Historique des pannes de Kevel

1. 20 juillet 2026 à 16:24 UTCinvestigating

   Nous enquêtons actuellement sur un problème qui cause un service publicitaire dégradé pour les clients de la région centrale de l'UE 1.

2. 20 juillet 2026 à 16:25 UTCidentified

   Le problème a été identifié et une solution est en cours de mise en oeuvre.

3. 20 juillet 2026 à 16:45 UTCidentified

   Nous continuons de travailler sur une solution à ce problème.

4. 20 juillet 2026 à 17:00 UTCmonitoring

   Un correctif a été mis en place et nous suivons les résultats.

5. 20 juillet 2026 à 17:31 UTCresolved

   Cet incident a été résolu.

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 11 juin 2026 à 14:52 UTCresolved

   We identified an issue that may have delayed delivery of beta log data via Data Shipping for some customers between June 9, 2026 at 21:16 UTC and June 11, 2026 at 14:18 UTC. June 11, 2026, 14:18 UTC: This issue has been resolved. Beta logs are now being shipped as expected, and no log data was lost. We are actively working to deliver any delayed logs from the affected period. June 12, 2026, 12:30 UTC: Confirming the backfill of all delayed data from the impacted period is complete.

1. 31 mars 2026 à 05:05 UTCinvestigating

   We are currently investigating degraded ad serving performance in the us-east-1 region.

2. 31 mars 2026 à 06:25 UTCinvestigating

   We are continuing to investigate the root cause. In the meantime, we have scaled additional capacity in the affected region, which has restored stability to ad serving. We will provide further updates as we learn more.

3. 31 mars 2026 à 09:00 UTCmonitoring

   A mitigation has been applied and we are currently monitoring the system to ensure continued stability.

4. 31 mars 2026 à 10:09 UTCmonitoring

   System performance has remained stable following the applied mitigation. We are continuing to monitor to ensure ongoing reliability.

5. 31 mars 2026 à 14:12 UTCresolved

   The issue has been resolved and system performance has remained stable. We will continue to monitor as a precaution.

1. 26 février 2026 à 19:47 UTCresolved

   Some reporting data did not process correctly between 2026-02-20 21:00:00 and 2026-02-20 22:00:00 UTC. We have since reprocessed the affected data. As a result, some reporting metrics for that time period may have increased. Additionally, in cases where clicks or conversions were fired late for ads delivered previously, the associated clicks and click-derived revenue were applied to the original delivery dates, resulting in updates to historical metrics.

1. 3 février 2026 à 21:09 UTCinvestigating

   We have identified an issue impacting ad serving. An investigation is underway with our team. We'll keep you updated along the way.

2. 3 février 2026 à 21:25 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

3. 3 février 2026 à 21:38 UTCresolved

   This incident has been resolved. Issue began: 1:55 PM ET Issue resolved: 4:37 PM ET

1. 3 décembre 2025 à 14:09 UTCinvestigating

   We are investigating an issue with delayed reporting data ingestion that is causing degraded reporting performance. Some customers may experience missing data in reports for December 2 - 3.

2. 3 décembre 2025 à 14:33 UTCidentified

   The issue has been identified and a fix is in-progress. To further clarify the impact, some customers may be temporarily missing data from December 2 - 3, as the ingestion of this data is delayed.

3. 3 décembre 2025 à 16:48 UTCidentified

   Work is still in progress on the fix for delayed data ingestion. We will update again as soon as we have further progress to share.

4. 3 décembre 2025 à 19:22 UTCmonitoring

   A fix has been implemented and we are monitoring results. We expect that it will take around 8 hours for the ingestion of the delayed reporting data to be completed. Once completed, the temporarily missing reporting data from December 2 - 3 will be available.

5. 3 décembre 2025 à 22:01 UTCresolved

   Confirming the ingestion of the delayed reporting data is now complete and this is now resolved. Newly created reports will contain correct and up to date metrics from December 2 - 3.

1. 26 novembre 2025 à 20:27 UTCresolved

   Customers in the eu-central-1 region may have experienced periods of high decision response times between 06:10 and 08:05 UTC on November 24, 2025. This was caused by anomalous behavior while the Kevel system scaled in response to irregular traffic patterns. Kevel's core ad serving systems are designed to mitigate these situations and we've adjusted our scaling algorithms following this investigation to reduce the likelihood of recurrence.

1. 20 octobre 2025 à 10:13 UTCidentified

   US-East-1 services including ad serving in the region, authentication, campaign management and reporting have been impacting by an underlying incident with one of our infrastructure providers. We're taking remediation steps and are continuing to monitor.

