Typesense incident

Power Outage in two data centers in the UAE

Major Ongoing
Started
Mar 01, 2026, 12:30 PM UTC
Resolved
Ongoing
Duration
● 63d 9h
Detected by Pingoru
Mar 01, 2026, 12:30 PM UTC

Affected components

TypesenseCluster Regions (UAE)

Update timeline

  1. identified Mar 01, 2026, 12:30 PM UTC

    Our infrastructure provider is experiencing a power outage in one of the three data centers in the UAE region. Customers who have enabled High Availability for their Typesense Cloud Cluster continue to operate normally even with one node in the affected data center being down, as traffic is being re-routed away from the affected data center to the other two nodes in the other data centers in the region. Customers who do not have High Availability enabled for their Typesense Cloud Cluster and have their node running in the affected data center are affected by this issue, and will remain unavailable until power is restored.

  2. identified Mar 01, 2026, 12:30 PM UTC

    A second data center in the UAE region has lost power, due to active military conflict in the region. An HA Typesense Cluster with 3 nodes can survive a maximum of 1 node failure, and with 2 data centers in the region now down, 2 nodes in HA clusters in the region are now inaccessible, which makes the 3rd node stop serving traffic to avoid split brain scenarios. Our infrastructure provider estimates that it might take at least a day to evaluate critical systems and their safety, before providing an ETA for full restoration. Given the severity and uncertainty of this event, we strongly recommend: 1. Provisioning a new cluster in a different region outside the Middle East. 2. Re-indexing data from your primary database. 3. Updating your application configuration to point to the new cluster. Please note: the cluster cloning feature is unavailable because the regional control plane is also affected.

  3. identified Mar 03, 2026, 03:31 AM UTC

    Our infrastructure provider has completed their initial assessment, and confirmed that the two data centers in the UAE region have sustained structural damage due to drone strikes. They estimate full recovery to take a prolonged amount of time given the extraordinary nature of the situation. Like mentioned earlier, we recommend that users with clusters in this region provision new clusters in a region outside the Middle East, reindex their data from their primary DB and update their applications to use the new regions.

  4. identified Mar 09, 2026, 07:22 PM UTC

    The situation has not improved in the Middle East regions affected by the recent drone strikes. Our infra provider does not have an ETA on when they will be able to restore these data center facilities, due to the extraordinary nature and fluidity of the situation. So we recommend that users with clusters in the UAE and Bahrain regions provision new clusters in a region outside the Middle East, reindex their data from their primary DB and update their applications to use the new regions. Since our infra provider's control plane is also down in these regions, it is not possible to access or export the existing data out of these regions. So it is not possible to clone or modify cluster configuration (like RAM, CPU, enabling HA, enabling SDN, etc) in these two regions. If you have have a multi-region (Search Delivery Network) cluster with at least one node outside of these two affected regions, please reach out to support at typesense dot org, for assistance in migrating your data out of your SDN cluster.

Looking to track Typesense downtime and outages?

Pingoru polls Typesense's status page every 5 minutes and alerts you the moment it reports an issue — before your customers do.

  • Real-time alerts when Typesense reports an incident
  • Email, Slack, Discord, Microsoft Teams, and webhook notifications
  • Track Typesense alongside 5,000+ providers in one dashboard
  • Component-level filtering
  • Notification groups + maintenance calendar
Start monitoring Typesense for free

5 free monitors · No credit card required