Error 401 API deploy on GCP Cloud Run

[woi-workstation1]

I'm getting a 401 authentication failed error when using the API.
I'm using Docker image 8.1.4 deployed on GCP Cloud Run.
This error occurs when the Apache service dies, causing Cloud Run to create a new instance.

My token's expiration date is still far off.

What could be causing the token to be unusable again?
My current solution is just to generate a new token.

[woi-workstation1]

hi dev,
any updates on this issue?
user and personal tokens are still available in the system, but cannot be used to authenticate ?
users must generate new tokens to successfully authenticate to the system

[uberbrady]

I think the keys get stored on the storage volume - if you're not re-attaching it, you will lose them - as well as any images or files that you uploaded.

[smaddock]

Same issue as OP. We are re-attaching /public/uploads/ and /storage/private_uploads/ which were the two documented paths to persist; are the keys stored elsewhere?

[smaddock]

The only file getting updated in the Snipe-IT directory when I create a new API token is my session cookie in /storage/framework/sessions/. Maybe it's getting stored outside of the project directory? The oauth_access_tokens database table is also getting a new entry, but it only has an ID. My other guess is the API token ID is combined with a salt somewhere on disk (perhaps /storage/oauth-private.key?) that is getting written once when a new instance is stood up.