1. Resources
  2. Detection rules
  3. PingOne Environment, Population or Users Deleted

PingOne Environment, Population or Users Deleted detection rules

An attacker with PingOne administrator rights deletes an environment, empties and deletes a population, removes many users at once or strips other administrators of their roles, to destroy the identity store or lock its owners out.

Platform
Ping Identity
Severity
High
Log source
PingOne audit activities
Last updated
Tactic
Impact
Rule formats
Sigmaand queries for: SplunkMicrosoft Sentinel
What an attacker needs
The Organization Admin role or an organization-level Environment Admin role for environments, or a role that can manage users and role assignments.
Impact
Applications that rely on the environment stop authenticating users, deleted users and sandbox environments are gone for good, and administrators lose access.
Response
Restore a soft-deleted production environment within the 30 day waiting period, suspend the actor and give removed administrator roles back.
Mitigation
Keep few Organization Admins with strong MFA, configure long-standing environments as PRODUCTION and keep configuration and users recoverable.
Action types
ROLE_ASSIGNMENT.DELETEDENVIRONMENT.DEMOTEDENVIRONMENT.DELETEDPOPULATION.DELETEDORGANIZATION.DELETEDENVIRONMENT.UPDATEDUSER.DELETED

How the attack works

An environment is the unit that holds everything in PingOne: users, populations, applications, policies and keys. Deleting one ends sign-on for every application that depends on it. Ping builds in a brake for environments of the type PRODUCTION: they cannot be deleted immediately. They go through a soft delete, the status DELETE_PENDING, in which the environment is not operational, and a waiting period of 30 days before the hard delete becomes possible. Within that period an Organization Admin, or an Environment Admin at the organization level, can restore it. Environments of the type SANDBOX have no brake: they are deleted immediately and cannot be restored, and Ping documents that demoting a production environment to a sandbox is refused unless a feature flag allows it for the organization.

Destruction can also happen below the environment. Users can be deleted one request at a time through the API, a population can be deleted once it holds no users and no groups, and an administrator who removes the role assignments of the other administrators leaves nobody to undo the rest. PingOne records these operations as ENVIRONMENT.DELETED, ENVIRONMENT.DEMOTED, POPULATION.DELETED, ORGANIZATION.DELETED, USER.DELETED and ROLE_ASSIGNMENT.DELETED audit activities. The page follows Ping's documentation of these operations: no public incident report describes them for PingOne at the audit event level.

Log source: the rules read PingOne audit activities. Include the Environment, Population, Organization, User and Role Assignment event types in the PingOne webhook that feeds the SIEM, or poll the Audit Activities API. Do this for the environment that holds the administrators first, because role assignments and environment operations are made from there.

What an attacker needs

  • For environments, the Organization Admin role or the Environment Admin role at the organization level, which Ping names for deleting and restoring production environments.
  • For users, populations and role assignments, a role that can manage users, populations or role assignments in the environment.

Impact

Applications that rely on the environment stop authenticating users the moment it is soft-deleted, and a sandbox environment and deleted users are gone for good. Removing the role assignments of other administrators delays the response by exactly the time it takes to find someone who can still sign on with enough rights. Recovery depends on what exists outside PingOne: a source directory to provision users from again, and configuration kept as code.

Detection rules

PingOne Administrator Role Assignment Deleted

Detects an administrator role assignment being removed in PingOne (audit activity ROLE_ASSIGNMENT.DELETED, described by Ping as Role Assignment Deleted) from a user, group, worker application or gateway. Removing the roles of the other administrators is how an attacker who holds one administrator account keeps its owners from undoing what follows, and it is also the normal end of every offboarding. The documented attributes do not include the role name or its scope as separate fields, so the rule fires on every removal: read both from the activity description, and check who still holds administrator roles in the environment and at the organization level.

