Outage recap · DeepL

DeepL outage — February 13, 2025

On February 13, 2025, DeepL had 3 reported incidents on DeepL's official status page. Below is what was reported — impact, timing, and the AI services the same upstream outage could affect.

Elevated error rates on DeepL Pro

minor
Started
Feb 13, 3:49 AM EST
Resolved
Feb 13, 4:02 AM EST
Duration
14 min

DeepL's updates (4)

  1. investigating · Feb 13, 3:49 AM EST

    We are currently investigating this issue.

  2. monitoring · Feb 13, 3:58 AM EST

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

  3. resolved · Feb 13, 4:02 AM EST

    This incident has been resolved.

  4. postmortem · Feb 13, 9:49 AM EST

    On February 13, we experienced fluctuating error rates starting at around 07:00 UTC, with a significant drop in traffic between 08:33 UTC and 08:54 UTC. One of our load balancers reached 100% CPU usage, which likely caused the elevated error rates for some DeepL Pro users. We received an alert at 08:46 UTC and promptly took the faulty load balancer out of the pool. The root cause is still under investigation, but we’ve implemented a fix, and service has returned to normal. We apologize for the disruption and are working to prevent future occurrences.

Updates are DeepL's own words from its status page (status: postmortem); we never add an inferred cause. The vendor's incident page ↗

Document Translation Pro

major
Started
Feb 13, 4:22 AM EST
Resolved
Feb 13, 4:43 AM EST
Duration
22 min

DeepL's updates (3)

  1. investigating · Feb 13, 4:22 AM EST

    We are currently investigating issues for PDF Documents not being translated.

  2. resolved · Feb 13, 4:43 AM EST

    This incident has been resolved.

  3. postmortem · Feb 13, 5:24 AM EST

    Today, February 13, we encountered a follow-up issue related to the incident that took place yesterday, February 12. From 08:55 UTC to 09:38 UTC, Document Translation for PDF files on DeepL Pro was temporarily unavailable. The issue arose as part of the ongoing recovery from yesterday’s incident that disrupted our document translation services. The root cause has been identified, and a fix has been implemented. We deeply regret the inconvenience caused by this continued disruption and are prioritizing efforts to ensure a full restoration of service.

Updates are DeepL's own words from its status page (status: postmortem); we never add an inferred cause. The vendor's incident page ↗

Major outage on api-us.deepl.com

major
Started
Feb 13, 7:47 AM EST
Resolved
Feb 13, 7:47 AM EST
Duration
—

DeepL's updates (2)

  1. resolved · Feb 13, 7:47 AM EST

    The US region-specific endpoint api-us.deepl.com was unavailable from 12:16 UTC to 12:17 UTC.

  2. postmortem · Feb 14, 2:44 AM EST

    During routine maintenance at 12:16 UTC, a seemingly unused component of our network infrastructure was removed. Unfortunately, this resulted in an interruption of requests to our US region \([api-us.deepl.com](http://api-us.deepl.com)\). After identifying and fixing the problem, we restored functionality for requests to our US region at approximately 12:17 UTC. We apologise for any inconvenience this may have caused. The problem was caused by the removed component still being configured in one place in our infrastructure. As a result, traffic wasn't being redirected to the US region. Once the fix was in place, we also identified the root cause of the initial incident and fixed it to prevent this exact problem from happening again. In addition, we are currently reviewing our traffic routing mechanisms to prevent this issue from recurring.

Updates are DeepL's own words from its status page (status: postmortem); we never add an inferred cause. The vendor's incident page ↗

AIDownCheck is not affiliated with DeepL. This recap summarizes incidents published on DeepL's official status page.