Beginning at approximately 14:23 UTC on October 19, 2026, as part of a planned internal maintenance, we migrated too many of the same routing resources at the same time. This resulted in route instability in certain network segments until the migration completed and routes were adopted and converged at 14:55 UTC on October 19, 2026.

To prevent this issue from reoccurring, we are modifying our maintenance planning process to ensure that critical components are taken offline in a phased manner that doesn’t introduce customer impact.

This summary provides an overview of our current understanding of the incident given the information available. Our investigation is ongoing and any information herein is subject to change.