Deribit’s 2020 Outage and Master-Node Failover
Summary
Deribit describes a platform outage on August 27, 2020, and the steps taken to restore trading. A hardware failure affected the master node that connected the platform’s gateways. Remote access could not restart it, and engineers at the data center were unable to bring it back online. The exchange instead promoted a regular node to serve as the new master. Clients regained access at 9:22 UTC; trading resumed at 9:27 UTC after final checks and a five-minute order cancellation period.
The account illustrates how a system with redundant load balancers and nodes can still depend on a single master component, and how failover can restore service when that component fails. Deribit said a Zurich disaster recovery facility was being prepared, but it was not yet ready to back up production trading. The statement is an exchange’s account of one incident; it gives no independent assessment of the outage’s market impact or detailed service-level data.
Key ideas
- A hardware failure in the master node caused the exchange outage.
- Remote access and on-site attempts could not restart the failed node.
- Promoting a regular node to master restored the platform.
- The planned Zurich recovery facility was not yet ready to support production trading.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.