LevelLowStatusExperimentalLog sourcePingOne audit activities
title: PingOne Administrator Role Assignment Deleted
id: 4affa30b-c5be-4932-b537-6a65842f1c07
status: experimental
description: >-
  Detects an administrator role assignment being removed in PingOne (audit
  activity ROLE_ASSIGNMENT.DELETED, described by Ping as Role Assignment
  Deleted) from a user, group, worker application or gateway. Removing the
  roles of the other administrators is how an attacker who holds one
  administrator account keeps its owners from undoing what follows, and it
  is also the normal end of every offboarding. The documented attributes do
  not include the role name or its scope as separate fields, so the rule
  fires on every removal: read both from the activity description, and check
  who still holds administrator roles in the environment and at the
  organization level.
references:
  - https://attack.mitre.org/techniques/T1531/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/roles/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.impact
  - attack.t1531
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type: ROLE_ASSIGNMENT.DELETED
  condition: selection
falsepositives:
  - Offboarding, role reviews and the removal of temporary access. Confirm with the request, and look closer when one actor removes the roles of several administrators within a short time or removes roles at the organization level.
  - Role assignments that disappear together with a deleted application, group or gateway, if your environment writes the activity for them.
level: low

False positives

  • Offboarding, role reviews and the removal of temporary access. Confirm with the request, and look closer when one actor removes the roles of several administrators within a short time or removes roles at the organization level.
  • Role assignments that disappear together with a deleted application, group or gateway, if your environment writes the activity for them.

PingOne Environment Demoted

Detects the audit activity ENVIRONMENT.DEMOTED, which Ping's event list describes as Environment Demoted. Ping documents two environment types and that an environment can be promoted or demoted between them. The type decides how deletion works: a PRODUCTION environment must pass through a soft delete and a waiting period of 30 days and can be restored, while a SANDBOX environment is deleted immediately and cannot be restored. Ping also documents that demoting a production environment to a sandbox is refused unless a feature flag allows it for the organization. A demotion therefore removes the protection against immediate deletion and should be rare. Ping's event list names the activity but shows no attributes for it, so find the actor in the activities around it if the activity itself does not carry one. Check which environment was demoted, by whom, and whether a deletion follows.

LevelMediumStatusExperimentalLog sourcePingOne audit activities
title: PingOne Environment Demoted
id: 202d52d8-6158-40af-96fc-612939db2822
status: experimental
description: >-
  Detects the audit activity ENVIRONMENT.DEMOTED, which Ping's event list
  describes as Environment Demoted. Ping documents two environment types and
  that an environment can be promoted or demoted between them. The type
  decides how deletion works: a PRODUCTION environment must pass through a
  soft delete and a waiting period of 30 days and can be restored, while a
  SANDBOX environment is deleted immediately and cannot be restored. Ping
  also documents that demoting a production environment to a sandbox is
  refused unless a feature flag allows it for the organization. A demotion
  therefore removes the protection against immediate deletion and should be
  rare. Ping's event list names the activity but shows no attributes for it,
  so find the actor in the activities around it if the activity itself does
  not carry one. Check which environment was demoted, by whom, and whether a
  deletion follows.
references:
  - https://attack.mitre.org/techniques/T1485/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/environments/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.impact
  - attack.t1485
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type: ENVIRONMENT.DEMOTED
  condition: selection
falsepositives:
  - An environment that was promoted by mistake and is set back to a sandbox, in organizations where the demotion is allowed. Confirm with the owner of the environment.
level: medium

False positives

  • An environment that was promoted by mistake and is set back to a sandbox, in organizations where the demotion is allowed. Confirm with the owner of the environment.

PingOne Environment, Population or Organization Deleted

Detects the deletion of a container of the PingOne directory (audit activities ENVIRONMENT.DELETED, POPULATION.DELETED and ORGANIZATION.DELETED, described by Ping as Environment Deleted, Population Deleted and Organization Deleted). An environment holds the users, applications, policies and keys that sign-on depends on. Ping documents that a production environment is first soft-deleted, which makes it not operational, and can be restored for 30 days, while a sandbox environment is deleted immediately and cannot be restored. A population can only be deleted once it holds no users and no groups, so its deletion is the last step of removing a part of the directory. Ping does not document whether the soft delete of a production environment writes ENVIRONMENT.DELETED or ENVIRONMENT.UPDATED: a separate rule of this page reads the update.

