Your browser does not seem to support JavaScript. As a result, your viewing experience will be diminished, and you may not be able to execute some actions.

Please download a browser that supports JavaScript, or enable it if it's disabled (i.e. NoScript).

k3s: The connection to the server localhost:8080 was refused - did you specify the right host or port?

morde last edited by

When I run any kubectl commands, I'm getting

The connection to the server localhost:8080 was refused - did you specify the right host or port?
Reply Quote 0
1 Reply Last reply
G
guilherme last edited by
This problem is likely caused by a bad ~/.kube/config perhaps you have a file from a different kubernetes install (minikube) or an older k3s. If the server is local you can fix this by running these commands,
```
mkdir ~/.kube
sudo k3s kubectl config view --raw | tee ~/.kube/config
chmod 600 ~/.kube/config
```
The contents of ~/.kube/config need to have the same information as /etc/rancher/k3s/k3s.yaml when the server was started (the keys, ip, and ports).

Note in order to tell k3s to use this config file https://devops.stackexchange.com/a/16044/18965 .
```
export KUBECONFIG=~/.kube/config
```
You should persist this by having it set in ~/.profile or ~/.bashrc
Reply Quote 0
1 Reply Last reply

2
Posts

0
Views

Suggested Topics

A

error validating data: ValidationError(Deployment): unknown field "spec" in
Continuous Integration and Delivery (CI, CD) • kubernetes yaml • • Avante

2

0
Votes

2
Posts

1
Views

E

Sure enough, everything and including the last spec: needed to be tabbed to the left once: apiVersion: apps/v1 kind: Deployment metadata: name: client-deployment spec: replicas: 3 selector: matchLabels: component: web template: metadata: labels: component: web spec: containers: - name: client image: ldco2016/multi-client ports: - containerPort: 3000 Unfortunately, you can't tell the difference with how yaml is formatted when pasted here, so it would not have been an easy catch, but if you would be looking at the original file configuration you would have noticed it like I did when I looked a third time.
Kubernetes: Custom second scheduler container unable to start
Continuous Integration and Delivery (CI, CD) • kubernetes • • morde

2

0
Votes

2
Posts

1
Views

K

In your manifest, you should set the spec.containers[0].livenessProbe.httpGet.scheme to HTTP not HTTPS. This is because you use the option --port=10251. As the documentation said, you should not use the deprecated --port option anymore, because it expose the API insecurely. You should set the following options: --port=0 --secure-port=10251 and then you would be able to set the spec.containers[0].livenessProbe.httpGet.scheme to HTTPS.
C

tools to gather variant information from thousands of servers and update into database
Continuous Integration and Delivery (CI, CD) • ansible monitoring systems • • courtlanda

2

0
Votes

2
Posts

1
Views

D

Doing something with collectd feeding into Prometheus or datadog is definitely a good idea for gathering metrics, but it doesn't help you with things like the kernel version. Handling process monitoring might be better handled with nagios or datadog as well. But for the bulk of what you're doing I'd say that it makes the most sense to stick with ansible. What you're doing sounds very similar to the ansible-cmdb project: Ansible-cmdb takes the output of Ansible's fact gathering and converts it into a static HTML overview page (and other things) containing system configuration information. It supports multiple types of output (html, csv, sql, etc) and extending information gathered by Ansible with custom data. For each host it also shows the groups, host variables, custom variables and machine-local facts. I've been using this for the last couple of years and it continues to evolve and improve. I hope you'll find it useful as well.
T

Newly installed k3s cluster on fresh OS install can not resolve external domains or connect to external resources?
Continuous Integration and Delivery (CI, CD) • dns k3s • • Trenton

