I just cannot bring up GPU optimized instances on GCP. Tried couple of different types/zones but always get rejected Unavailable with the following explanation.
Does anybody else experienced this? Is it possible that the instance-types are not available whole day?
I tried to configure different timezones, increasing/decreasing the instance size and tried to use a VPN when accessing GCP.
Related
I am facing this issue from yesterday. This is the exact error: Failed to start feature-config: A e2-micro VM instance is currently unavailable in the us-central1-a zone. Alternatively, you can try your request again with a different VM hardware configuration or at a later time. For more information, see the troubleshooting documentation.
I had scheduled Google Compute Engine to TURN on & off at specific time using Instance scheduler but now I am locked out of it. I cannot even create a machine image to deploy on another zone
I changed the Machine Configuration. As from your answer I could figure out that resources might not be available for the US Central Zone possibly due to traffic. I changed configuration to - n2-highcpu-2 vCPU 2 & Memory -2 GB
At the end, it seems this was a general issue that multiple users experienced in us-central1 among other regions.
In this thread more is talked and it seems it got worse during the weekend.
As some suggestions in the comments, changing the zone/region/hardware can help but not always since this also depends on any constraints you may have.
As the error suggests, there aren't any available resources in that regions. I contacted GCP support after facing the same issue and got the following response:
Google Cloud Support, : Upon further checking, the reason that the e2-medium VM instance is currently unavailable is because there are limited VMs available to a specific zone and regions. Best we can do is to try another time or select a different zone so that the VMs that you desire to use will start. Rest assured that there is nothing wrong with your account and it was on the us-central1 zone who do not have available VMs you selected as of the moment.
If possible, try deploying to a different instance. For those who need an instance in us-central1 (for Qwiklabs?) might have to wait until more instances are available.
Similar issue here, but coming from a terraform apply. I've tried multiple zones and every one says both 'e2-small' and 'e2-micro' instances are unavailable. Seems google completely fumbled the "cloud game" here since AWS doesn't have this problem EVER! (not that I like using AWS, it's just "ick" compared to google).
I am trying to create a GPU instance (n1-standard-2 with 1 NVIDIA T4 GPU) on Compute Engine and I have been getting this error since yesterday:
Operation type [insert] failed with message "The zone 'projects/deep-learning-xxxx/zones/us-central1-a' does not have enough resources available to fulfill the request. Try a different zone, or try again later."
It seems that this region of Google Cloud doesn't have enough GPU resources, but I am getting the same error with other zones too, and after trying multiple times. Regular non-GPU instances are working fine though. I am trying to figure out if I'm doing something wrong or if there is a just a huge demand for GPU instances on GCP right now.
The reasons for GPU not being created on a VM in a particular region/zone can be,
1.Resource Unavailability. Check Resource availability here GPU availability across regions and zones.
2.Quota overuse can restrict the creation of GPUs. Refer Checking project quota for details.
3.Few GCP Restrictions, you can refer to the list of Restrictions here.
You can Check GPU Quota in Create VM with GPU's
Alternatively, GCP offers a feature called Reserving Compute Engine zonal resources to ensure that your project has resources for future use.
Finally, I was able to launch a preemptible GPU instance without a problem. So it really seems like Google Cloud doesn't have enough GPU resources to reserve an on-demand GPU VM at the moment.
I'm trying some stuff on Google Cloud and I have the following issue. Some days ago I created a Deep Learning VM with Compute Engine, with 8 vCPU and 1 Tesla K80 GPU. All worked fine, but now I want to try another GPU with different memory size. So, I deleted the VM instance (from Compute Engine -> VM instances) and I also deleted the deployment from Deployment manager. Nevertheless, when I try to create a new VM, I get an error message referring to the fact that I no more resources available and in fact, in the quotas page, I still see the GPU usage to 1 (with a limit of 1, that's why I can't create a new instance). Does anyone knows what could be the problem? Do I just have to wait? Thank you everyone!
If you receive a resource error (such as ZONE_RESOURCE_POOL_EXHAUSTED or ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS) when requesting new resources, it means that the zone cannot currently accommodate your request.
This error is due to the availability of Compute Engine resources in the zone, So, you could try to create the resources in another zone in the region or in another region.
You can search another available zone on this document: Available regions and zones
If possible, change the shape of the VM you are requesting. It's easier to get smaller machine types than larger ones. A change to your request, such as reducing the number of GPUs or using a custom VM with less memory or vCPUs, might allow your request to proceed.
Also, you can create reservations for Virtual Machine (VM) instances in a specific zone, using custom or predefined machine types, with or without additional GPUs or local SSDs, to ensure resources are available for your workloads when you need them.
Additionally, you can found more information to troubleshoot this issue in the following link
My VM in google cloud can't run due to below error has shown.
"Starting VM instance failed. Error: The zone does not have enough
resources available to fulfill the request. Try a different zone, or
try again later."
Then I what to start VM from other zones by changing the zone of my VM by this method but it's required VM running.
The problem is I can't run the VM. How can I use another solution?
It looks like Google's having issues with limited External IPs. Try removing the external IP before starting the instance. Then create an external IP and attach it your instance.
You’ve just encountered a stockout issue. A Stockout means that the particular GCP datacenter in that zone has reached its resource limit.
The Google Cloud Platform team are there to make sure that there are available resources in all zones. This type of issue is rare. When a situation like this occurs or is about to occur, the team is notified immediately, the issue is investigated and quickly fixed.
I recommend deploying and balancing your workload across multiple zones or regions to reduce the likelihood of a stockout. Please review the documentation which outlines how to build resilient and scalable architectures on the Google Cloud Platform.
You may also try again later, once resources will be available again in the region.
This being said, I see that you’ve posted your question on November 9th. That was a long time ago. Can you confirm if your issue is fixed now? It is very rare for stockouts to last this long.
I’m looking to use a cloud computing instance which will give me best value for my particular use case.
What I need to do is fire off a script periodically which performs some actions via selenium.
My questions are:
Will the packages installed be “remembered” if I reboot the instance?
Does the instance even need to be rebooted sometimes?
Do I get persistent storage or something else?
Am I charged when my instance is running but idle (between cron jobs)?
Any recommendations on which type of service would provide best value for my use case.
EDIT: I made some edits because I offended people by mentioning specific vendors and asking about pricing. The question probably could have been worded better initially but what I really wanted to know was which cloud computing solutions would be best value for a particular use case which seems to be a bit niche.
I'm with GCP Support team, so I will provide information on Google Cloud’s part.
Google Cloud Compute Engine offers a free usage for an amount up to a specific limit.
Will the packages installed be “remembered” if I reboot the instance? Do I get persistent storage or something else?
When you create a new VM instance you get assigned at least 10 Gb of persistent disk. Everything that is stored on that disk will stay there even in case of rebooting the instance.
Does the instance even need to be rebooted sometimes?
Instances doesn’t have to be rebooted unless you want so. However, occasionally they get rebooted automatically by the Google Compute Engine service.
Am I charged when my instance is running but idle (between cron jobs)?
You are charged for the instance when it is up and running. You can find more information about Google Cloud Compute pricing here.
To have an approximation of what you will be paying, based on what you will be using, you can use GCP Pricing Calculator.
After some investigation I found a couple of suitable options for running a selenium script a few times a day.
1) GCP, which as mentioned in another answer has a free tier.
2) AWS provides lambdas for running code without needing to provision servers. They also provide a free tier and I found some pre-compiled AWS lambda packages using Python + Selenium and some pretty clear instructions on how to get them running in AWS on a schedule: https://github.com/ryfeus/lambda-packs.