We recently addressed an issue that affected Box Search and related Query functionality, including the Public APIs and Box Apps features that rely on them. We would like to take the opportunity to explain what happened and the steps we have taken to prevent it from happening again.

On July 7, 2026 between 06:22 AM and 08:53 AM PT, some customers may have experienced elevated latencies and intermittent errors when using the Box Search and Query functionality. Public APIs for these features, Box Apps, Web App Search, and other functionalities that depends on Search and Query APIs were also impacted. This incident was caused by a sudden surge in indexing traffic that overwhelmed services in one of our subsystems, before they could be scaled up or rate limiting could kick in. We were able to remediate the issue by isolating the offending traffic and allowing the surge of indexing work to finish processing and drain. Regular traffic routing was restored once the affected infrastructure was healthy again.

Analysis

This incident revealed an opportunity to improve how we protect our search indexing subsystem from sudden~~,~~ and concentrated surges of load. While our stateless services can quickly autoscale horizontally, our stateful systems cannot be scaled as quickly to handle sudden surges in load. We’ve historically focused on query traffic driven load and over-provisioned to handle any increases in internally generated indexing load. This incident revealed that sudden and large indexing surges can impact stability more than we anticipated. Our safeguards did not sufficiently slow or throttle a large burst of re-indexing traffic, and a temporary configuration change in place at the time of this incident caused files to be re-indexed in full even when only non-content attributes had changed, amplifying the load.

Corrective Actions

Box has initiated the following corrective actions:

• Add per source rate limiting to the indexing pipeline so a large burst of work cannot overwhelm the subsystem.
• Optimize indexing load, so files are only re-indexed when their content has actually changed.
• Set safer limits on how much work the pipeline will accept at once.
• Improve monitoring and alerting to detect unusual spikes in processing earlier and respond before customers are impacted.

‌

We are continuously working to improve Box and want to make sure we are delivering the best product and user experience we can. We hope this provides clarity on what occurred and the steps we are taking. We would be happy to answer any questions you may still have regarding this matter.

Sincerely,
The Box Team