Atlassian·

Atlassian outage: a cleanup script deletes 883 customer sites

A script meant to remove an old app was given the IDs of whole customer sites instead, deleting 883 sites; some customers waited up to 14 days for restoration.

ResolvedEnded Apr 18, 2022
Security:
Explicitly ruled out
Started
Apr 5, 2022, 07:38 UTC
All sites restored
Apr 18, 2022
Duration
Not computable from sources
Full recovery
Complete
Geographic scope
Global
Last verified
1 source

What happened

Systems and services affected
Jira productsConfluenceAtlassian AccessOpsgenieStatuspage
Kind of failure
Human error
Sector
Workplace software
Provider report
Published
Status and recovery
Resolved. All customer sites were restored by April 18. No customer lost more than five minutes of data, though 57 customers had some databases restored to an earlier point and were helped to recover that data.[1]

Deletions ran from 07:38 to 08:01 UTC on April 5. Sites were restored progressively; the last were back by April 18.

What caused it?

Root cause: Confirmed

A team ran a script to remove a legacy app (Insight). The team requesting it supplied the IDs of entire cloud sites rather than of the app, and the deletion API gave no warning about whether it was removing a site or an app.[1]

Provider report: Published. Atlassian published a post-incident review on April 29, 2022.[1]

Was it malicious?

Explicitly ruled out

Atlassian: "This was not a result of a cyberattack and there was no unauthorized access to customer data."[1]

How the failure happened

  1. 1TriggerScript run to remove a legacy app (Insight)
  2. 2FailureScript given IDs of entire sites
  3. 3SpreadDeletion API gave no warning
  4. Affected5 services and systems

Evidence for each step

  1. 1
    Script run to remove a legacy app (Insight)[1]
    Starting point.
  2. 2
    Script given IDs of entire sites. Not the IDs of the app to be removed[1]
    Led from step 1: Stated in a provider report or official record.
  3. 3
    Deletion API gave no warning. It did not distinguish a site from an app; 883 sites were deleted[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.

In April 2022, hundreds of companies lost access to their Jira and Confluence workspaces, some for up to two weeks.

Atlassian ran a cleanup job to remove an old add-on. The list it was given contained the identifiers of customers' whole sites instead of the add-on, so the job deleted the sites.

Atlassian had backups, but restoring the sites took until April 18. It says it was not a cyberattack.

Confirmed and unknown

Confirmed

  • Confirmed
    883 sites belonging to 775 customers were deleted.[1]
    Said by: Atlassian · Evidence: Official
  • Confirmed
    The script was given the IDs of entire sites instead of the app to be removed.[1]
    Said by: Atlassian · Evidence: Official
  • Confirmed
    The outage lasted up to 14 days for a subset of customers; all sites were restored by April 18.[1]
    Said by: Atlassian · Evidence: Official
  • Confirmed
    No customer lost more than five minutes of data.[1]
    Said by: Atlassian · Evidence: Official

Denied

  • Denied
    It was a cyberattack.[1]
    Said by: Atlassian · 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 observedApr 5, 2022, 07:38 UTC
  2. AcknowledgedNot documented
  3. MitigationNot documented
  4. RecoveryNot documented
  5. ResolvedApr 18, 2022
  6. Incident reportApr 29, 2022
  1. First observed
    Deletion script begins removing customer sites.[1]
  2. Deletions finish: 883 sites removed.[1]
  3. Resolved
    All customer sites restored.[1]
  4. After the incident ended
    Incident report
    Atlassian publishes its post-incident review.[1]

Related incidents

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

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.
    Post-Incident Review on the Atlassian April 2022 outage (opens in a new tab)
    Atlassian · Provider report · Published · Accessed

Last verified against these sources on Nov 14, 2026.