Not enough resources to fulfill for the past 14 hours

[np]

Hello, 

I'm just wondering if anyone has been able to resolve the problem "/zones/us-west1-b' does not have enough resources available to fulfill the request. Try a different zone, or try again later." if it persists for a long period of time. I've been trying to access my VM this weekend with no luck. 

I've also tried to save a snapshot and create a new VM instance elsewhere, but had problems that it couldn't be migrated? 

Hoping to get some insight on when resources will be available again or what to do in the mean time.

Thanks! 

[Digil (Google Cloud Platform Support)]

Hello,

I deployed a basic GCE VM in us-west1-b without any issue. At this time, it is difficult to tell what could have happened within your project's infrastructure. However, I would highly recommend you to re-deploy the workload in different location where the resources are currently available. 

A similar concern is stated and discussed in this thread. 

[np]

Thanks for your response. I'm having trouble when attempting to spin up the VM with a GPU. It doesn't matter which GPU or which zone; it just doesn't work. I understand this might be a resource issue and this has happened before momentarily but never for such a prolonged period of time. 

Just wondering what I can do at this point -- trying to make any reservations will return the same error and migration or creating a new instance in any other zone is not working for me. As far as I can tell I haven't changed anything about the project or the VM since my last successful deployment. 

[Digil (Google Cloud Platform Support)]

Before deploying an instance with GPU, you need to make sure that particular GPU model is available(you can check the availability here) in the specified zone. 

As I mentioned before, you need try different zone considering the current availability of specific GPU model in that zone.  

[np]

I am deploying with the NVIDIA Tesla K80. It should be available in all of the zones I am trying to deploy with. I also have tried multiple zones, just to create a new instance in case it was something on my current VM with no luck. 

Is there any known outage with GPUs in the US, Europe, and Asia zones at the moment? It seems like I'm not the only one with this problem. 

[Digil (Google Cloud Platform Support)]

As per the help center article, NVIDIA Tesla K80 is currently available in the following zones.

=======

asia-east1-a

asia-east1-b

europe-west1-b

europe-west1-d

us-central1-a

us-central1-c

us-east1-c

us-east1-d

us-west1-b

======= 

If the issue keep persisting for the deployment of NVIDIA Tesla K80 in any of the above mentioned zones, I would strongly recommend you to report it using issue tracker to get additional view from Google Cloud's Compute Engine team. 

Additional note: If you are having a business impact connected with this concern, then contacting the Google Cloud Platform's sales team is also a recommendable approach. 

[Kirill Katsnelson]

On Monday, May 11, 2020 at 2:07:46 PM UTC-7, np wrote:

Thanks for your response. I'm having trouble when attempting to spin up the VM with a GPU.

X-ref: https://groups.google.com/forum/#!topic/gce-discussion/34zBBmTV8Tg

Looks like there is a shortage.

 -kkm