Google sign-in outage: a quota error stops Gmail, Workspace and Cloud logins for 47 minutes

A half-finished move to a new quota system reported Google's user ID service as using nothing, so automation cut its storage quota and signed-in Google services failed worldwide.

ResolvedEnded Dec 14, 2020
Security:
No determination
Started
Dec 14, 2020, 11:46 UTC
Impact ended
Dec 14, 2020, 12:33 UTC
Duration
47 min
Full recovery
Complete
Geographic scope
Global
Last verified
1 source

What happened

Systems and services affected
Google sign-in (User ID Service)GmailGoogle CalendarGoogle Workspace appsCloud ConsoleBigQueryCloud StorageGoogle Kubernetes Engine
Kind of failure
Configuration change
Sector
Cloud software
Provider report
Published
Status and recovery
Resolved. Google's issue summary gives an impact window of 03:46 to 04:33 US/Pacific, 47 minutes.[1]

Google gives times in US/Pacific (03:46 to 04:33). Its incident header lists a wider window (04:07 to 06:23) and its preliminary statement 03:45 to 04:35; the issue summary gives 47 minutes.

What caused it?

Root cause: Confirmed

During a migration of the User ID Service to a new quota system, parts of the old system remained in place and "incorrectly reported the usage for the User ID Service as 0." Automated quota management then reduced the account database's quota, blocking the Paxos leader from writing.[1]

Provider report: Published. Google posted an issue summary on December 18, 2020, corrected on December 22.[1]

Was it malicious?

No determination

Google attributes the outage to quota management and a migration error. Its summary does not mention an attack.[1]

How the failure happened

  1. 1TriggerMigration to a new quota system
  2. 2FailureUsage reported as 0
  3. 3SpreadAccount database quota reduced
  4. Affected8 services and systems

Evidence for each step

  1. 1
    Migration to a new quota system. Parts of the old system remained in place[1]
    Starting point.
  2. 2
    Usage reported as 0. For the User ID Service[1]
    Led from step 1: Stated in a provider report or official record.
  3. 3
    Account database quota reduced. The Paxos leader could not write[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.

On December 14, 2020, people around the world could not use Gmail, Google Calendar and other signed-in Google services for about 47 minutes.

Google was moving the system that tracks sign-ins to a new way of managing resource limits. Leftovers from the old system reported that it was using nothing, so automation took away the storage it needed.

Without storage to write to, the sign-in system could not work, and every service that needs to know who you are failed with it.

Confirmed and unknown

Confirmed

  • Confirmed
    Parts of the old quota system "incorrectly reported the usage for the User ID Service as 0."[1]
    Said by: Google · Evidence: Official
  • Confirmed
    Automated quota management reduced the account database's quota, blocking the Paxos leader from writing.[1]
    Said by: Google · Evidence: Official
  • Confirmed
    Impact lasted 47 minutes.[1]
    Said by: Google · Evidence: Official

Sources disagree

  • Disputed
    The incident window is given as 03:46–04:33, 03:45–04:35 or 04:07–06:23 US/Pacific in different parts of Google's own record.[1]
    Said by: Google · 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 observedDec 14, 2020, 11:46 UTC
  2. AcknowledgedDec 14, 2020
  3. MitigationNot documented
  4. RecoveryNot documented
  5. ResolvedDec 14, 2020, 12:33 UTC
  6. Incident reportDec 18, 2020
  1. First observed
    Authentication failures begin (03:46 US/Pacific).[1]
  2. Resolved
    Impact ends (04:33 US/Pacific).[1]
  3. Acknowledged
    Google posts a preliminary statement.[1]
  4. After the incident ended
    Incident report
    Google posts its issue summary (corrected December 22).[1]
  5. Start to: Impact ended47 min

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.
    Google Cloud Infrastructure Components Incident #20013 (opens in a new tab)
    Google Cloud · Provider report · Published · Accessed
    Issue summary posted December 18, 2020 and corrected December 22; preliminary statement December 14.

Last verified against these sources on Nov 14, 2026.