Cloudflare·

Cloudflare outage: an oversized configuration file crashes its proxy worldwide

A database permission change doubled the size of a Bot Management file, which crashed Cloudflare's core proxy and returned errors for a large share of the web for about three hours.

ResolvedEnded Nov 18, 2025
Security:
Explicitly ruled out
Started
Nov 18, 2025, 11:20 UTC
All services restored
Nov 18, 2025, 17:06 UTC
Duration
3 h 10 minto: Main impact resolved5 h 46 min to: All services restored
Full recovery
Complete
Geographic scope
Global
Last verified
2 sources

What happened

Systems and services affected
CDN and security proxyTurnstileWorkers KVCloudflare DashboardCloudflare AccessEmail Security
Kind of failure
Configuration change
Sector
Internet infrastructure
Provider report
Published
Status and recovery
Resolved. Main impact ended at 14:30 UTC; Cloudflare says all services were restored at 17:06 UTC.[1]

What caused it?

Root cause: Confirmed

A change to database permissions made a ClickHouse query return duplicate rows, doubling the Bot Management feature file past a size limit in the proxy software, which then failed.[1]

Provider report: Published. Cloudflare published its report on November 18, 2025.[1]

Was it malicious?

Explicitly ruled out

Cloudflare: "The issue was not caused, directly or indirectly, by a cyber attack or malicious activity of any kind."[1]

How the failure happened

  1. 1TriggerDatabase permissions change
  2. 2FailureFeature file doubled in size
  3. 3SpreadProxy failed
  4. Affected6 services and systems

Evidence for each step

  1. 1
    Database permissions change[1]
    Starting point.
  2. 2
    Feature file doubled in size. A ClickHouse query returned duplicate rows[1]
    Led from step 1: Stated in a provider report or official record.
  3. 3
    Proxy failed. The file exceeded a size limit in the proxy software[1]
    Led from step 2: Stated in a provider report or official record.

Only steps the sources state are drawn. Affected services are listed under What happened.

Explanation

Both versions summarize the sourced claims on this page. Our own interpretation appears only under Independent analysis.

Plain language: what happened and why people were affected.

Many websites that use Cloudflare showed error pages for several hours on November 18, 2025.

Cloudflare sends its servers a file that helps tell bots from people. A database change made that file twice as big as normal. The software reading it had a size limit and crashed.

Cloudflare first thought it was under a huge attack, partly because its status page went down at the same time by coincidence. It was not an attack.

Confirmed and unknown

Confirmed

  • Confirmed
    A database permissions change caused the Bot Management feature file to contain duplicate entries and double in size.[1]
    Said by: Cloudflare · Evidence: Official
  • Confirmed
    The proxy software had a feature-file size limit below the doubled size, and failed.[1]
    Said by: Cloudflare · Evidence: Official
  • Confirmed
    Cloudflare's status page, hosted outside Cloudflare, went down at the same time by coincidence.[1]
    Said by: Cloudflare · Evidence: Official

Denied

  • Denied
    It was a hyper-scale DDoS attack.[1]
    Said by: Cloudflare (its own initial suspicion) · Evidence: Official

Still unknown

No open questions: the provider has published a full account.

Timeline

Times in UTC. Dates without a time mean the sources give no time. Only events a source documents are listed.

  1. First observedNov 18, 2025, 11:20 UTC
  2. AcknowledgedNot documented
  3. MitigationNov 18, 2025, 14:30 UTC
  4. RecoveryNot documented
  5. ResolvedNov 18, 2025, 17:06 UTC
  6. Incident reportNov 18, 2025
Nov 18, 2025, 11:05 UTCNov 18, 2025, 17:06 UTC
  1. Database access control change deployed.[1]
  2. First observed
    Core traffic impact begins.[1]
  3. First customer-facing HTTP errors observed.[1]
  4. Mitigation
    Main impact resolved after a known-good feature file is deployed.[1]
  5. Resolved
    All services restored.[1]
  6. Incident report
    Cloudflare publishes its incident report.[1]
  7. Start to: Main impact resolved3 h 10 min
  8. Start to: All services restored5 h 46 min

Related incidents

Links stated by providers or officials carry their evidence label. “Similar failure” is a comparison of documented causes, not a causal link.

  • Similar failure (comparison)

    A valid input triggered a latent bug in a CDN's edge software and took down a large share of the web.[1][2]

Evidence and sources

Grouped by type of evidence. Numbers match the citations above.

Primary: provider and official sources

Published by the provider or agency itself: incident reports, status records, advisories, statements.

  1. 1.
    Cloudflare outage on November 18, 2025 (opens in a new tab)
    Cloudflare · Provider report · Published · Accessed
  2. 2.
    Summary of June 8 outage (opens in a new tab)
    Fastly · Provider report · Published · Accessed

Last verified against these sources on Nov 14, 2026.