Replies posted by michaelhaigh

4 Replies

T
ThewanandonlyExplorer
asked in Nutanix Kubernetes Engine

Playing with Karbon

Hi Thewanandonly, Please see this Kubernetes documentation for more information: [url=https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/]https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/[/url] In summary, you have 3 options: [list=1] [*]Place the kubeconfig file in ~/.kube/config [*]Set it as an environment variable ($ export KUBECONFIG=karbon-kubectl.cfg) [*]Specify it each time you run a kubectl command ($ kubectl --kubeconfig karbon-kubectl.cfg get pods) [/list]

2
M
Nutanix Employee
4 years ago
G
guifAdventurer
asked in Self-Service

Error when install an Application Blueprint

Hi Guif! First off, apologies about the lengthy delay. Generally that sort of error indicates the lack of a DNS server set on the CentOS machine. As a test, you can add a nameserver to /etc/resolv.conf to see if it resolves your issue. Longer term, DNS configuration is best handled during IP assignment, whether that's DHCP, AHV IPAM, or static.

1
M
Nutanix Employee
4 years ago
aluciani
alucianiChevalier
published in Community Blog

Creating a CI/CD Pipeline with Nutanix Karbon and JenkinsBlog

Hi Kazi!Are you asking about the agents that perform the pipeline builds?  If so, they’re spun up dynamically for each build.  If you (or a group of users) creates multiple builds at once, multiple agents will spin up.If that’s not what you’re asking about, can you please clarify?Thanks,Michael Haigh

44375
K
4 years ago
aluciani
alucianiChevalier
published in Community Blog

Creating a CI/CD Pipeline with Nutanix Karbon and JenkinsBlog

Thanks for the clarification Kazi.  So one option would be to change the agent.podRetention field to “Always” -- that way the agents stay alive after a successful build:https://github.com/helm/charts/blob/master/stable/jenkins/values.yaml#L379Another option would be to change the agent.idleMinutes field to a non-zero value -- that way the agents are not destroyed immediately.  However they will eventually be destroyed, unlike the previous setting.https://github.com/helm/charts/blob/master/stable/jenkins/values.yaml#L412For the maximum agent guardrail, you can set the agent.containerCap value.  By default this value is 10, so if your previous example of 2-10 is accurate, then you can leave it to the default.https://github.com/helm/charts/blob/master/stable/jenkins/values.yaml#L407All of these values should be set in the jenkins-config.yaml file that is described at the beginning of this blog.Regards,Michael Haigh

44375
K
4 years ago

Badges

Show all badges
Terms & Conditions