2

0
Votes

2
Posts

0
Views

R

My problem was basically that I had multiple default routes, from ip route default via 172.16.42.1 dev ens5 proto dhcp src 172.16.42.135 metric 100 default via 172.16.42.1 dev ens3 proto dhcp src 172.16.42.95 metric 100 default via 10.2.64.1 dev ens4 proto dhcp src 10.2.67.51 metric 100 10.2.64.0/19 dev ens4 proto kernel scope link src 10.2.67.51 169.254.169.254 via 172.16.42.2 dev ens5 proto dhcp src 172.16.42.135 metric 100 169.254.169.254 via 172.16.42.2 dev ens3 proto dhcp src 172.16.42.95 metric 100 169.254.169.254 via 10.2.64.11 dev ens4 proto dhcp src 10.2.67.51 metric 100 172.16.42.0/24 dev ens5 proto kernel scope link src 172.16.42.135 172.16.42.0/24 dev ens3 proto kernel scope link src 172.16.42.95 The cause of this was not using no_gateway = true in my Terraform stanza, resource "openstack_networking_subnet_v2" "subnet_project" { name = "subnet_project" network_id = openstack_networking_network_v2.net_project.id cidr = "172.16.42.0/24" ip_version = 4 } Without no_gateway = true. I can fix this by lowering the metric on the default route on the host, sudo ip route replace default via 10.2.64.1 dev ens4 metric 90 Which will add a new route to this, default via 10.2.64.1 dev ens4 metric 90 Now running nslookup google.com as in the question will work fine, I can re-break it with sudo ip route del default via 10.2.64.1 dev ens4 metric 90 Other diagnostics These are taken before bringing up k3s. My ip -o addr shows, 1: lo inet 127.0.0.1/8 scope host lo\ valid_lft forever preferred_lft forever 1: lo inet6 ::1/128 scope host \ valid_lft forever preferred_lft forever 2: ens3 inet 172.16.42.95/24 brd 172.16.42.255 scope global dynamic ens3\ valid_lft 40534sec preferred_lft 40534sec 2: ens3 inet6 fe80::f816:3eff:fecd:6722/64 scope link \ valid_lft forever preferred_lft forever 3: ens4 inet 10.2.67.51/19 brd 10.2.95.255 scope global dynamic ens4\ valid_lft 24333sec preferred_lft 24333sec 3: ens4 inet6 2620:0:28a4:4140:f816:3eff:fed2:72c7/64 scope global dynamic mngtmpaddr noprefixroute \ valid_lft 2591979sec preferred_lft 604779sec 3: ens4 inet6 fe80::f816:3eff:fed2:72c7/64 scope link \ valid_lft forever preferred_lft forever 4: ens5 inet 172.16.42.135/24 brd 172.16.42.255 scope global dynamic ens5\ valid_lft 40534sec preferred_lft 40534sec 4: ens5 inet6 fe80::f816:3eff:fe3d:cde8/64 scope link \ valid_lft forever preferred_lft forever And ip link shows 1: lo: mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: ens3: mtu 1492 qdisc fq_codel state UP mode DEFAULT group default qlen 1000 link/ether fa:16:3e:cd:67:22 brd ff:ff:ff:ff:ff:ff 3: ens4: mtu 1500 qdisc fq_codel state UP mode DEFAULT group default qlen 1000 link/ether fa:16:3e:d2:72:c7 brd ff:ff:ff:ff:ff:ff 4: ens5: mtu 1492 qdisc fq_codel state UP mode DEFAULT group default qlen 1000 link/ether fa:16:3e:3d:cd:e8 brd ff:ff:ff:ff:ff:ff
O

Kubernetes show secrets for namespace
Security, Penetration Testing • kubernetes documentation • • Oba22

2

0
Votes

2
Posts

1
Views

S

Found the solution. Searched at the wrong place: -n is part of the core kubectl command: https://kubernetes.io/docs/reference/kubectl/kubectl/ -n, --namespace string If present, the namespace scope for this CLI request So it shows the results for a given namespace if applied.
R

How do I know the name pods?
Software Programming • kubernetes • • Raziyah00

2

0
Votes

2
Posts

0
Views

N

On the team. port-forward Please. namespace Your subway:kubectl -n main_sp port-forward main-ms-hc-78469b74c-7lfdh 8080:80
K

What is the difference between using brackets (`[` `]`) and just specifying the command in the CMD option in Dockerfiles?
Continuous Integration and Delivery (CI, CD) • dockerfile • • kalyn

2

0
Votes

2
Posts

1
Views

B

CMD with brackets is contained within an array. Without brackets, the command is executed as /bin/sh -c mysqld. Best practice afaik is to always use brackets.
How to connect to postgres container from a node container using docker-compose?
Continuous Integration and Delivery (CI, CD) • postgresql docker networking • • jeanid

2

0
Votes

2
Posts

1
Views

K

docker-compose contains an internal DNS. In the example the service names are also the DNS names. In this case, the DNS of the database is db. If one configure this as a hostname in the node app then one should be able to access the database. Back in the day, the wordpress example of docker-compose helped me understanding this. When I changed WORDPRESS_DB_HOST: db:3306 to WORDPRESS_DB_HOST: db2:3306, the wordpress was not able to access the mysql db anymore as depicted in the log. When it was changed back and forth to db and to db2 it was not able to access it and with db it was. Then I realized that the service names are actually added to an internal DNS that can be resolved by other services that are defined in the docker-compose.yml. Recently I realized that running docker-compose up and specifying multiple files using -f also ensures that all services get added to the internal DNS. Initially I thought that the resolving was restricted to the same docker-compose.yml, but probably the docker network is responsible. If multiple apps reside in the same network, in this case the default, then they can resolve each other.
C

Running services on different ports in development environment and production environment
Continuous Integration and Delivery (CI, CD) • ports microservices • • chanisef

