Fireworks AIService Degradation for Kimi K3 US
We've resolved the incident and the affected service is now operational.
Resolved
Outages and recoveries from the services you depend on.
40 incidents
We've resolved the incident and the affected service is now operational.
Resolved
This incident has been resolved.
Resolved
We've resolved the incident and the affected service is now operational.
Resolved
**Incident Summary** On August 12, 2026, between approximately 10:31 AM and 10:45 AM PDT, the Lever platform experienced partial unavailability. During this 14-minute window, some customers were unable to access the platform, and others experienced elevated error rates for API requests. **Detection** The issue was identified through internal monitoring. A Datadog alert for high database page faults triggered at 10:37 AM PDT, notifying the engineering team. This allowed responders to isolate the specific database nodes experiencing high load and identify the responsible operations. **Root Cause** The incident was triggered by a safety mechanism within an internal administrative tool designed to prevent expensive database queries. When an operator searched for unindexed data, this safety check requested a "plan" from the database to see if the query was safe to run. However, because the data was unindexed, the database attempted to evaluate every possible retrieval method, resulting in the same large account being scanned 11 times for a single request. These diagnostic checks bypassed standard time limits, consuming the database read capacity required for normal customer traffic. **Resolution** Engineering teams identified and terminated the long-running diagnostic operations, which immediately restored database performance and platform stability. Following the cleanup, a database-level safeguard was applied to force these specific queries into a single, efficient path, preventing the database from attempting multiple expensive scans in the future. **Preventative Measures** * **Redesigning Safety Checks:** We are replacing the current diagnostic method with a new system that verifies query safety by inspecting database definitions directly, eliminating the need to run expensive tests on live data. * **Advanced Alerting**: We are implementing new monitoring for database read churn and long-running operations to detect and resolve similar issues before they affect customers. * **Enhanced Logging:** We are improving internal audit logs to ensure faster identification and attribution of administrative database activity. * Database Optimization: We are adding new indexes to properly support these search patterns, ensuring they are inherently efficient.
Resolved
The connection between Linear and OpenAI was failing and has now been rectified.
Resolved
**Incident Summary** On August 7, 2026, between approximately 6:00 AM and 5:30 PM ET, some customers were unable to send offer letters, and some document and resume previews failed to generate. Files uploaded successfully; only the step that converts them into a viewable PDF or preview failed. No data was lost. **Detection** Identified on August 7 at approximately 11:47 AM ET following customer reports to our Support team. **Root Cause** A third-party service we use to convert documents into PDFs experienced an outage lasting most of the day. During it, the service accepted conversion requests but never completed them. Our system is designed to switch to a backup conversion method when that service reports a failure — but because these requests neither succeeded nor failed, they simply waited, and the backup was never used. Each waiting request also occupied one of a limited number of conversion slots, so delays spread to other documents. **Resolution** We added a strict overall time limit to document conversions: if the third-party service does not finish within that window, we stop waiting and convert the document using our own backup method instead. We also corrected a related defect that had prevented the backup from being used reliably. Both fixes were released to production on the afternoon of August 7, while the third-party outage was still ongoing, and conversion success rates returned to normal levels the following day. We then re-processed the documents that had failed to convert during the outage. **Preventative Measures** · We fixed a defect that had prevented this service's health measurements from reaching our monitoring system and are adding alerts on the document conversion failure rate, so issues like this are detected automatically and quickly, rather than through customer reports. · The new time limit can be adjusted without a code release, and we are documenting a procedure to route all conversions away from the third-party service during a vendor outage.
Resolved
The incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
We've resolved the incident and the affected service is now operational.
Resolved
We can confirm that the issue was mitigated at 08:30 UTC on August 14, 2026 and the service has resumed normal operation. Customers and partners can view additional details about the incident by logging in to: https://community.akamai.com/customers/s/feed/0D5a700001NAvXBCA1 or reaching out to Akamai Support. We apologize for the impact and thank you for your patience and continued support. We are committed to making continuous improvements to make our systems better and prevent a recurrence of this issue.
Resolved
This issue is now resolved
Resolved
The issue is being marked as resolved. We're seeing improved latencies and our systems have recovered.
Resolved
Live Chat behavior is stable across the board, so we're closing this out. Thank you for your patience while we worked on this one!
Resolved
We've resolved the incident and the affected service is now operational.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved. The affected webhook events remained queued and were delivered after service recovered; no events were lost. The queue is now clear, and no customer action is required.
Resolved
This incident has been resolved. Between August 12, 17:12 UTC and August 13, 20:33 UTC, some Vercel deployments could return 404 errors from dynamic API routes. Deployments built on Vercel during this period were affected only if they met all three conditions: used Next.js 16.2 or earlier, used the Pages Router with internationalization (i18n configured in next.config.js), and used dynamic API routes, such as pages/api/[slug].ts or pages/api/[...path].ts. Deployments built outside Vercel (with vercel deploy --prebuilt) were not affected, and existing deployments not rebuilt during this period were not affected. If your project matches the conditions above and you deployed during the impact window, redeploy your project. Affected deployments are not fixed automatically; a new deployment is required to restore the affected routes.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
The issue resulting in users being unable to reach status.claude.com has been resolved.
Resolved
This incident has been resolved.
Resolved
As of 4:47 PST / 11:47 UTC the errors have returned to baseline.
Resolved
As of 7:34 PST / 14:34 UTC the errors have returned to baseline.
Resolved
This issue has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
As of 15:20 UTC (8:20am PT), error rates on Claude Sonnet 5 have returned to baseline. This incident is resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This issue has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This issue has been resolved.
Resolved
The issue affecting Claude Mythos 5 and Claude Fable 5 has been resolved.
Resolved
The issue affecting Claude services has been resolved. Impact occurred from 23:42 to 00:50 UTC.
Resolved
A fix has been rolled out by our vendor, and we’re seeing improved performance for Lovable Cloud creation. The incident is now considered resolved, and we’ll continue to monitor performance.
Resolved
This incident has been resolved.
Resolved