Recovering lost connections and data after upgrading GCP hosted instance without migrate flag

Slack Knowledge Base Q&A Platform Questions
, , , ,
1

Summary

After upgrading GCP hosted instance without migrate flag, connections are lost. User is asking if it’s possible to recover the connections and data.


Question

We upgraded our GCP hosted instance today w/out the migrate flag set. All of our connections are now gone. Is there anyway that we can recover the connections? Can we recover our data?


This topic has been created from a Slack thread to give it more visibility.
It will be on Read-Only mode here. Click here if you want
to access the original thread.

Join the conversation on Slack

["gcp-hosted-instance", "connections", "data", "recover", "upgrade", "migrate-flag"]
2

Do you use external database or built-in pod/container for PostgreSQL used by Airbyte?

3

Can you connect to database and check if there are any records in connection table?

4

For the future,

  1. I’d recommend backups. especially before maintenance
  2. storing connections configuration in terraform
  3. ideally, external database if you can afford it
5

> Can we recover our data?
What kind of data exactly do you want recover apart from connections’ configuration? What destination connectors do you use?

6

OK. I’m looking to just recover my connection settings

7

Another thing to check is whether there are any snapshots or images of the system disk (I’m assuming this is a Compute Engine VM).

If there are either instant snapshots or scheduled standard snapshots, you could use one of those to restore from. (Machine Images or Disk Clones would also work, but it’s unlikely there would be one as this is usually a manual operation.)

I’m not sure if there’s any safety net here, but once you’re through this I would recommend putting a GitHub issue in that they should check/warn in the case of existing data to prevent this in the future (probably with some type of flag to force the overwrite).

8

We used the built-in container