Git Product home page Git Product logo

Comments (6)

koderian avatar koderian commented on June 3, 2024 1

Also the script does not work when the preemptible cpu's quota is anything OTHER then 0. I requested 1, but the script stopped working, and it was the quota mismatch when I dug into the problem

from gcloudrig.

putty182 avatar putty182 commented on June 3, 2024

I'm not sure what to make of Google's statement there, since I'm literally doing this now to stretch out how long the one-time $300 free credits lasts.

It's likley they mean that preemptible instances don't take advantage of the "always free" tier limits, as listed here: https://cloud.google.com/free/docs/gcp-free-tier#always-free-usage-limits

If you're sure you don't want a preemptible instance, as a quick fix just remove this line:
https://github.com/putty182/gcloudrig/blob/master/globals.sh#L433

Will look at adding a preemptible option in the setup.sh prompts to do just this.

from gcloudrig.

koderian avatar koderian commented on June 3, 2024

Are you going to add an option in the setup.sh process to toggle preemptible instances? That would solve people's problems if the preemptible isn't working in there case, as it did in mines.

from gcloudrig.

putty182 avatar putty182 commented on June 3, 2024

originally from #50

Also, this is unrelated to this issue, but when setting up an instance using the tool, it only seems to start and set up correctly if I remove the --preemptible line in globals.sh. I have waited for a more than enough time for the preemptible instance to actually start running, but it always keeps restarting (and the setup log linked in the end of setup,sh doesn't actually run any of the scripts windows-start-up scripts / boot scripts) The instance just keeps restarting. My quota is good, and I've tried deleted the repo, destroying everything (destroy.sh) from my shell directory and cloning and running the whole setup again, with the same results, which sucks because I'd really want to take advantage of preempts.

from gcloudrig.

putty182 avatar putty182 commented on June 3, 2024

Paid billing is a requirement to use GPU instances; i.e. "upgrading" a free trial is required, but credits (if available) are still consumed first. Not sure when that kicked in, but that's the case now.

Since the goal of this project is to minimise costs, the default behaviour is still to use pre-emptible instances but it's now editable at the top of globals.sh (which will be documented one day!)

from gcloudrig.

SwiftWinds avatar SwiftWinds commented on June 3, 2024

For future visitors: I contacted GCP support, and indeed:

It's likley they mean that preemptible instances don't take advantage of the "always free" tier limits, as listed here: https://cloud.google.com/free/docs/gcp-free-tier#always-free-usage-limits

What @putty182 is saying here is correct. Preemptible GCP instances take advantage of GCP free credits, but they are not part of the GCP "always free" tier. Meaning that if you have free credits left, you are not charged; otherwise, you are guaranteed to be charged.

from gcloudrig.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.