Detection rules › Panther

Databricks Group Deleted

Status
Experimental
Severity
low
Log types
Databricks.Audit
Tags
Databricks, Impact
Reference
https://github.com/databricks-solutions/cybersec-workspace-detection-app/blob/main/base/detections/event-based/group_deleted.py
Source
github.com/panther-labs/panther-analysis

Detects group deletions in Databricks accounts. While often part of normal cleanup processes, unauthorized group deletions could indicate access control dismantling. Successful deletions are elevated to HIGH severity.

MITRE ATT&CK coverage

TacticTechniques
ImpactT1531 Account Access Removal

Rule body yaml

AnalysisType: rule
Filename: databricks_group_deleted.py
RuleID: "Databricks.Audit.GroupDeleted"
DisplayName: "Databricks Group Deleted"
Enabled: true
Status: Experimental
LogTypes:
  - Databricks.Audit
Tags:
  - Databricks
  - Impact
Reports:
  MITRE ATT&CK:
    - TA0040:T1531 # Account Access Removal
Severity: Low
Description: >
  Detects group deletions in Databricks accounts. While often part of normal cleanup processes,
  unauthorized group deletions could indicate access control dismantling. Successful deletions
  are elevated to HIGH severity.
Runbook: |
  1. Query audit logs for all members who were in this group in the 24 hours before deletion
  2. Check if this group had admin privileges or access to sensitive resources based on past 30 days of activity
  3. Find all group deletions by this actor in the past 30 days to identify bulk deletion patterns
Reference: https://github.com/databricks-solutions/cybersec-workspace-detection-app/blob/main/base/detections/event-based/group_deleted.py
Tests:
  - Name: Successful Group Deletion
    ExpectedResult: true
    Log:
      timestamp: 1234567890000
      serviceName: "accounts"
      actionName: "removeGroup"
      userIdentity:
        email: "admin@example.com"
      sourceIPAddress: "198.51.100.1"
      requestParams:
        targetGroupId: "group-123"
      response:
        statusCode: 200
  - Name: Group Deletion from Permission Assignment
    ExpectedResult: true
    Log:
      timestamp: 1234567890000
      serviceName: "accounts"
      actionName: "removeGroup"
      userIdentity:
        email: "admin@example.com"
      sourceIPAddress: "198.51.100.1"
      requestParams:
        endpoint: "permissionAssignment"
        targetUserName: "developers-group"
      response:
        statusCode: 200
  - Name: Failed Group Deletion
    ExpectedResult: true
    Log:
      timestamp: 1234567890000
      serviceName: "accounts"
      actionName: "removeGroup"
      userIdentity:
        email: "admin@example.com"
      requestParams:
        targetGroupId: "group-456"
      response:
        statusCode: 403
  - Name: Wrong Service
    ExpectedResult: false
    Log:
      timestamp: 1234567890000
      serviceName: "workspace"
      actionName: "removeGroup"
      userIdentity:
        email: "admin@example.com"

Detection logic

Condition

not (userAgent contains "Databricks-Service/driver" or userAgent contains "Databricks-Runtime" or userAgent contains "Delta-Sharing-SparkStructuredStreaming" or userAgent contains "RawDBHttpClient" or userAgent contains "mlflow-python" or userAgent contains "obsSDK-scala" or userAgent contains "wsfs" or userAgent contains "feature-store" or requestParams.path contains "/telemetry" or requestParams.path contains "/delta-commit" or requestParams.path contains "/health" or requestParams.path contains "/metrics" or requestParams.path contains "/status" or userIdentity.email regex_match "[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}")
serviceName eq "accounts"

This rule also runs imperative logic the parser cannot express as a filter; the conditions above are the structured part it could extract.

Exclusions

Top-level NOT(...) conjuncts: predicates this rule actively suppresses.

FieldKindExcluded values
requestParams.pathcontains/delta-commit
requestParams.pathcontains/health
requestParams.pathcontains/metrics
requestParams.pathcontains/status
requestParams.pathcontains/telemetry
userAgentcontainsDatabricks-Runtime
userAgentcontainsDatabricks-Service/driver
userAgentcontainsDelta-Sharing-SparkStructuredStreaming
userAgentcontainsRawDBHttpClient
userAgentcontainsfeature-store
userAgentcontainsmlflow-python
userAgentcontainsobsSDK-scala
userAgentcontainswsfs
userIdentity.emailregex_match[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}

Indicators

Each row is a field, operator, and value that the rule matches. The corpus column counts how many other rules in the catalog look for the same combination: high numbers point to widely-used, community-vetted indicators. Blank or 1 shows that the indicator is specific to this rule.

FieldKindValues
serviceNameeq
  • accounts

Output fields

Fields the rule emits when it matches. Chronicle authors list these in the outcome block; they appear on the detection and $risk_score drives alerting. Sentinel / Defender XDR rules build them up through project / summarize / extend stages. Sentinel maps these into alert fields via entityMappings and customDetails; Defender XDR custom detections surface them as alert fields directly.

FieldSource
emailuserIdentity.email