Last 12 weeks · 0 commits
2 of 6 standards met
Hi, I’m wondering if you experience the same issue I have. If two of the 3 pods get killed and recreated they won’t be able to rejoin the cluster and the UI won’t respond. I checked the logs and all 3 pods trying to elect a leader using the old IPs of the two dead pods instead of the new pod IPs. I spent days trying to figure out why. If you could provide any assistance I would greatly appreciate it. Thanks.
Given that the Google Cloud Free Trial only offers 8 CPUs in a single region, it's impossible to run the entire cluster with 2-CPU machines with a new account. I changed the machine types from to to reduce CPUs from 2 to 1 for each machine, and halved the resources in the files to match. I did have to add a fourth node to the in the nomad cluster to account for all running Nomad containers. I didn't dig too deep to figure out if there's a better layout, but given that all the services can run with reduced resources, I think this is a good start. Please let me know if you have any issues with my proposed changes here. Having limited experience with k8s and virtually none with Vault, Consul, and Nomad, I'm eager to learn more.
Repository: kelseyhightower/nomad-on-kubernetes. Description: Tutorial on running Nomad on Kubernetes. Stars: 355, Forks: 49. Primary language: Shell. Languages: Shell (66.6%), Go (20.9%), HCL (12.6%). License: Apache-2.0. Open PRs: 1, open issues: 1. Last activity: 8y ago. Community health: 42%. Top contributors: kelseyhightower, ianmiell, mehzer.