2

0
Votes

2
Posts

1
Views

F

This should be fine. It's something you may even have to do if ports are already in use on localhost. Even development languages like .net core are now supporting different environments that can be used to configure the different ports. Example: 1433 is the standard sql server port. If I wanted to test a local sql server container, but did not want to stop my locally installed sql server, I could run it on another port (e.g. 1431).
Helm Repo: can it host dockerfiles and other image-building artifacts
Continuous Integration and Delivery (CI, CD) • docker images kubernetes containers helm • • Marcee

2

0
Votes

2
Posts

1
Views

E

Well that's a legit question IMHO, but there are multiple way of doing things (and no one is worth than other). 1- Everything from dev to prod at the same place If you want to grab everything from dev to production in a single project, it's possible, but you cannot build the docker image from hell command. Helm is only here to organize, use template in yaml and sync releases following kubernetes file structure. You can follows helm repository structure and add other scripts, but you have to execute steps in the right order: build, push, deploy 2- The "regular" way to do I more often saw a separation of production configuration files and application files. What I do is: One repository which carry my application source code + everything for the CI (test, build, docker file) One repository which carry my helm.charts for all my infrastructure. Choose what suits the best for you, but keep in mind that helm is a templating system for kubernetes yaml files, so it will not build the docker image.
E

Connect external Jenkins Master to GKE with Kubernetes plugin
Continuous Integration and Delivery (CI, CD) • jenkins kubernetes cicd • • emerson

2

0
Votes

2
Posts

1
Views

You could install Jenkins using helm to install the k8s plugin as well. https://github.com/helm/charts/tree/master/stable/jenkins
E

Gitlab CI/CD: deploy docker image to existing kubernetes cluster
Continuous Integration and Delivery (CI, CD) • docker kubernetes amazon web services gitlab continuous integration • • esther

2

0
Votes

2
Posts

1
Views

A

Your error is most likely from referencing the wrong directory in your script block of your job. When GitLab executes your job in a container, it mounts the project directory to the /builds directory in the container. You can reference this either by specifying /builds/scheduler-deployment.yaml or by using the predefined environment variable $CI_PROJECT_DIR with $CI_PROJECT_DIR/scheduler-deployment.yaml.
Z

Startup of ECS Fargate task fails with ResourceInitializationError mount.nfs4 Connection timed out
Continuous Integration and Delivery (CI, CD) • amazon ecs fargate amazon efs • • zymir

2

0
Votes

2
Posts

1
Views

D

Underlying reason: The ECS service's network access security group did not have permission to access EFS. Solution: Add an inbound rule for type NFS in the security group as described in this tutorial: https://docs.aws.amazon.com/AmazonECS/latest/developerguide/tutorial-efs-volumes.html#efs-security-group
C

Files with variables are evil?
Software Programming • kubernetes continuous integration web based programming ci cd cross cutting • • chanisef

2

0
Votes

2
Posts

0
Views

D

The answer to this question is based on: https://12factor.net/config I'd rather read it. https://12factor.net https://en.wikipedia.org/wiki/Twelve-Factor_App_methodology https://habr.com/ru/post/261171/ (Configuration is the properties of the implementation environment). This is the most important phrase for this reply. The text can be foundNow to the details:As soon as the configuration is the nature of the environment (i.e. OS), it is the variables that must come from it.The attachment must be inferential, who and how the variables put it. I mean, the actual application doesn't know anything about the configuration files and should not know.Approach with .env-Files, we're going to bury at the assembly stage, and we're gonna have to decide that we're loaded, whether it's a configuration or a variable environment.If you have any https://www.vaultproject.io/ and https://www.consul.io/ the variables will be rolled out as variable environments.There's a loader in the code that won't be able to test normally because it's not a library, it's a founic that's too limited. What to do with him is not clear. It's a potential vulnerability. Plus additional code. Plus a spot for the bagos that will be hard to catch.In general, such an approach could be. But if we have a separate library that handles the loading of variables, configurations and everything. Yes, it can produce different types of loads: different types of configurations, variables, etc. This library may be responsible for several types of media: files, configurations, env-Files, variable environments, etc.On the mark. There's a great one. https://plugins.jetbrains.com/plugin/7861-envfile for Goland. Plagin can download variables
How to update pods if version hasn't changed?
Continuous Integration and Delivery (CI, CD) • docker kubernetes deployment • • irl

2

0
Votes

2
Posts

1
Views

Y

The main problem here is, that from a Kubernetes point-of-view nothing has changed to your deployment. So nothing gets updated. The best solution here would be, that you add something to your image tag, like the git short hash, for example. The your update will just work. Also you will make your deployments re-creatable, when you do it this way. If you really want to keep it the way it is, you could add some other information to the deployment, that changes on each deployment, but is not used, like an annotation. But I would not recommend that, since I saw this leading to strange behavior with old pods no being shutdown. On a side note: The imagePullPolicy is just evaluated, when there is change to your deployment. So basically this flag says: If you do deploy something (e.g. bc there is a new deployment), no matter if you have the image already locally, reload it from the registry.

k3s: The connection to the server localhost:8080 was refused - did you specify the right host or port?