LevelHighStatusExperimentalLog sourcePingOne audit activities
title: PingOne Environment, Population or Organization Deleted
id: f63b4367-49d2-4215-a220-692faf7c4f47
status: experimental
description: >-
  Detects the deletion of a container of the PingOne directory (audit
  activities ENVIRONMENT.DELETED, POPULATION.DELETED and ORGANIZATION.DELETED,
  described by Ping as Environment Deleted, Population Deleted and
  Organization Deleted). An environment holds the users, applications,
  policies and keys that sign-on depends on. Ping documents that a production
  environment is first soft-deleted, which makes it not operational, and can
  be restored for 30 days, while a sandbox environment is deleted immediately
  and cannot be restored. A population can only be deleted once it holds no
  users and no groups, so its deletion is the last step of removing a part of
  the directory. Ping does not document whether the soft delete of a
  production environment writes ENVIRONMENT.DELETED or ENVIRONMENT.UPDATED:
  a separate rule of this page reads the update.
references:
  - https://attack.mitre.org/techniques/T1485/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/environments/index.html
  - https://developer.pingidentity.com/pingone-api/platform/populations/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.impact
  - attack.t1485
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type:
      - ENVIRONMENT.DELETED
      - POPULATION.DELETED
      - ORGANIZATION.DELETED
  condition: selection
falsepositives:
  - Cleanup of sandbox and trial environments and of populations created for a test. Add a filter on actors.user.name or actors.client.name for the identities that create and remove test environments, after you have reviewed them, and keep the alert for everything else.
level: high

False positives

  • Cleanup of sandbox and trial environments and of populations created for a test. Add a filter on actors.user.name or actors.client.name for the identities that create and remove test environments, after you have reviewed them, and keep the alert for everything else.

PingOne Environment Updated

Detects the audit activity ENVIRONMENT.UPDATED, which Ping's event list describes as Environment Updated. The rule exists because of the soft delete. Ping documents that a production environment is deleted by setting its status to DELETE_PENDING with the Update Environment Status request, after which the environment is not operational, and that it is restored by setting the status back to ACTIVE. Ping does not document whether that status change is written as ENVIRONMENT.UPDATED or as ENVIRONMENT.DELETED, so this rule reads the update and another rule of the page the deletion. The rule matches on the activity alone: it also fires when an environment is renamed or its license or products change, and the documented attributes do not say which property changed. Read the status of the environment after each match.

LevelLowStatusExperimentalLog sourcePingOne audit activities
title: PingOne Environment Updated
id: d291f0e6-ad2b-4a3d-8fd4-d273cd7fee8d
status: experimental
description: >-
  Detects the audit activity ENVIRONMENT.UPDATED, which Ping's event list
  describes as Environment Updated. The rule exists because of the soft
  delete. Ping documents that a production environment is deleted by setting
  its status to DELETE_PENDING with the Update Environment Status request,
  after which the environment is not operational, and that it is restored by
  setting the status back to ACTIVE. Ping does not document whether that
  status change is written as ENVIRONMENT.UPDATED or as ENVIRONMENT.DELETED,
  so this rule reads the update and another rule of the page the deletion.
  The rule matches on the activity alone: it also fires when an environment
  is renamed or its license or products change, and the documented
  attributes do not say which property changed. Read the status of the
  environment after each match.
references:
  - https://attack.mitre.org/techniques/T1485/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/environments/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.impact
  - attack.t1485
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type: ENVIRONMENT.UPDATED
  condition: selection
falsepositives:
  - Renaming an environment, changing its description, license or products, and restoring a soft-deleted environment. Environments change rarely, so read the status after each activity instead of excluding actors.
level: low

False positives

  • Renaming an environment, changing its description, license or products, and restoring a soft-deleted environment. Environments change rarely, so read the status after each activity instead of excluding actors.

