Linode Outage History

On February 17th, 2026, between 21:25 UTC and 22:50 UTC, and on February 18th, 2026, between 11:44 UTC and 12:16 UTC, we experienced connectivity issues in our FR-PAR \(Paris\), FR-PAR-2 \(Paris 2\), and ES-MAD \(Madrid\) data centers. During the impact windows, customers may have experienced intermittent connection timeouts and errors for all services deployed in those data centers. ‌ Our investigation identified the cause as an unstable interface link between Washington and Paris data centers. To resolve the issue, we removed the affected interface from service at 12:16 UTC on February 18th, 2026. ‌ To help prevent similar issues in the future, Akamai will review and enhance our monitoring and procedures. ‌ 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 recurrence. ‌ This summary provides an overview of our current understanding of the incident, given the information available. Our investigation is ongoing, and any information herein is subject to change.

Between 07:40 and 10:52 UTC on February 12, 2026, some customers experienced issues accessing their Block Storage volumes in the Stockholm region. During this period, storage operations stalled, resulting in a temporary denial of service for affected users. Monitoring detected a complete drop in storage throughput, and impacted customers were unable to access data on their volumes during the impact window. The incident was traced to a gap in the execution sequence during a recent storage system upgrade, which resulted in the storage environment entering a degraded state. The scenario was documented, however it was not clearly incorporated into the upgrade workflow, as it had not been encountered in previous upgrades. To mitigate the impact, Akamai initiated recovery actions and restored normal working service. In response, the upgrade process has been updated to include clearer guidance for handling similar situations. This summary provides an overview of our current understanding of the incident given the information available. Our investigation is ongoing and any information herein is subject to change

Starting around 11:04 UTC on February 10, 2026, some customers were unable to access the Longview graph dashboard. Longview is a system data graphing service that tracks metrics for CPU, memory, and network bandwidth on both an aggregate and per-process basis. It also provides real-time graphs that can help expose performance problems \(more details about Longview are available [here](https://techdocs.akamai.com/cloud-computing/docs/longview)\). The investigation revealed that the issue was caused by an internal certificate expiry for the [longview.linode.com](http://longview.linode.com) hostname. The impact was limited to reading the existing reporting data only, while the write operations remained unaffected, and there was no permanent reporting data loss due to this issue. Akamai's internal automated tool rotated the expiring certificate before the certificate expired; however, the internal certificate tool which was supposed to auto-reload the system, did not reload the Longview servers after the previous certificate was rotated. The issue started when the previous certificate expired. To mitigate the impact, we manually reloaded the Longview servers for the new certificate to take effect. The impact was mitigated at 23:44 UTC on February 10, 2026, following this action. We will continue to investigate the root cause and will take appropriate preventive actions. 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 recurrence. This summary provides an overview of our current understanding of the incident given the information available. Our investigation is ongoing, and any information herein is subject to change.

On February 4, 2026, starting at 14:43 UTC, users accessing [gb-lon-1.linodeobjects.com](http://gb-lon-1.linodeobjects.com) experienced increased error rates due to an issue with one of the storage backends supporting the London Object Storage service. During this period, success rates dropped to between 93% and 95%. Our team identified that one of the six storage backends was experiencing problems, which affected service availability. The affected components recovered on their own, and service success rates returned to normal by 15:19 UTC. We are continuing to investigate the underlying cause of the backend issue and will implement additional safeguards to help prevent similar incidents. Thank you for your patience as we work to enhance service reliability. This summary reflects our current understanding of the incident. Our investigation is ongoing, and details may be updated as more information becomes available.