{"source_checked_on":"2026-10-19","schema_version":1,"key":"enterprise/T1578.004","id":"T1578.004","name":"Revert Cloud Instance","domain":"enterprise","reviewed_on":"2026-09-28","source_url":"https://attack.mitre.org/techniques/T1578/004","source_file":{"file":"enterprise.json","url":"https://raw.githubusercontent.com/mitre-attack/attack-stix-data/20664a889568dce884a1dac0ec7b12ae3a199f51/enterprise-attack/enterprise-attack-20.0.json","sha256":"81113ba98afa2928f3dcff62022539f7ce30c315c32e14422f11ff95ce534352","bytes":55271014},"source_object_sha256":"588911820a4ed5078ece11bc89e769785a26d83cf540cea5a7c415b1707d8b55","description":"An adversary may revert changes made to a cloud instance after they have performed malicious activities in attempt to evade detection and remove evidence of their presence. In highly virtualized environments, such as cloud-based infrastructure, this may be accomplished by restoring virtual machine (VM) or data storage snapshots through the cloud management dashboard or cloud APIs.\n\nAnother variation of this technique is to utilize temporary storage attached to the compute instance. Most cloud providers provide various types of storage including persistent, local, and/or ephemeral, with the ephemeral types often reset upon stop/restart of the VM.(Citation: Tech Republic - Restore AWS Snapshots)(Citation: Google - Restore Cloud Snapshot)","references":[{"source_name":"mitre-attack","url":"https://attack.mitre.org/techniques/T1578/004","external_id":"T1578.004"},{"source_name":"Google - Restore Cloud Snapshot","description":"Google. (2019, October 7). Restoring and deleting persistent disk snapshots. Retrieved October 8, 2019.","url":"https://cloud.google.com/compute/docs/disks/restore-and-delete-snapshots"},{"source_name":"Tech Republic - Restore AWS Snapshots","description":"Hardiman, N.. (2012, March 20). Backing up and restoring snapshots on Amazon EC2 machines. Retrieved October 8, 2019.","url":"https://www.techrepublic.com/blog/the-enterprise-cloud/backing-up-and-restoring-snapshots-on-amazon-ec2-machines/"}],"platforms":["IaaS"],"environments":["cloud"],"visibility":"source_mapped","visibility_note":null,"parent":{"key":"enterprise/T1578","id":"T1578","name":"Modify Cloud Compute Infrastructure","domain":"enterprise","level":"technique","parent_id":null,"tactics":["defense-impairment"],"platforms":["IaaS"],"environments":["cloud"],"page":"techniques/enterprise/T1578/","constraints":["isolated_authorized_lab","procedure_review_required","telemetry_configuration_required","disposable_tenant_and_cost_controls","potential_disruption_requires_individual_review"],"telemetry_tags":["Cloud Service Metadata","Instance Start","Instance Stop","Snapshot Creation","Snapshot Deletion","Snapshot Modification","Volume Creation","Volume Deletion","Volume Modification"],"telemetry_references":[{"id":"DC0070","name":"Cloud Service Metadata","tag":"Cloud Service Metadata","page":"telemetry/DC0070/","kind":"attack_component"},{"id":"DC0080","name":"Instance Start","tag":"Instance Start","page":"telemetry/DC0080/","kind":"attack_component"},{"id":"DC0089","name":"Instance Stop","tag":"Instance Stop","page":"telemetry/DC0089/","kind":"attack_component"},{"id":"DC0057","name":"Snapshot Creation","tag":"Snapshot Creation","page":"telemetry/DC0057/","kind":"attack_component"},{"id":"DC0049","name":"Snapshot Deletion","tag":"Snapshot Deletion","page":"telemetry/DC0049/","kind":"attack_component"},{"id":"DC0058","name":"Snapshot Modification","tag":"Snapshot Modification","page":"telemetry/DC0058/","kind":"attack_component"},{"id":"DC0097","name":"Volume Creation","tag":"Volume Creation","page":"telemetry/DC0097/","kind":"attack_component"},{"id":"DC0098","name":"Volume Deletion","tag":"Volume Deletion","page":"telemetry/DC0098/","kind":"attack_component"},{"id":"DC0092","name":"Volume Modification","tag":"Volume Modification","page":"telemetry/DC0092/","kind":"attack_component"}],"detections":{"page":"detections/enterprise/T1578/","status":"rule_source_available","counts":{"sigma":1,"atlas_basic":0,"atlas_anomaly":0,"strategies":1,"analytics":1,"live_validated":0},"anomaly_types":[]},"detection_status":"rule_source_available","anomaly_tags":[],"tool_tags":[],"tool_references":[],"visibility":"source_mapped","telemetry_status":"mapped","classification":"cannot_simulate_yet","candidate_count":0,"implementation":"not_integrated","methods":[]},"children":[],"telemetry":[{"id":"DC0073","name":"Instance