2. 20 octobre 2025 à 11:03 UTCidentified

   Compute resources for ad serving on US-East-1 are still experiencing increased error rates while the underlying service is being restored. Authentication, Campaign Management and Reporting are operational, but you may see elevated response times while previous failed requests are retried.

3. 20 octobre 2025 à 13:05 UTCidentified

   Confirming we are still working to address the impact from an incident with one of our infrastructure providers. Ad serving on US-East-1 is still experiencing increased error rates while the underlying service is being restored. Authentication, Campaign Management and Reporting are operational, but may still experience elevated response times.

4. 20 octobre 2025 à 14:09 UTCidentified

   We are continuing to work with our infrastructure provider to fully resolve the incident. Ad serving in the US-East-1 region remains partially impacted, with elevated error rates still observed. Authentication, Campaign Management, and Reporting remain operational, though some customers may continue to experience higher-than-normal response times. Our teams remain actively engaged, and we’ll share another update as soon as more progress is made.

5. 20 octobre 2025 à 15:05 UTCidentified

   We are continuing to actively respond to this incident. Ad serving in the US-East-1 region remains impacted, with elevated error rates and latency still observed. Authentication and Campaign Management remain operational, though some customers may continue to experience higher-than-normal response times. Reporting, including Data Shipping, are continuing to experience degraded performance and customers may see reports failing. Completed reports may also lack recent data. Forecast is also experiencing degraded performance and recently completed forecasts may yield less accurate results.

6. 20 octobre 2025 à 17:34 UTCidentified

   We are continuing to work with our infrastructure provider to fully resolve the incident. Ad Serving is now stable. If you continue to experience issues with Ad Serving, please reach out to Support or your to Customer Success Manager. Authentication and Campaign Management remain operational, though customers may continue to experience elevated response times. Reporting, including Data Shipping, continue to experience degraded performance and customers may continue to see failed reports. Completed reports may also lack recent data. Forecast is also continuing to experience degraded performance and completed forecasts may yield inaccurate results. For Audience, event tracking is operational, however, purchase request syncing may be degraded for some customers.

7. 20 octobre 2025 à 20:45 UTCidentified

   We are continuing to engage with our infrastructure provider to resolve the incident. Ad Serving continues to be stable. Management is stable. Reporting, including Data Shipping, is still experiencing degraded performance and customers may continue to see failed reports. In addition, reports from today may lack recent data. Forecast is also experiencing degraded performance and forecasts from today may show inaccurate results. For Audience, purchase request syncing may be degraded for some customers.

8. 20 octobre 2025 à 22:19 UTCidentified

   We are continuing to work with our infrastructure provider to address the impact of the incident. Ad Serving continues to be stable. Management is operational, has returned to pre-incident status and has worked through the backlog of changes. Reporting and Data Shipping are still experiencing degraded performance and customers may continue to see failed reports. In addition, today's reports may lack recent data. Forecast is also experiencing degraded performance and today's forecasts may show inaccurate results. For Audience, purchase request syncing may continue to be degraded for some customers.

9. 20 octobre 2025 à 23:56 UTCidentified

   Ad Serving has now been returned to an operational status. Audience has also been returned to an operational status. Management also continues to be operational. We are continuing to work with our infrastructure provider to address the impact on Reporting, Data Shipping and Forecast. Reporting and Data Shipping are still experiencing degraded performance and customers may continue to see failed reports. In addition, today's reports may lack recent data. Forecast is also experiencing degraded performance and today's forecasts may show inaccurate results.

10. 21 octobre 2025 à 01:01 UTCidentified

    Ad Serving, Management and Audience continue to be operational. Data Shipping has resumed normal operations, however, customers may be missing data from today, October 20th. Confirming we will evaluate backfilling this data. Reporting and Forecasting are still operating with degraded performance. We are waiting for recovery of the related service from our infrastructure provider to resume normal operations. We anticipate this will take hours.

11. 21 octobre 2025 à 09:07 UTCmonitoring

    All Kevel services are now operational including Ad Serving, Audience, Data Shipping, Management, Forecast and Reporting. We are continuing to monitor all services. We are also still working to backfill missing data from yesterday's incident for Data Shipping and Reporting.

12. 21 octobre 2025 à 17:02 UTCresolved

    We confirm that all Kevel services have remained operational for several hours, and we are moving the incident to resolved. Please note that we are still working to backfill missing data for Reporting and Data Shipping. We anticipate this will be completed by the end of the week. If you have any questions about the incident or the ongoing backfill, please reach out to your Customer Success Manager or Kevel Support.

1. 14 octobre 2025 à 18:37 UTCinvestigating

   We identified an issue impacting Reporting. Reports are processing through the system. We'll keep you updated along the way.

2. 14 octobre 2025 à 18:49 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

3. 14 octobre 2025 à 19:42 UTCmonitoring

   We are continuing to monitor for any further issues.

