Data loss incident

During disaster recovery testing today, our most recent nightly backup was mistakenly restored to our production GitLab instance instead of a testing instance.

This means that all data from today, Friday 29th March 2019 has been lost and, as we restored the latest backup, taken at approximately midnight, there is no later backup to restore from.

We recognise that this level of data loss is extremely aggravating to our users. It should never have happened and, more importantly, our processes should have been such that it could not have happened. We shall prepare a post-mortem report next week to give a timeline of the incident and mitigation strategies we will adopt to make sure this does not happen again. Resiliancy measures such as this are a requirement for us to transition to a beta service.

The GitLab team offers their sincere apologies for this incident and assures you that lessons will be learned from it.