Modification","page":"telemetry/DC0073/","kind":"attack_component","category":"Cloud control plane","description":"Changes to a compute instance configuration.","collection_focus":"Collect compute update/modify APIs and before/after state, including attached role, network interfaces and launch parameters where exposed.","providers":[{"name":"AWS CloudTrail","scope":"Management-plane API activity; supported resource/data events require separate selectors.","source_id":"cloudtrail"},{"name":"Azure Monitor Activity Log","scope":"Azure resource control-plane operations; resource data-plane logs are separate.","source_id":"azure"},{"name":"Google Cloud Audit Logs","scope":"Admin Activity and other categories; configure Data Access where required.","source_id":"gcp"}],"configuration":["In the isolated account/project/subscription, route administrative audit logs to a protected destination with retention and access controls.","Select the relevant service, regions and read/write operations. Confirm whether the API below is a management or data event in the provider documentation.","Preserve actor/session, resource, request ID, service, operation, region and error details. Check delivery latency and avoid logging credentials or sensitive request bodies."],"configuration_example":null,"required_fields":["instance_id","attribute","before","after"],"example":{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0073","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","attribute":"instance_type","before":"small-lab","after":"medium-lab"}},"example_note":"Project-normalized synthetic JSON, not a native provider log, captured event, attack verdict or validated detection. A parser/adapter is required for a real SIEM. Example names, IPs and values are fictional.","limitations":"AWS, Azure and Google expose different APIs and event schemas. An accepted API request does not prove the final resource state; join asynchronous outcomes and inventory.","sources":[{"id":"cloudtrail","title":"AWS CloudTrail event categories","url":"https://docs.aws.amazon.com/awscloudtrail/latest/userguide/cloudtrail-events.html"},{"id":"ec2","title":"AWS EC2 API logging with CloudTrail","url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/monitor-with-cloudtrail.html"},{"id":"azure","title":"Azure Monitor Activity Log","url":"https://learn.microsoft.com/en-us/azure/azure-monitor/fundamentals/activity-log"},{"id":"gcp","title":"Google Cloud Audit Logs","url":"https://docs.cloud.google.com/logging/docs/audit"}],"association":"attack_component","anomaly_design":"Per account, role, region and API: rare actions, new role-resource relationships, enumeration bursts and configuration changes. Separate deployment automation and organization-level administrative activity."},{"id":"DC0080","name":"Instance Start","page":"telemetry/DC0080/","kind":"attack_component","category":"Cloud control plane","description":"A request to start a stopped compute instance and its resulting state.","collection_focus":"Collect start APIs and instance-state transitions with request IDs; distinguish accepted from completed.","providers":[{"name":"AWS CloudTrail","scope":"Management-plane API activity; supported resource/data events require separate selectors.","source_id":"cloudtrail"},{"name":"Azure Monitor Activity Log","scope":"Azure resource control-plane operations; resource data-plane logs are separate.","source_id":"azure"},{"name":"Google Cloud Audit Logs","scope":"Admin Activity and other categories; configure Data Access where required.","source_id":"gcp"}],"configuration":["In the isolated account/project/subscription, route administrative audit logs to a protected destination with retention and access controls.","Select the relevant service, regions and read/write operations. Confirm whether the API below is a management or data event in the provider documentation.","Preserve actor/session, resource, request ID, service, operation, region and error details. Check delivery latency and avoid logging credentials or sensitive request bodies."],"configuration_example":null,"required_fields":["instance_id","action","state_before","state_after"],"example":{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0080","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","action":"start","state_before":"stopped","state_after":"pending"}},"example_note":"Project-normalized synthetic JSON, not a native provider log, captured