Debug School: prashant

Learning in Kubernetes

prashant — Fri, 11 Nov 2022 05:57:36 +0000

ETCD-Kubernetes cluster uses etcd as its back up plan for data.
Controller Manager:-responsible for onboarding new nodes to the cluster handling situations where nodes become unavailable or get destroyed to keep our application running
API Server:-determines if a request is valid and then processes it
Schedule:- a control plane process which assigns Pods to Nodes
POD- Pod is not a physical entity. It just like isolate containers.
Configmap:- store key value pair.
Secret:- store key value in encoded form
PV:-have a lifecycle independent of any individual Pod that uses the PV
PVC:-PersistentVolumeClaim is a request for storage by a user
Job:- run the process once
Cronjob:- run the process in repeated interval

How Kubernetes works? All components and roles of their components

prashant — Wed, 09 Nov 2022 07:44:50 +0000

k8s cluster contains master and workers node. Master nodes distributes the workloads to the worker nodes and Worker nodes hosts the containers/pods.

Master Node components:
apiserver: Api server is used to establish any communication with the k8s cluster through REST APIs. Any input to k8s cluster is received in JSONs or .yml.
Cluster storage: Its used to store all configuratons and state information of the cluster. It uses etcd to store the data. apiserver will connect with cluster store to fetch these informations.
Kube Contoller manager- It runs in infinite loop and watches for any changes in the cluster. Status is collected and shared with apiserver thus ensuring apiserver has the latest status. Each component in the cluster has a separate controller. Eg: Node, Endpoint, Namespace etc.
Kube scheduler - It watches apiserver for any new pods and schedules them in the available nodes.

Worker Nodes components:
Kubelet: kublet is an agent deployed in worker nodes and communicates with master node through apiserver. Kubelet instantiates pods in the nodes. Any nodes to be added to cluster requires kubelet to be setup.
Container Engine: Container engine is used for Container management like pulling images and starting and stopping containers.
Kube Proxy: Kube-proxy handles all networking requirements including load balancing across all pods, POD IP addresses.

What is POD?

prashant — Wed, 09 Nov 2022 07:36:24 +0000

atomic unit of scheduling.
containers always run inside the pod.
pods is an virtual object.
pods can have multiple containers and each container have different port.
all the container in a pod share the pod environment.
if a pod contain single container known as loose coupling.
if a pod contain multiple container known as tight coupling.
we cant restart or stop the pods.
pod life-cycle contains pending, running, succeeded/failed phase.
inter pod communication can be done using pod network.
intra pod communication is done using pod ip address.
Ephemeral - no Pod is ever “redeployed”
Atomicity - they’re there or NOT

Why we need Kubernetes?

prashant — Wed, 09 Nov 2022 07:23:12 +0000

high deployment speed.
ability to absorb change quickly.
ability to recover quickly.
hide complexity in the cluster.
Automated rollouts and rollbacks.
service discovery mechanism.
load balancing.
Storage orchestration.
Self-healing

What is Kubernetes?

prashant — Wed, 09 Nov 2022 07:19:16 +0000

container orchestrator.
workload placement.
infrastructure abstraction.
desired state.
open-source system for automating deployment.
management of containerized applications.
scaling of containerized applications.
It groups containers that make up an application into logical units for easy management and discovery.

diff between docker pause and docker unpause?

prashant — Mon, 07 Nov 2022 11:16:21 +0000

Docker pause cmd:- suspends all processes in the specified container that means allocated cpu to that process is 0%.

Docker unpause cmd:- un-suspends all processes in the specified containers. On Linux, it does this using the freezer cgroup.

[root@localhost CONTAINER ID IMAGE c91a0f994f47 httpd 66d23993e0e4 httpd c213671a8288 httpd 708a7964a45f httpd b21cb4ba98f8 httpd a5151e54aa53 httpd [root@localhost c213671a8288 [root@localhost ~]# docker ps -a CONTAINER ID IMAGE c91a0f994f47 httpd 66d23993e0e4 httpd c213671a8288 httpd 708a7964a45f httpd b21cb4ba98f8 httpd a5151e54aa53 httpd [root@localhost ~]# docker stats CONTAINER ID NAME c213671a8288 [root@localhost c213671a8288 [root@localhost ~]# docker ps CONTAINER ID IMAGE c213671a8288 httpd 708a7964a45f httpd b21cb4ba98f8 httpd [root@localhost ~]# docker stats CONTAINER ID NAME c213671a8288 708a7964a45f romantic_kepler b21cb4ba98f8 raj2 ~]# docker ps -a
COMMAND CREATED STATUS PORTS NAMES
"httpd-foreground" 21 minutes ago Exited (137) 12 minutes ago kind_meitner
"httpd-foreground" 22 minutes ago Exited (0) 11 minutes ago clever_leavitt
"httpd-foreground" 22 minutes ago Up 22 minutes 0.0.0.0:80->80/tcp, :::80- >80/tcp admiring_faraday
"httpd-foreground" 23 minutes ago Up 23 minutes 80/tcp romantic_kepler
"httpd-foreground" About an hour ago Up 50 minutes 80/tcp raj2
"httpd-foreground" About an hour ago Created objective_austin
~]# docker pause c213671a8288
COMMAND CREATED STATUS PORTS NAMES
"httpd-foreground" 22 minutes ago Exited (137) 12 minutes ago kind_meitner
"httpd-foreground" 23 minutes ago Exited (0) 12 minutes ago clever_leavitt
"httpd-foreground" 23 minutes ago Up 23 minutes (Paused) 0.0.0.0:80->80/tcp, :::80- >80/tcp admiring_faraday
"httpd-foreground" 23 minutes ago Up 23 minutes 80/tcp romantic_kepler
"httpd-foreground" About an hour ago Up 51 minutes 80/tcp raj2
"httpd-foreground" About an hour ago Created objective_austin
CPU % MEM USAGE / LIMIT MEM % NET I/O BLOCK I/O PIDS
admiring_faraday 0.00% 8.781MiB / 3.701GiB 0.23% 656B / 0B 0B / 0B 82
~]# docker unpause c213671a8288
COMMAND CREATED STATUS PORTS NAME S
"httpd-foreground" 23 minutes ago Up 23 minutes 0.0.0.0:80->80/tcp, :::80->80/tcp admi ring_faraday
"httpd-foreground" 24 minutes ago Up 24 minutes 80/tcp roma ntic_kepler
"httpd-foreground" About an hour ago Up 51 minutes 80/tcp raj2
CPU % MEM USAGE / LIMIT MEM % NET I/O BLOCK I/O PIDS
admiring_faraday 0.00% 8.781MiB / 3.701GiB 0.23% 656B / 0B 0B / 0B 82
0.00% 8.777MiB / 3.701GiB 0.23% 656B / 0B 0B / 0B 82
0.00%