0

We had sort of a disastrous event with a database at work (managed on Cloud SQL, it's a MySQL db) and thankfully we have point in time recovery enabled.

We went ahead and cloned the production database to a certain point in time to be able to recover the data just before the disaster, but it's now been running for more than 5 hours for a 69GB database (the size advertised on the GCP Cloud SQL panel, so the real size of the DB is probably less than that).

Does anyone have experience with this ? The status of the operation when querying it with the gcloud CLI says "RUNNING". We checked the logs of the instance to see if anything was off but there just isn't any log at all.

Weev
  • 36
  • 1
  • 8
  • The performance depends on the setup of your Cloud SQL instance. Have you checked the RAM usage when performing PITR? There are instances that there would be high performance impact when doing data recovery especially if the instance is using minimal setup as this would have a performance bottleneck when recovering large files. – Robert G Jan 23 '23 at 22:51

0 Answers0