For Question#19 Taints and Tolerations doesn't really guarantee that a POD with Toleration will GET SCHEDULED on a specific Node with Taint, these pods can still get scheduled on different Nodes without any Taints. Taints just repels the pods without the tolerations from being scheduled on the specific Node. So, you would also need to use Node Affinity along with Taints and Tolerations to ensure that a pod or group of pods get scheduled on required Nodes. Alternatively, there are options like "nodeSelector" which can directly be used into the Pod definition to schedule a pod onto a specified Node under the "nodeSelector".
K8s provides a way to share volumes between containers and persist data outside of the containers as well (7:30 mentioned little wrong). This is achieved using the concept of Persistent Volumes (PV) and Persistent Volume Claims (PVC).
Question no. 19 If we want to use taint and toleration to deploy specific pod in specific node then we must taint all node. Otherwise the pod may be deploy on that node who has no taint. Best practice for that situation is use of node affinity or node selector. Is that right?
Rolling updates that kubernetes has is different from rolling back of change that the speaker is referring to. Rolling update just means if there are 10 replicas of a deployment not all 10 will be shut down and replaced with new version at the same time. Instead it's done in batches of a few replicas at a time so that there is no downtime.
@helloworld If you wish to learn Kubernetes from top experts and get certification, then kindly call our course advisors on IND: +91-7022374614 US: 1-800-216-8930 (Toll-Free).
You can share a storage volume with multiple containers in multiple pods in K8’s with rwx or rom access modes, not sure what you mean? and the controller manager is not a daemon set its a static pod on the master. do your own research people
Kubelet is managing the state of each node and ensuring that the containers running on the node are healthy. ---------- Kube-Proxy is managing network communications between the pods (groups of one or more containers) running on the node and the other pods in the cluster.
don't we have any alternative to get back our information from unexpectedly deleted pod instead of persistent volume. because PV is used by only one pod which is missing High availability or failover service by other running cluster pods to take care?
Unfortunately, without using a persistent volume, there's no other direct way to retrieve data from a deleted pod, impacting high availability and failover capabilities.
Hi thank you so much, I have an Interview question 1. How many pods, master nodes, Worker nodes are there in your project? Please answer this question approximatetely thanks in advance
For Question#19 Taints and Tolerations doesn't really guarantee that a POD with Toleration will GET SCHEDULED on a specific Node with Taint, these pods can still get scheduled on different Nodes without any Taints. Taints just repels the pods without the tolerations from being scheduled on the specific Node. So, you would also need to use Node Affinity along with Taints and Tolerations to ensure that a pod or group of pods get scheduled on required Nodes.
Alternatively, there are options like "nodeSelector" which can directly be used into the Pod definition to schedule a pod onto a specified Node under the "nodeSelector".
K8s provides a way to share volumes between containers and persist data outside of the containers as well (7:30 mentioned little wrong). This is achieved using the concept of Persistent Volumes (PV) and Persistent Volume Claims (PVC).
Do like, share & subscribe to our channel to get updates on upcoming videos.
a pod can be scheduled on the specific node using nodeSelector as-well
Yes, you can assign a pod to a particular node using nodeSelector too.
Question no. 19
If we want to use taint and toleration to deploy specific pod in specific node then we must taint all node. Otherwise the pod may be deploy on that node who has no taint.
Best practice for that situation is use of node affinity or node selector.
Is that right?
@dnyaneshwar shiral We appreciate your active participation😊. Subscribe to Intellipaat to know the right answer.
@@Intellipaat I know what is answer, still I subscribed this channel
i am so happy for these questions... finally i got answers to some of my die hearted questions
Glad to hear that! Keep learning with Intellipaat😊
Rolling updates that kubernetes has is different from rolling back of change that the speaker is referring to. Rolling update just means if there are 10 replicas of a deployment not all 10 will be shut down and replaced with new version at the same time. Instead it's done in batches of a few replicas at a time so that there is no downtime.
@helloworld If you wish to learn Kubernetes from top experts and get certification, then kindly call our course advisors on IND: +91-7022374614 US: 1-800-216-8930 (Toll-Free).
Thanks a lot Prakhar for this video
@Kenneth Antony John Most welcome 😊 Do like, share & subscribe to our channel to get updates on upcoming videos.
Good content, but very fast explanation, for the freshers or who is new to kubernetes is difficult to understand that much fast
Glad you liked it 😊 Do like, share & subscribe to our channel to get updates on upcoming videos.
Can you make video on setup and monitoring metrics that need to be considered for prometheus and grafana
Okay Rishabh, Stay tuned to Intellipaat and press the bell icon so that you will not miss that video from us.😊
You can share a storage volume with multiple containers in multiple pods in K8’s with rwx or rom access modes, not sure what you mean? and the controller manager is not a daemon set its a static pod on the master. do your own research people
For personal support, you can reach out to our course advisors on IND: +91-7022374614 US: 1-800-216-8930 (Toll-Free).
I have a doubt ?
Is kubelet enables the communication between master and slave nodes ? Or the kube proxy?
If you wish to get Personal support Kindly call our course advisors on IND: +91-7022374614 US: 1-800-216-8930 (Toll-Free).
Kubelet is managing the state of each node and ensuring that the containers running on the node are healthy.
----------
Kube-Proxy is managing network communications between the pods (groups of one or more containers) running on the node and the other pods in the cluster.
Useful. Thank you !!
Glad it was helpful! @Yashwant Pawar Do like, share & subscribe to our channel to get updates on upcoming videos.
Good video quite knowledgeful
Glad you think so! @Rohan Sharma Do like, share & subscribe to our channel to get updates on upcoming videos.
don't we have any alternative to get back our information from unexpectedly deleted pod instead of persistent volume. because PV is used by only one pod which is missing High availability or failover service by other running cluster pods to take care?
Unfortunately, without using a persistent volume, there's no other direct way to retrieve data from a deleted pod, impacting high availability and failover capabilities.
So this is for Kubernetes admin right?
yes
Cool😊
Very helpful.
Glad it was helpful. Keep learning with Intellipaat😊
Nice
Thank you😊
Hi thank you so much, I have an Interview question
1. How many pods, master nodes, Worker nodes are there in your project?
Please answer this question approximatetely thanks in advance
@AVINASH K Do like, share & subscribe to our channel to get updates on upcoming videos.😊
Very useful
Thanks a lot
anyone faced these questions?
@Rajesh Gubbala Do like, share & subscribe to our channel to get updates on upcoming videos.
Sorry if you are experienced don't expect these questions ...these are for fresher's those who know just k8s
These sudden microphone switches are triggering, I jump every time - hate it.
Sorry for the inconvenience caused. We'll work on it.