Many PingOne Users Deleted Through One Client

Counts, per client, the PingOne users that were deleted (audit activity USER.DELETED). Twenty or more through one client within 10 minutes are more than an administrator removing single accounts and fit the bulk removal of identities, either to destroy the directory or to lock a group of people out. Ping documents the client actor as the client the actor used to perform the action. For a person that is the application they work in, such as the administrator console, and for deletions made through the API with client credentials it is the worker application itself. Counting per client covers both. The administrator, where there is one, is in actors.user.name of the counted activities. The count and the window are starting points to tune.

Fires when one actors.client.id has at least 20 matching events within 10 minutes.

Threshold ruleLevelHighStatusExperimentalLog sourcePingOne audit activities
title: PingOne User Deleted (Bulk Deletion Base Event)
id: 8152e6c7-6d56-44be-8ab2-9da6362a73b2
name: pingone_user_deleted
status: experimental
description: >-
  Base event of the threshold rule in this file. The audit activity
  USER.DELETED, which Ping's event list describes as User Deleted.
logsource:
  product: pingone
  service: audit
detection:
  selection:
    action.type: USER.DELETED
  condition: selection
level: low
---
title: Many PingOne Users Deleted Through One Client
id: 7264153b-49b5-42df-8986-a0c4b45513d5
status: experimental
description: >-
  Counts, per client, the PingOne users that were deleted (audit activity
  USER.DELETED). Twenty or more through one client within 10 minutes are more
  than an administrator removing single accounts and fit the bulk removal of
  identities, either to destroy the directory or to lock a group of people
  out. Ping documents the client actor as the client the actor used to
  perform the action. For a person that is the application they work in, such
  as the administrator console, and for deletions made through the API with
  client credentials it is the worker application itself. Counting per client
  covers both. The administrator, where there is one, is in actors.user.name
  of the counted activities. The count and the window are starting points to
  tune.
references:
  - https://attack.mitre.org/techniques/T1531/
  - https://developer.pingidentity.com/pingone-api/platform/reference/audit-reporting-events.html
  - https://developer.pingidentity.com/pingone-api/platform/audit-activities.html
  - https://developer.pingidentity.com/pingone-api/platform/users/users-1/index.html
author: Forestall
date: 2026-11-18
tags:
  - attack.impact
  - attack.t1531
correlation:
  type: event_count
  rules:
    - pingone_user_deleted
  group-by:
    - actors.client.id
  timespan: 10m
  condition:
    gte: 20
falsepositives:
  - Provisioning and lifecycle jobs that remove leavers or expired accounts in batches, and the planned cleanup of a test population. Exclude the client IDs of those jobs after you have reviewed them, or give them a higher count.
  - Customer-facing environments where many users delete their own account through a self-service application in the same period.
  - The values of 20 deletions and 10 minutes are starting points, not tested limits.
level: high

False positives

  • Provisioning and lifecycle jobs that remove leavers or expired accounts in batches, and the planned cleanup of a test population. Exclude the client IDs of those jobs after you have reviewed them, or give them a higher count.
  • Customer-facing environments where many users delete their own account through a self-service application in the same period.
  • The values of 20 deletions and 10 minutes are starting points, not tested limits.

Detection logic

Patterns that need more than one event, a baseline or knowledge of your own environment. Each is a hypothesis to build and tune in your SIEM, not a ready rule.

Many users deleted and their population deleted afterwards

Detection hypothesisSequence of events

Ping documents that a population can only be deleted once it contains no users and no groups. Someone who wants a population gone therefore deletes its users first. A burst of user deletions followed by the deletion of a population by the same actor is the removal of a whole part of the directory, not housekeeping on single accounts.