Suggested Topics

error validating data: ValidationError(Deployment): unknown field "spec" in Continuous Integration and Delivery (CI, CD) • kubernetes yaml • • Avante

Kubernetes: Custom second scheduler container unable to start Continuous Integration and Delivery (CI, CD) • kubernetes • • morde

tools to gather variant information from thousands of servers and update into database Continuous Integration and Delivery (CI, CD) • ansible monitoring systems • • courtlanda

Newly installed k3s cluster on fresh OS install can not resolve external domains or connect to external resources? Continuous Integration and Delivery (CI, CD) • dns k3s • • Trenton

Kubernetes show secrets for namespace Security, Penetration Testing • kubernetes documentation • • Oba22

How do I know the name pods? Software Programming • kubernetes • • Raziyah00

What is the difference between using brackets (`[` `]`) and just specifying the command in the CMD option in Dockerfiles? Continuous Integration and Delivery (CI, CD) • dockerfile • • kalyn

How to connect to postgres container from a node container using docker-compose? Continuous Integration and Delivery (CI, CD) • postgresql docker networking • • jeanid

Running services on different ports in development environment and production environment Continuous Integration and Delivery (CI, CD) • ports microservices • • chanisef

Helm Repo: can it host dockerfiles and other image-building artifacts Continuous Integration and Delivery (CI, CD) • docker images kubernetes containers helm • • Marcee

Connect external Jenkins Master to GKE with Kubernetes plugin Continuous Integration and Delivery (CI, CD) • jenkins kubernetes cicd • • emerson

Gitlab CI/CD: deploy docker image to existing kubernetes cluster Continuous Integration and Delivery (CI, CD) • docker kubernetes amazon web services gitlab continuous integration • • esther

Startup of ECS Fargate task fails with ResourceInitializationError mount.nfs4 Connection timed out Continuous Integration and Delivery (CI, CD) • amazon ecs fargate amazon efs • • zymir

Files with variables are evil? Software Programming • kubernetes continuous integration web based programming ci cd cross cutting • • chanisef

How to update pods if version hasn't changed? Continuous Integration and Delivery (CI, CD) • docker kubernetes deployment • • irl

error validating data: ValidationError(Deployment): unknown field "spec" in
Continuous Integration and Delivery (CI, CD) • kubernetes yaml • • Avante

Kubernetes: Custom second scheduler container unable to start
Continuous Integration and Delivery (CI, CD) • kubernetes • • morde

tools to gather variant information from thousands of servers and update into database
Continuous Integration and Delivery (CI, CD) • ansible monitoring systems • • courtlanda

Newly installed k3s cluster on fresh OS install can not resolve external domains or connect to external resources?
Continuous Integration and Delivery (CI, CD) • dns k3s • • Trenton

Kubernetes show secrets for namespace
Security, Penetration Testing • kubernetes documentation • • Oba22

How do I know the name pods?
Software Programming • kubernetes • • Raziyah00

What is the difference between using brackets (`[` `]`) and just specifying the command in the CMD option in Dockerfiles?
Continuous Integration and Delivery (CI, CD) • dockerfile • • kalyn

How to connect to postgres container from a node container using docker-compose?
Continuous Integration and Delivery (CI, CD) • postgresql docker networking • • jeanid

Running services on different ports in development environment and production environment
Continuous Integration and Delivery (CI, CD) • ports microservices • • chanisef

Helm Repo: can it host dockerfiles and other image-building artifacts
Continuous Integration and Delivery (CI, CD) • docker images kubernetes containers helm • • Marcee

Connect external Jenkins Master to GKE with Kubernetes plugin
Continuous Integration and Delivery (CI, CD) • jenkins kubernetes cicd • • emerson

Gitlab CI/CD: deploy docker image to existing kubernetes cluster
Continuous Integration and Delivery (CI, CD) • docker kubernetes amazon web services gitlab continuous integration • • esther

Startup of ECS Fargate task fails with ResourceInitializationError mount.nfs4 Connection timed out
Continuous Integration and Delivery (CI, CD) • amazon ecs fargate amazon efs • • zymir

Files with variables are evil?
Software Programming • kubernetes continuous integration web based programming ci cd cross cutting • • chanisef

How to update pods if version hasn't changed?
Continuous Integration and Delivery (CI, CD) • docker kubernetes deployment • • irl