Compute Engine

Welcome to the Google Compute Engine idea forum. You can submit and vote on ideas here to tell the Google Compute Engine team which features you’d like to see.

This forum is for feature suggestions, and is visited regularly by Google’s Product Managers. It is not actively monitored by our technical support team. If you’re looking for help forums, look here:

If you are looking to file a bug report for which you do not need 1:1 technical support, you can also visit our Public Issue Tracker

We can’t wait to hear from you!

  1. Support HTTP(S) Internal Load Balancers

    Currently internal load balancing only works with TCP/UDP. HTTP Internal Load Balancing would simplify our QA env setup.

    33 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, our networking team has this on their roadmap. This item will be updated as it nears planning and implementation/release.

  2. Disable firewall rules

    Now we can only delete firewall rules. It would be great if we could disable/enable firewall rules as well.

    45 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    8 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, this is a good idea. Our team is looking at making this enhancement.

  3. Make pricing by region easier to understand

    Use the same names for regions on the pricing page (https://cloud.google.com/compute/pricing) as you do when looking at instances.

    Display the actual hourly pricing, not including discounts when you select or edit a product.

    9 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  4. Increase IP address allocation limit

    On AWS I can assign 12 external IPs to a t2.large. I would like to exceed that limit with GCP. Even If I paid per IP I wouldn't care.

    30 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    8 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, we’re considering ways this might be increased. Currently you can only assign one external IP per network device, although we have recently added the ability to use multiple network devices. 12 would be a lot though just to get multiple IPs. Please stay tuned for further developments.

  5. Add "User" to each host in ~/ssh/config with `gcloud compute config-ssh` command

    Connections to the VM with OpenSSH only succeed if the correct username is given. Otherwise it returns "Permission denied (publickey)".
    Some external tools, e.g. Ansible, use the SSH config file resp. the SSH agent.

    Adding the "User" to each "Host" entry would be help here. Then it's also not required to add it somewhere in the tool config (if possible at all) and therefore these configs can easier be shared without user state in it.

    11 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Flag idea as inappropriate…  ·  Admin →
  6. Set scheduled autoscaling for instance groups bases on time for scale up and down?

    Should be a possibility to autoscale based on scheduled time in conjunction with other metrics. This is because, for example, I'm expecting several events during the month on specific days during specific range hours.

    38 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    11 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    We have generally not found scheduled autoscaling to be necessary in Google Cloud, unlike some other clouds. The technology behind our Load Balancers is the same as Google uses on all its sites, and we don’t need to do these types of tricks.

    We do get this request from time to time, but most customers find it is not needed, and several large web hosters have scrapped their own custom scheduling and pre-spinning approaches when migrating to GCP. Our instance provisioning and boot times are low enough that they work well for many use cases, and App Engine in particular is well suited to very fast and high scaling.

    The one thing we can’t affect would be if your particular application server requires a long time to configure itself at instance startup time. The best way to deal with that, at present, would be to try not to do…

  7. add "Shutdown script (Optional)" field in addResourceForm on Google Cloud Console

    In Google Cloud Console
    In Create an instance template and Create an instance (addResourceForm)
    in Automation section

    add "Shutdown script (Optional)" text area like "Startup script (Optional)"

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  8. Clarify the use of IAM roles and GCE access scopes in the inline help

    Please add the fact that you need to provide both IAM roles and GCE Access scopes to grant the right permissions to a Service Account to the inline help texts in the Cloud Console.

    You need to set both the right role AND access scope. It literaly took us literally days to find this.

    This is the particular part that cleared things up: https://cloud.google.com/compute/docs/access/service-accounts?hl=en_US#usingroles

    This would have helped to have inline or directly referenced to in the inline help text:

    "The level of access a service account has is determined by a combination of access scopes and IAM roles so…

    5 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  9. GUI

    Vm instances needs a simple GUI PRE-INSTALLED

    5 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    We can only imagine that you mean you want a desktop user interface installed inside the guest OS, for example X windows or some desktop windows manager inside a Linux VM? Because Cloud IaaS is largely used to provide servers, this is not the default configuration of any of our pre-provided Linux images. Our suggestion currently would be that you create a VM, install the specific desktop packages you would like and configure them just they way you want, and then create your own disk Image from that VM’s disk. You can then use that Image to start any additional VMs you’d like over time.

    Note that Windows of course comes with a desktop environment that you can use via RDS for administrative and testing purposes (or you may use RDS as a workstation for one or more users, if you properly license it).