4. 14 octobre 2025 à 20:12 UTCresolved

   This incident has been resolved.

1. 9 octobre 2025 à 12:22 UTCinvestigating

   We are currently investigating an issue that is causing degraded Ad Serving for customers in the EU Central 1 region.

2. 9 octobre 2025 à 12:44 UTCidentified

   The issue has been identified and a fix is being implemented.

3. 9 octobre 2025 à 12:53 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

4. 9 octobre 2025 à 13:41 UTCresolved

   This incident has been resolved.

1. 12 août 2025 à 16:39 UTCinvestigating

   We are currently investigating an issue impacting Kevel Ad Serving. We will keep you updating along the way.

2. 12 août 2025 à 17:13 UTCidentified

   We have identified the issue and are working on a fix.

3. 12 août 2025 à 19:48 UTCidentified

   A fix has been implemented for most impacted customers. We are deploying the fix to the remaining impacted customers and monitoring.

4. 12 août 2025 à 20:37 UTCresolved

   This incident has been resolved.

1. 31 juillet 2025 à 16:09 UTCmonitoring

   We identified an issue impacting Forecast. Some instances were not responding to Forecast requests or Forecast API requests from ~15:15 UTC to ~15:55 UTC on July 31, 2025. A fix has already been deployed and we are monitoring.

2. 31 juillet 2025 à 16:24 UTCresolved

   This incident has been resolved.

1. 29 juillet 2025 à 17:11 UTCresolved

   We identified an issue with our ad publication and balancing systems starting at 22:23 UTC on July 23, 2025. The issue was resolved at 17:12 UTC on July 24, 2025. Some customers may have experienced underdelivery on affected ads.

1. 29 juillet 2025 à 17:14 UTCresolved

   We identified an issue with our ad publication and balancing systems that began on July 21, 2025, at 16:55 UTC. The issue was resolved on July 28, 2025, at 21:41 UTC. As a result, some customers may have experienced overdelivery of affected ads.

1. 2 juillet 2025 à 21:02 UTCinvestigating

   We identified an issue impacting ad publication starting 2025-07-02 20:10 UTC. Users might experience issues starting, stopping, and pausing ads; however, ad balancing is not impacted. An investigation is underway with our team. We'll keep you updated along the way.

2. 2 juillet 2025 à 21:03 UTCidentified

   The issue has been identified and a fix is being implemented.

3. 2 juillet 2025 à 21:13 UTCresolved

   This incident has been resolved.

1. 8 juin 2025 à 18:09 UTCinvestigating

   We identified an issue impacting ad publication starting 2025-06-08 16:00 UTC. An investigation is underway with our team. We'll keep you updated along the way.

2. 8 juin 2025 à 18:11 UTCinvestigating

   A fix has been implemented and we are monitoring the results.

3. 8 juin 2025 à 18:30 UTCinvestigating

   We are continuing to monitor for any further issues.

4. 8 juin 2025 à 18:31 UTCresolved

   This incident has been resolved.

1. 25 avril 2025 à 19:09 UTCinvestigating

   We identified an issue with our ad publication & balancing systems starting 2025-04-25 18:29 UTC. This is being investigated with high priority. This impacts us-east only.

2. 25 avril 2025 à 19:31 UTCidentified

   The issue has been identified and a fix is being implemented.

3. 25 avril 2025 à 19:42 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

4. 25 avril 2025 à 20:03 UTCmonitoring

   We are continuing to monitor for any further issues.

5. 25 avril 2025 à 20:23 UTCresolved

   This incident has been resolved.

1. 16 avril 2025 à 16:47 UTCresolved

   We identified an issue with our ad publication & balancing systems starting 2025-04-15 14:00 UTC. This was resolved on 2025-04-16 15:50 UTC. Some customers may have experienced a delay in new ads serving, as well as a delay in ads restarting after a daily cap was met prior day.

1. 10 avril 2025 à 13:39 UTCresolved

   We identified an issue with our ad publication & balancing systems starting April 9 at 18:00 UTC. This was resolved on April 10 at 14:30 UTC. Some customers may have experienced a delay in new ads serving, as well as a delay in ads restarting after a daily cap was met prior day.

1. 25 mars 2025 à 13:35 UTCinvestigating

   We identified an issue impacting ad serving in us-east-1, beginning at 13:10 UTC. An investigation is underway with our team. We'll keep you updated along the way.

2. 25 mars 2025 à 13:59 UTCidentified

   We have identified the issue and are working on a fix.

3. 25 mars 2025 à 14:18 UTCmonitoring

   A fix has been implemented and we are monitoring the results.

4. 25 mars 2025 à 14:29 UTCresolved

   This incident has been resolved.