A Small SaaS Team With Client Websites: Compose, Swarm, Nomad Or K8s?

architecturecontainerskubernetes31 replies · 718 views · started
  1. Were a small SaaS team running client websites. Today its Docker Compose on two VMs, deployed by a script 😂 A new hire is pushing hard for Kubernetes: self-healing, rolling updates, industry standard. i worry well spend all our time running teh platform instead of the product. Wheres teh tipping point??

    nobody on the team is a dedicated ops person... we all write code, and whoeevr is around handles teh servers... that's the context i'm trying to weigh... any war stories from teams of a similar size would be really helpful. im about to throw this thing out the window lol

    1
  2. yeahhh managed Kubernetes removes the control plane work, but you still own node upgardes, ingress, cert management, secrets, monitoring and the version upgrade treadmill lmao Budget real time for that every quarter.

    0 5 helpful
  3. spent since tuesday on client and im still not done thx in advance!!

    0 2 helpful
  4. been there man compose on a VM with a reverse proxy, health checks, restart: unless-stopped, and a simple deploy script covers a surprising amount. What it lacks is multi-host scheduling and zero-downtime deploys without extra work.

    4 1 helpful
  5. have u checked your kubernetes settings? mine were reset after an update. +1 same here fingers crossed 🤞

    0 0 helpful
  6. which vresion are u on?. have u checked ur websites settings?? mine were reset after an update ??

    5 0 helpful
  7. @heapfox what did u end up doing in the end

    1 2 helpful
  8. @heapfox thank u this worked!! 🤷 dumb question but how do i even check teh driver

    0 1 helpful
  9. oh also i come back to this thread every week just for teh drama, make a backup before u touch the extension, learned that teh hard way

    2 2 helpful
  10. hmmmm quick q any update on this??

    7 0 helpful
  11. guys pls help told my boss it would take 10 minutes... its been all weekend. @router_nomad366 did u ever fix it?? any ideas??

    2 1 helpful
  12. The tipping point is usually multiple services across multiple hosts that need to scale independently, or many teams deploying independently 👍 If a few internal tools fits on one or two machines, the benefits are small. worked for me anyway 🤷

    1 2 helpful
  13. ahh classic exact same thing happened to my mac last month worked for me anyway 🤷

    25 2 helpful
  14. @tensor_tinker works on paper, not in real life

    0 1 helpful
  15. oof following, having the same problem

    4 0 helpful

Reply to this topic

Sign in or create an account to reply.