  1. Count USER.DELETED activities per actor (actors.client.id, and actors.user.id where a user is the actor) in windows of one hour.
  2. Look for a POPULATION.DELETED activity by the same actor within 24 hours after a burst.
  3. Read resources.population.id of the deleted users, which Ping lists among the attributes of the activity, and compare it with the deleted population named in resources.id of the population activity.

EventsUSER.DELETEDPOPULATION.DELETED

LimitsA planned decommissioning of a test population produces the same order, so confirm with the change record. The windows are starting points. Users can also be moved to another population instead of deleted, which empties a population without any deletion.

Triage

  • Identify the actor (actors.user.name or actors.client.name), the source address and what was deleted, named in the activity's resources, and check for an approved decommissioning or cleanup.
  • For an environment, read its status and type in the console of the organization. A production environment in DELETE_PENDING can be restored, so note the time left.
  • For deleted users, count them, find out which population they were in and whether a provisioning job, a lifecycle rule or a bulk cleanup explains the number.
  • For a removed role assignment, list who still holds administrator roles in the environment and at the organization level, and whether the actor removed others while keeping their own.
  • Review what the actor did before the deletion, such as new role assignments, new worker applications, or changes to webhooks that would have reported the activity.

Response

Once the attack is confirmed:

  1. Restore a soft-deleted production environment by setting its status back to ACTIVE, which needs an Organization Admin or an organization-level Environment Admin and the license ID.
  2. Suspend the actor, end its sessions and remove its roles. For a worker application, disable it and replace its secret.
  3. Give the removed role assignments back to the administrators who lost them, and verify each of them through a channel outside PingOne first.
  4. Provision deleted users again from the source system and tell the application owners which accounts lost their history, group memberships and MFA devices.
  5. Review every environment of the organization for further deletions, demotions and role changes by the same actor.

Mitigation

  • Keep the number of Organization Admins and organization-level Environment Admins small, protect them with phishing-resistant MFA, and keep at least two so that one cannot lock the other out unnoticed.
  • Configure every long-standing environment as PRODUCTION, as Ping advises, so that a deletion passes through the 30 day waiting period.
  • Keep the environment's configuration as code or as regular exports, and keep users in a source system from which they can be provisioned again.
  • Limit the roles that can delete users, and the worker applications that hold them, since they can remove users in bulk through the API.
  • Alert on these event types and make sure the alert reaches people who do not depend on PingOne to sign on to the alerting tool.

Frequently asked questions

Can a deleted PingOne environment be restored?

It depends on its type. Ping documents that a PRODUCTION environment is first soft-deleted: its status becomes `DELETE_PENDING`, it is not operational, and for 30 days it can be restored by setting the status back to ACTIVE. After the waiting period it can be deleted for good, and Ping states that such environments are deleted automatically. A SANDBOX environment is deleted immediately and cannot be restored.

Which audit activity does the soft delete of a production environment write?

Ping does not say. The soft delete is made with the Update Environment Status request, and the event list has `ENVIRONMENT.DELETED` and `ENVIRONMENT.UPDATED` without describing when each is written. The rules here read both, with a low level for the update because renames and license changes use it too. Soft-delete a test environment once to see what your log shows.

Why is the demotion of an environment worth an alert?

The waiting period protects only PRODUCTION environments. A sandbox is deleted immediately and for good. Ping documents that demoting a production environment to a sandbox is refused unless a feature flag allows it for the organization, so an `ENVIRONMENT.DEMOTED` activity for an environment that mattered means the brake was taken off, whatever happens next.

Why does the user deletion rule count per client and not per administrator?

Bulk deletion is usually done through the API by a worker application, which appears in the activity as the client. An administrator in the console appears as the user, with the console as the client. Counting per client covers both cases with one rule. The administrator's name, where there is one, is in `actors.user.name` of the counted activities.

References

We respect your privacy

We use cookies to keep this site secure and working properly. With your permission, we also use optional cookies to understand usage and improve the experience. Cookie Policy

You can change your choice at any time.