medium  names  value 
 Yes  docker20  Node local port  32436  Access packet loss rate  100%.

delay_node_network_by_names.yaml  Content ：
apiVersion: chaosblade.io/v1alpha1
kind: ChaosBlade
metadata:
  name: delay-node-network-by-names
spec:
  experiments:
  - scope: node
    target: network
    action: delay
    desc: "delay node network loss"
    matchers:
    - name: names
      value: ["docker20"]
    - name: interface
      value: ["ens33"]
    - name: local-port
      value: ["32436"]
    - name: time
      value: ["3000"]
    - name: offset
      value: ["1000"]

 Carry out orders ：
$ kubectl apply -f delay_node_network_by_names.yaml

 Check the status 
 perform  kubectl get blade delay-node-network-by-names -o json  command , Check the experiment status .

 View results 
#  Access... From the experiment node  Guestbook
$ time echo "" | telnet 192.168.1.129 32436
Trying 192.168.1.129...
Connected to 192.168.1.129.
Escape character is '^]'.
Connection closed by foreign host.
echo ""  0.00s user 0.00s system 35% cpu 0.003 total
telnet 192.168.1.129 32436  0.01s user 0.00s system 0% cpu 3.248 total

 stop it 
 Carry out orders ：kubectl delete -f delay_node_network_by_names.yaml
 Or delete  blade  resources ：kubectl delete blade delay-node-network-by-names

Node network packet loss scenario

The goal is ： Of the specified node 32436 Packet loss rate of port injection 100% Failure of

 Select a node , modify  loss_node_network_by_names.yaml  Medium  names  value .

loss_node_network_by_names.yaml  Content ：
apiVersion: chaosblade.io/v1alpha1
kind: ChaosBlade
metadata:
  name: loss-node-network-by-names
spec:
  experiments:
  - scope: node
    target: network
    action: loss
    desc: "node network loss"
    matchers:
    - name: names
      value: ["docker20"]
    - name: percent
      value: ["100"]
    - name: interface
      value: ["ens33"]
    - name: local-port
      value: ["32436"]

 Carry out orders , Start ：
$ kubectl apply -f loss_node_network_by_names.yaml

 Check the status 
 perform  kubectl get blade loss-node-network-by-names -o json  command , Check the status .

 View results 
 The port is  Guestbook nodeport  The port of , No response when accessing the experimental port , However, the port that accesses the unopened experiment can be used normally ：

#  Access to the node  IP
$ kubectl get node -o wide
NAME       STATUS   ROLES                      AGE     VERSION   INTERNAL-IP     EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION       CONTAINER-RUNTIME
docker20   Ready    worker                     3d16h   v1.17.6   192.168.1.129   <none>        Ubuntu 18.04.4 LTS   4.15.0-101-generic   docker://19.3.11
kk         Ready    controlplane,etcd,worker   4d16h   v1.17.6   192.168.4.210   <none>        Ubuntu 18.04.4 LTS   4.15.0-101-generic   docker://19.3.11
#  Access from the operation node  Guestbook -  cannot access 
$ telnet 192.168.1.129 32436
Trying 192.168.1.129...
telnet: connect to address 192.168.1.129: Operation timed out
telnet: Unable to connect to remote host
#  Access from non operational nodes  Guestbook -  Normal visit 
$ telnet 192.168.4.210 32436
Trying 192.168.4.210...
Connected to 192.168.4.210.
Escape character is '^]'.

 You can also access address verification directly from the browser .

 stop it 
 Carry out orders ：kubectl delete -f loss_node_network_by_names.yaml
 Or delete  blade  resources ：kubectl delete blade loss-node-network-by-names

Node domain name access exception scenario

The goal is ： This experiment is modified Node Of hosts, Tamper with domain name address mapping , simulation Pod Abnormal scenario of internal domain name access .

 Select a node , modify  dns_node_network_by_names.yaml  Medium  names  value .

dns_node_network_by_names.yaml  Content ：

apiVersion: chaosblade.io/v1alpha1
kind: ChaosBlade
metadata:
  name: dns-node-network-by-names
spec:
  experiments:
  - scope: node
    target: network
    action: dns
    desc: "dns node network by names"
    matchers:
    - name: names
      value:
      - "docker20"
    - name: domain
      value: ["www.baidu.com"]
    - name: ip
      value: ["10.0.0.1"]

 Carry out orders , Start the experiment ：
$ kubectl apply -f dns_node_network_by_names.yaml

 Check the experiment status 
 perform  kubectl get blade dns-node-network-by-names -o json  command , Check the experiment status .

 View experiment results 
#  Enter the experiment  node
$ ssh [email protected]
# Ping www.baidu.com
$ ping www.baidu.com
#  No response 
 You can see  Node  Of  /etc/hosts  The document was modified , Simulated  dns  Analyze abnormal scenarios .

 stop it 
 Carry out orders ：kubectl delete -f dns_node_network_by_names.yaml
 Or delete  blade  resources ：kubectl delete blade dns-node-network-by-names