Resources unavailable in us-east1-b

[Matt Miller]

Even though when I look at the Zones in my console it says us-east1-b is available:

When I try to start one of my VMs in that zone, it tells me there aren't enough resources:

Like most people here, I need to be able to get the data stored on my VMs or GCE is about completely worthless to me and the money I've been billed for creating the data on those VMs was basically thrown away. When will this issue be resolved?

Regards,

Matt L. Miller

[Justin Reiners]

I don't know specifics, but are the VM's not starting preemptible?

That's about the only time I see this error. I have around 100 VM's on GCE, only one is preemptible.

--
© 2018 Google Inc. 1600 Amphitheatre Parkway, Mountain View, CA 94043
 
Email preferences: You received this email because you signed up for the Google Compute Engine Discussion Google Group (gce-dis...@googlegroups.com) to participate in discussions with other members of the Google Compute Engine community and the Google Compute Engine Team.
---
You received this message because you are subscribed to the Google Groups "gce-discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gce-discussio...@googlegroups.com.
To post to this group, send email to gce-dis...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gce-discussion/48b65379-c49f-4cd7-a006-379189856c66%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jason]

Hi,

The error message means that resources (vCPU, Memory, PDs, etc...) are exhausted at that time within that specific zone. When this issue occurs, a team is notified and are working to resolve the issue; however, I cannot provide an ETA on when the issue will be resolved. If you need to start a VM instance right away, then you can create the VM instance in any other zone (us-east1-c or us-east1-d) or wait until the team adds more resources to the specified zone (us-east1-b in your use case).

If you are using a Preemptible VM instance, then this can occur more frequently as they might not always be available. I have attached a link to our documentation in regards to Preemptible instance limitations to the bottom of this message [1].

[1] https://cloud.google.com/compute/docs/instances/preemptible#limitations

[Matt Miller]

None of my VMs are preemptible.

On Thursday, May 9, 2019 at 9:53:57 AM UTC-7, Justin Reiners wrote:

I don't know specifics, but are the VM's not starting preemptible?

That's about the only time I see this error. I have around 100 VM's on GCE, only one is preemptible.

On Thu, May 9, 2019 at 11:45 AM Matt Miller <matt....@gmail.com> wrote:

Even though when I look at the Zones in my console it says us-east1-b is available:

When I try to start one of my VMs in that zone, it tells me there aren't enough resources:

Like most people here, I need to be able to get the data stored on my VMs or GCE is about completely worthless to me and the money I've been billed for creating the data on those VMs was basically thrown away. When will this issue be resolved?

Regards,

Matt L. Miller

--
© 2018 Google Inc. 1600 Amphitheatre Parkway, Mountain View, CA 94043
 
Email preferences: You received this email because you signed up for the Google Compute Engine Discussion Google Group (gce-dis...@googlegroups.com) to participate in discussions with other members of the Google Compute Engine community and the Google Compute Engine Team.
---
You received this message because you are subscribed to the Google Groups "gce-discussion" group.

To unsubscribe from this group and stop receiving emails from it, send an email to gce-dis...@googlegroups.com.

[Matt R]

Yes, I'm having the same issue today and yesterday.   All I'm trying to do is spin up a 2core8GB VM. Seems a little crazy... Mine are not preemptible, but they are running on free google credit. 

[Matt Miller]

Unfortunately, the question is not whether I need VMs running, but that I need those specific VMs running to get the data on them. I found a workaround ... snapshotted the disks and created new instances in another zone that I could spin up and then download the data. But it definitely seems like we should know if zones are available or not -- the Zone list itself said east1-b was available at the time that all of us couldn't spin anything up.

I am now able to start VMs in that zone, but overall the experience has been pretty suboptimal. I think (a) transparency into what zones are actually operable, (b) some insight into how often each zone has problems like this, and (c) faster recovery when this happens would be something you would want to provide in a production-level site. Like, I had no guarantee when spawning from my snapshots that the zone I was creating VMs in would actually work (it did, but it looked like east1 should have, too).

Regards,

Matt L. Miller

[Matt Miller]

Mine were running on a full paid account. I actually never had problems like this when I was on free credit, so to have it happen within a few days of actually putting real money on the line was pretty disappointing, to say the least.

[Jason]

Hi,

To be clear, this issue does not have anything to do with the type of account (Paid or Free Tier) a user has. 

I understand that this type of issue can be frustrating; however, keep in mind that GCP continues to strive to provide the most uptime possible for its users. 

[Justin Reiners]

Good to know Jason thanks, preemptive as well?

--

© 2018 Google Inc. 1600 Amphitheatre Parkway, Mountain View, CA 94043
 
Email preferences: You received this email because you signed up for the Google Compute Engine Discussion Google Group (gce-dis...@googlegroups.com) to participate in discussions with other members of the Google Compute Engine community and the Google Compute Engine Team.
---
You received this message because you are subscribed to the Google Groups "gce-discussion" group.

To unsubscribe from this group and stop receiving emails from it, send an email to gce-discussio...@googlegroups.com.

To post to this group, send email to gce-dis...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/gce-discussion/8df87196-caab-493f-b08d-fa930c478784%40googlegroups.com.

[Ryan Waldron]

Well, I am kind of surprised that a company like Google can't be bothered to rack enough servers to maintain their cloud to the point that you can start an instance. As a senior engineer with a SaaS company, I was exploring using GCE as our company was looking to diversify from AWS to a multi-cloud infrastructure. It's also perfectly clear from a search using google that this is far from an isolated incident. I can tell you what....after this I will never recommend GCE to my own company or any other. I'm not willing to let your extremely poor business practices become ours that is for sure.

On Thursday, May 9, 2019 at 12:45:25 PM UTC-4, Matt Miller wrote:

[Kyle Chen]

Just want to let all my friends in US know that you are not alone. Same situation here in Hong Kong (asia-east2-a), and even more unfortunate that I haven't been able to start my VM instance since TWO WEEKS ago. I tried all the 3 zones in Hong Kong but I cannot even create an instance to migrate.

To Google, indeed we are not asking for a copy-and-paste reply that your team is looking into it. What we need is your concrete plan, timeline and commitment of resolving this and getting our VM instance back online. I don't want to use AWS to be honest, but you leave me no choice if the situation continues. Please, timeline, plan and commitment.

Ryan Waldron於 2019年5月11日星期六 UTC+8上午11時30分35秒寫道：