  10. Specify swap space size in "Create an instance"

    Currently, after a new VM instance is created, the swap space size is 0 and, if desired, we have to add swap space post-creation. It would be nice to be able to enter a desired swap space size in the instance creation form and it would be ready and active when the instance becomes available. Note: I'm mostly using CentOS images.

    23 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    4 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks for the suggestion! Swap space is not a very common use case, so far as we know, and it might not rise to the level of something that makes sense to burden the API with for all time, but we’ll take a look.

  11. 3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks for your suggestion. This would likely be quite difficult technically, but we’ll evaluate it. In the meantime, an effective way to do this would be to create a snapshot of the disk and create a new disk of the other type based from that snapshot.

  12. Committed Use Discounts for shared CPU instance

    It seems that Committed Use Discounts is only available for at least 1vCPU. But I also want to have Committed Use Discounts for shared CPU instance like g1-small or f1-micro.

    24 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    6 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks for your suggestion, this is definitely something we will consider adding in the future!

  13. Move files from GCE to Drive and back.

    I wish there was a way to easily move files from my compute instance to my Google Drive account. That way I could more easily transfer files between my compute instances rather than having to zip/host/download/upload/unhost/delete.

    52 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    5 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, we’ll consider this. Drive is more of a consumer oriented product, but there are cases where we see users wanting to do this.

    The common way to share files between machines would be GCS, have you investigated that? There are a number of solutions like gcsfuse that let you treat a bucket like a network drive, too. This should eliminate most of the steps, just upload to bucket, download from bucket.

  14. Please add 'update date' column to see when the instance has been updated

    Please add 'update date' column to see when the instance has been updated

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, we’ll consider this. In the meantime however, please clarify via comments what your expectations would be for what “updated” means. Do you mean the last time you changed a setting? Last time you used our Updater feature to start the VM? Or, the last time the (OS specific) internal autoupdating might have installed packages, like from yum or apt-get?

  15. Add ability to "force" shutdown/deletion of instance

    When rolling large instance groups, the wait time between the ACPI shutdown signal and the actual removal of the instance from the group is significant (and accrues quickly when performing staggered rollouts).

    It would be great if instances could be "force-killed", since a clean shutdown isn't a requirement in many cases.

    14 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, that’s a very valid point – it mixes with another recent item about selecting which instances to take down. The ideal goal would be to quickly takedown any idle instances that don’t need clean shutdown but wait for clean shutdown (completion of request processing) for any busy instances. I’m sharing this feedback with the team for consideration.

  16. Provide Tests for Health Check Configurations

    Health Checks presently are disconnected from the instance, but in no dialogues connected to either the instance or the healhcheck do you have the ability to say "test the healthcheck" the way you can with uptime monitors in Stackdriver.

    It would be particularly nice if there were a way to do this.

    2 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Flag idea as inappropriate…  ·  Admin →
  17. By Default enable Compute Api for billed projects

    The compute API should be enabled by default if billing is enabled for a project it is giving me error every time I try to start it

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, this is something we’ve been evaluating (not all customers want this behavior). Regarding the error you mention, please open a case with support if you have a support agreement with us, or post to the gce-discussion Google Group to contact someone who can provide interactive support (we’re not staffed to tackle individual issues here on UserVoice). Thx!

  18. Ability to permanently delete projects

    As of now projects deleted from console go to shutdown state and stay there for a while. Also there is a limit of 5 projects per user. It would be nice to be able to permanently delete a project by the user instead of waiting.

    38 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    10 comments  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    Thanks, we’re looking at how we can improve this. As customers develop a history of usage and payments over time, limits like project count and other quotas can be raised.

  19. When a rolling update of an instance group is performed, the new instance should not receive any traffic until the health check passes.

    When a rolling update of an instance group is performed, the new instance should not receive any traffic until the health check passes.

    4 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
    under review  ·  Paul Nash responded

    The product team thought this should already be the case, so we’re looking into it, and will be reaching out directly.

  20. add tar functionality to gsutil

    tar functionality is highly appreciated when you've uploaded a tar.gz compressed raw VM image for 6 hours and then get the error message "ERROR: The file inside the tar archive was named 'something-other-than-disk.raw'. It should be named 'disk.raw'. The documentation for importing VirtualBox images doesn't say that using the name 'disk.raw' is required. So now I have to rename the contents of the tar.gz container locally and wait again for 6 hours until the upload has finished again.

    14 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  • Don't see your idea?

Compute Engine

Categories

Feedback and Knowledge Base