Questions tagged [kubernetes]
For questions about Kubernetes, an open-source system for automating deployment, scaling, and management of containerized applications.
908 questions
0
votes
0
answers
15
views
Kibana expose through K8s Gateway API
I have a python app running and I already managed to expose it through Gateway API and access it through "http:///pythonapp".Next I deployed a Elasticsearch and Kibana instance so I can gain ...
0
votes
0
answers
11
views
Do I still need watchdog built-in feature if pgpool ii is already deployed inside a kubernetes cluster?
I have a kubernetes cluster with three nodes, and two nodes DB (Master/ & Slave) that outside of the cluster.
Pgpool ii is deployed inside the cluster with kind of Deployment with replicas:3.
So, ...
0
votes
0
answers
10
views
issues with cross-cluster Vault Agent Injector setup
We have vault running in one kubernetes cluster. It is running behind an ingress controller with tls enables with a signed certificate by sectigo.
kubernetes auth method is enabled and configured with ...
0
votes
0
answers
11
views
+50
504 Gateway timout for all new requests at specific minute each hour in Kubernetes cluster
We're running our own Kubernetes cluster and it works well for a while but at minute :03 each hour, public web services stop responding (504 gateway timeout) and some commands such as kubectl top ...
0
votes
0
answers
6
views
Authentication with github when using sonatype nexus in CI builds only for NodeJS
We have a NodeJS project in our kubernetes cluster. We are looking at using Sonatype Nexus(self hosted) for caching packages. One package is a private package that we host on github npm registry. We ...
0
votes
0
answers
23
views
How to change the ingress port of Ubuntu Canonical k8s
I have a pretty bare bone snap install k8s on Ubuntu. Just longhorn is up and running. I would like the cluster not to listen to port 80, but 8080. This should be fairly easy, with the most straight ...
0
votes
0
answers
18
views
Kubernetes: Long-lived TCP connections dropping through kube-proxy / Service LoadBalancer. Direct node connection better?
I’m building a service that needs to maintain a very large number of long-lived TCP connections (persistent sockets). Low latency and stability are essential.
We’re running on a managed Kubernetes ...
0
votes
1
answer
11
views
Do all CNI allow the host to directly communicate with the service and pod networks?
This may seem a bit of an odd question but I've been going over the theory of it and I'm still not sure. From what I understand of Linux theory, a host machine accesses the pod network via a virtual ...
0
votes
1
answer
22
views
Why is my etcd didn't tolerate single node failure with 3 etcd members?
I installed k3s on three nodes:
GCP (control-plane, etcd, master)
AWS (control-plane, etcd, master)
LXC (dedicated etcd only installed from doc of k3s, but it didn’t register in the Kubernetes ...
0
votes
0
answers
10
views
unable to get kong ingress request level metrics in kubernetes
Kong Prometheus plugin only exposes node / nginx metrics — no request-level metrics (status codes, latency, RPS)so that i can see use them in grafana dashboard.
helm values used
Kubernetes cluster.
...
0
votes
0
answers
33
views
Error when deploying SQL Server on k3d using helm charts
I'm trying to deploy SQL Server on k3d using a Helm chart.
I ran the following command:
$ helm install mssql-latest-deploy . --set ACCEPT_EULA.value=Y --set MSSQL_PID.value=Developer
But I’m getting ...
0
votes
1
answer
22
views
How to persist coreDNS custom hosts in k8s(LKE)
How to persist CoreDNS custom host entries in Linode Kubernetes Engine (LKE) so they don’t get reset?
I’m running a cluster on Linode Kubernetes Engine (LKE) (Kubernetes v1.34).
I want some of my ...
0
votes
0
answers
11
views
canary upgrade of hybrid openshift cluster using custom mcp
I am working on canary upgrade of openshift cluster.
my cluster is a 3 node hybrid, where each node act as a worker and master.
NAME STATUS ROLES AGE ...
0
votes
0
answers
13
views
Kubeflow Behind Istio Returns 403 / 503 at Ingress, Dex & OIDC Gatekeeper Behaving Differently Across Two Identical Installations
I am troubleshooting an odd behavior between two Kubeflow installations that were both deployed using Juju charms. One installation is on-premise and working correctly, while the second one (hosted on ...
0
votes
0
answers
19
views
GCP GKE GatewayAPI Client Authentication (serverTlsPolicy)
I use GCP, GKE and GatewayAPI. I created Gateway resources in order to create an Application Load Balancer in GCP in order to get my applications (which are in an Istio mesh) exposed to the world.
...