kubegres VS HomeLab

A good comparison with other operators here [2022]: https://blog.palark.com/cloudnativepg-and-other-kubernetes-o...

Personally I've only used Kubegres (https://www.kubegres.io/), which didn't even make the above list. It's ok for a personal project.

All k8s solutions for postgres take subtly different approaches. It seems that they've all converged on the Operator pattern. The basics are easy: run a database process which persists data to a cloud disk of your choice. The hard parts are how to update, migrate, backup, restore, monitor, failover, replicate, etc. These kubernetes "operators" promise to fulfill the role of a DBA but, just like hiring a DBA, it requires buy-in to their approach.

Reactive-Tech / Kubegres (opérateur)

I like https://www.kubegres.io, it takes a little getting used to how it works. I probably wouldn't use it for a huge database, but for things under a few TB it works great.

I don't know if you're using Kubernetes, but Kubegres is great for self hosting postgres on a cluster.

Any experience with https://www.kubegres.io/ ?

How does this compare with something like kubegres?

https://www.kubegres.io/

https://www.kubegres.io/ has worked well for me. You can use custom images, provided they follow the conventions of the official postgres image. I haven't tried a major-version upgrade yet - my impression is that it will probably be a bit painful and involve some downtime and manual ops.

Some of the more advanced operators might better handle the upgrade problem? But in general, I've been happy with Kubegres for a ~10GB database with low traffic.

There are some great operators out there that make deployment and management easy. I use kubegres managed with pgadmin and it has been working well for me.

reactive-tech/kubegres (operator)

And if you are interested how I've done it: https://github.com/Michaelpalacce/HomeLab

Example of a service that is being backed up (see backup.velero.io/backup-volume): https://github.com/Michaelpalacce/HomeLab/blob/master/Helm/apps/uptimekuma/templates/deployment.yaml

But for me https://github.com/Michaelpalacce/HomeLab/tree/master/cluster/homelab/apps

So I've been trying to get botkube to work, but it seems to me like the helm chart documentation is outdated? I've been looking at the values and trying to make things work and it didn't really work for me.. https://github.com/Michaelpalacce/HomeLab/blob/master/cluster/homelab/apps/botkube/helmrelease.yaml is what I tried

And here is some documentation and specifics I've ran into. https://github.com/Michaelpalacce/HomeLab/blob/master/docs/Backups.md

Also for configuration storage and git, I recommend taking a gitops approach to things. You can checkup FluxCD2. Here is my repo for reference https://github.com/Michaelpalacce/HomeLab

A few suggestions and thinks I wish I knew when I started ( personal opinions incoming ): 1. Get a single well sized server ( I'd go for 8 cpus .. And like more than 16 gigs of ram and 250+ ssd, ideally m.2 nvme ). Don't bother with raspberry pis.. Check my profile for my previous post to see a comparison of raspberry pi cluster and a x86 cluster... 2. Install proxmox to do virtualization 3. The majority of People in this sub hate kubernetes and the minority is scared to speak, but install k3s on the vms you spin up, this way you can scale resources and not worry about some elaborate setup of which containers go to which node and how to setup tls offloading, load balancing and much much much more as kubernetes will handle it for you. ( bonus since you are a devops engineer so you are probably used to kube clusters ) 4. Use Velero to do restic backups to s3 for services that are critical like dbs and password managers etc. 5. Do everything with ansible or GitOps! My personal homelab is here: https://github.com/Michaelpalacce/HomeLab utilizing ansible, flux for GitOps and renovate to keep my services up to date ( note to say my cluster spin up isn't as nice as I want it to be since when provisioning I have a Dependency on some secrets which I do manually.. But this is only first cluster setup ) 6. Setup a VPN and administrative services on the raspberry pi you have lying around. Personally my VPN is on the router but if yours doesn't support it, do that. 7. After you are done setting everything up and you have a backup, format the drives and reinstall the vms and re set everything up. Document the process, write automation scripts until satisfied... I genuinely recommend you do this like 3 times...... 8. Vpn I guess a lot of people like wireguard or adguard home... Doesn't really matter as long as it's not open vpn and it supports multiple threads. 9. Setup nodered for some automation around your new home, you'll love it. 10. Firefly is amazing! 11. Make sure that your CSI ( Container storage interface ) supports replication. Keep 3 replicas of critical volumes! 11. If you need any help, don't forget to ask :) personally shoot me a dm and I'll do my best to help, but I'm sure most of us here would