Cloud SQL instances are not starting or restarting. It's stuck

[Shaharia Azam]

I was just trying to restart my Google Cloud SQL instances from command line. But it got stuck since 2 hours. I don't know why.

And from gcloud console I can't do any operation for my Cloud SQL instances. It's saying ERROR: (gcloud.sql.instances.patch) HTTPError 409: The instance or operation is not in an appropriate state to handle the request.

So is there anything I can perform a fresh restart? Even I couldn't get in touch with any technical support channel rather than here.

[Adam (Cloud Platform Support)]

This may be related to an ongoing issue. If you send me a private message with your full instance connection name I can confirm for you.

[Bulama Yusuf]

@Adam

I tried connecting to my cloud sql instance and it refused to connect. I stopped it and it refused to start when I asked it to start. The I tried to restart and it has been stuck in restart ever since.

PROJECT ID: ra-climate-change

CLOUD SQL INSTANCE ID: ra-cc-data

As expected with data, this is a very critical situation.

Thanks in advance.

Bulama.

[Adam (Cloud Platform Support)]

@Bulama The instance was restarted on 03/17. In order to do so the activation policy had to be set back to ALWAYS. Please confirm if you do want to keep this instance running. Also, can you let us know how the restart was issued (eg. via gcloud or Cloud Console)?

[Pragatheesh Kumar]

Hi Adam,

Similar issue here. The instance was suspended due to billing. Once after the billing is cleared out, the instance is still in the suspended state and I am not able to manually restart the service.

Project ID: piyestra-collections-app-test

Project number: 826791900584

[Bulama Yusuf]

Thanks @Adam. I have taken a back up of the database. I had started another instance so I would not want to keep this anymore, so I will shut it down myself. 

I restarted the instance from the Cloud Console but that was after I had stopped it and it failed to start (all on the Cloud Console).

How can I set the activation policy of my current and future instances to ALWAYS in order to avoid this from happening again?

Thanks.

[Adam (Cloud Platform Support)]

@Pragatheesh Your issue is not related to those described in the thread. The instance was suspended due to billing. If the billing account state has been resolved there can be some propagation delay before the service is reactivated.

If you still need to resolve your billing issue you may file a billing support request.

[zhen ping wong]

@Adam, I am facing the similar issue for a few days. Had emailed to cloud-sql since few days ago, but no reply yet.

The stuck I'm facing is not only few hours, it is to days' extend.

I thought using cloud sql will be hassle free, compare to hosting by myself in a compute engine.

Desperately need help, really appreciate it.

Thank you very much.

[Alex (Cloud Platform Support)]

@Bulama, the Activation Policy of a 2nd Generation Cloud SQL instance is normally set to “ALWAYS” upon being started from a stopped state or upon creation. Alternatively, you can make sure of this by calling [gcloud sql instances patch [INSTANCE_NAME] --activation-policy ALWAYS] command (more information can be found on this Starting an instance section of the Starting, Stopping, and Restarting Instances guide).

@Zhen, in case you still experience this issue, can you reply to me privately with your ProjectID and Instance ID so we can help you further?

[Adam (Cloud Platform Support)]

All of the affected instances should now be restored. If anyone is experiencing further issues please file a issue on the Google Issue Tracker with the project and instance details.

[Sebastian Scotti]

Hello,

We're having the same issue that people describe in this thread on a productive DB.

Project-id: debut-backend-eu

Cloud sql instance name: debut-backend-prod-eu-db

It's a MySQL 2nd gen DB hosted in europe-west1 with a failover replica. We started to have problems about 2pm BST.

Are you having a problem now?

Thank you,

Sebastian.-

[Rishikesh Chiliveri]

@Adam

Facing the same issue instance got stuck while updating.Since last 7 days its being updated and still not able to do any operations like edit,restart or delete.

Thanks in advance.

[Ethan Tan]

Hi Adam,

Same issue here:

PROJECT ID: pigeon-express-143907

CLOUD SQL INSTANCE ID: pexdb1

i have a sql instance with:

- db-n1-standard-1

- one failover

- one replica

i was failed to connect to the instance and then i tried to restart it, since then it stuck at restart stage, from 8.15am until 10.05am.

i tried to promote my replica and same things happen to my replica after i tried to "enable binary logging".

your prompt reply is much appreciated! Thank you Google Support Team!

[siva kottapalli]

Hi Adam,

One of the cloud sql instances failed to restart after it failed. I'm unable to start, stop or restart it.
Instance Name : citibikemysql
project id : fast-ai-178518

Please let me know if you need more details.

Thank you,
Siva

[Huy Nguyen]

Hi Adam,

My project id: id-vatgia-com

Cloud SQL instance: main-db-4gb

I restarting this instance but stuck, what happend?

Tks,