exe.dev status

Approximately 600 VMs are currently unreachable. We are investigating.

This is now resolved. Please let us know if you see any problems.

We had an issue that is causing some connectivity trouble for some VMs. Our services are designed to have a clean handoff of current state when they are updated. Unfortunately when rolling out a new instance an error caused some of the service state to remain with the old instance, rather than being transferred to the new one. An unrelated error caused the new instance to not behave properly. To fix these combined problems, we shut down both the old and new instances, and restarted the service from a rolled back version. This worked, but meant that the newly started instance no longer started with the state transferred from the old one. This made some VMs inaccessible in a few regions. The problem was fixed by synchronizing the full state in the new instance. Sincere apologies for the down time.

A network issue is causing VM connectivity issues for some number of VMs.

An upstream network issue is causing VM connectivity issues for a limited number of VMs.

We are having issues with our certificate registration that are causing connectivity and system availability. Update: we've identified the issue and mitigated on our side. The exe.dev service is now available however, due to an on-going issue with Let's Encrypt (https://letsencrypt.status.io/) custom domain certificate issuance may be delayed or have issues. Please reach out to support@exe.dev if you have trouble.

We are investigating an issue that could cause issues with starting or managing VMs on a small set of users.

We are investigating a network issue that is causing connection issues.

We are investigating an issue with one of our compute hosts that could impact VM availability.

We have identified an issue with TLS certificate generation for custom domains. Update: We've deployed the fix and the issue should be resolved.

A network error caused one of our compute hosts to have connectivity issues affecting a small subset of VMs. The compute host had to be restarted. All VMs are now back online.

Testing.