Several of our PR Environments were deleted at the same time and we're trying to figure out why
Project ID: e93207a4-8a4a-4754-852b-af5a639274b1
We pushed out a deployment to our Production environment yesterday (merged a PR into our
production branch in GitHub), which started the deployment process on Railway in the Production environment at around Sep 30, 9:19 AM UTC.
At Sep 30, 9:20 AM UTC, most of our PR Environments were deleted, including those which still had PRs open.
We're trying to figure out if this was because of something we hadn't configured correctly in Railway, a user error from one of our team members or some other underlying problem. I can't seem to find any audit logs for deleted environments or user activity in our dashboard, so any idea how we can investigate this?
I'm simply trying to understand what wen't wrong here so we don't run into any further disruptions in the future.3 Replies
Project ID:
e93207a4-8a4a-4754-852b-af5a639274b1In addition to this, we're noticing that any PR environment being created now has a completely new instance of our Postgres and Redis service as well, meaning the databases have no data and are in a blank state.
Up until a couple of days ago, every time a PR environment was created, it used to share the same database services and volumes as the configured base environment (called
development in our project) which meant the same database instance was shared among all of them.
Has this been a recent change?While I don't know what happened to those PR environments, I can assure you with utmost certainty that databases created in PR or cloned environments are empty, they always have been, they are completely new deploys and do not copy over any data from the database in the base environment, we have never copied the data over for you.