event, attack verdict or validated detection. A parser/adapter is required for a real SIEM. Example names, IPs and values are fictional.","limitations":"AWS, Azure and Google expose different APIs and event schemas. An accepted API request does not prove the final resource state; join asynchronous outcomes and inventory.","sources":[{"id":"cloudtrail","title":"AWS CloudTrail event categories","url":"https://docs.aws.amazon.com/awscloudtrail/latest/userguide/cloudtrail-events.html"},{"id":"ec2","title":"AWS EC2 API logging with CloudTrail","url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/monitor-with-cloudtrail.html"},{"id":"azure","title":"Azure Monitor Activity Log","url":"https://learn.microsoft.com/en-us/azure/azure-monitor/fundamentals/activity-log"},{"id":"gcp","title":"Google Cloud Audit Logs","url":"https://docs.cloud.google.com/logging/docs/audit"}],"association":"attack_component","anomaly_design":"Per account, role, region and API: rare actions, new role-resource relationships, enumeration bursts and configuration changes. Separate deployment automation and organization-level administrative activity."},{"id":"DC0089","name":"Instance Stop","page":"telemetry/DC0089/","kind":"attack_component","category":"Cloud control plane","description":"Stopping a compute instance without necessarily deleting it.","collection_focus":"Collect stop API requests and resulting state; correlate with approved maintenance and resource health.","providers":[{"name":"AWS CloudTrail","scope":"Management-plane API activity; supported resource/data events require separate selectors.","source_id":"cloudtrail"},{"name":"Azure Monitor Activity Log","scope":"Azure resource control-plane operations; resource data-plane logs are separate.","source_id":"azure"},{"name":"Google Cloud Audit Logs","scope":"Admin Activity and other categories; configure Data Access where required.","source_id":"gcp"}],"configuration":["In the isolated account/project/subscription, route administrative audit logs to a protected destination with retention and access controls.","Select the relevant service, regions and read/write operations. Confirm whether the API below is a management or data event in the provider documentation.","Preserve actor/session, resource, request ID, service, operation, region and error details. Check delivery latency and avoid logging credentials or sensitive request bodies."],"configuration_example":null,"required_fields":["instance_id","action","state_before","state_after"],"example":{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0089","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","action":"stop","state_before":"running","state_after":"stopping"}},"example_note":"Project-normalized synthetic JSON, not a native provider log, captured event, attack verdict or validated detection. A parser/adapter is required for a real SIEM. Example names, IPs and values are fictional.","limitations":"AWS, Azure and Google expose different APIs and event schemas. An accepted API request does not prove the final resource state; join asynchronous outcomes and inventory.","sources":[{"id":"cloudtrail","title":"AWS CloudTrail event categories","url":"https://docs.aws.amazon.com/awscloudtrail/latest/userguide/cloudtrail-events.html"},{"id":"ec2","title":"AWS EC2 API logging with CloudTrail","url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/monitor-with-cloudtrail.html"},{"id":"azure","title":"Azure Monitor Activity Log","url":"https://learn.microsoft.com/en-us/azure/azure-monitor/fundamentals/activity-log"},{"id":"gcp","title":"Google Cloud Audit Logs","url":"https://docs.cloud.google.com/logging/docs/audit"}],"association":"attack_component","anomaly_design":"Per account, role, region and API: rare actions, new role-resource relationships, enumeration bursts and configuration changes. Separate deployment automation and organization-level administrative activity."}],"detection":{"key":"enterprise/T1578.004","id":"T1578.004","name":"Revert Cloud Instance","domain":"enterprise","platforms":["IaaS"],"environments":["cloud"],"source_url":"https://attack.mitre.org/techniques/T1578/004","technique_page":"techniques/enterprise/T1578.004/","page":"detections/enterprise/T1578.004/","status":"guidance_only","counts":{"sigma":0,"atlas_basic":0,"atlas_anomaly":0,"strategies":1,"analytics":1,"live_validated":0},"sigma_rules":[],"atlas_basic":[],"atlas_anomaly":[],"anomaly_types":[],"strategies":[{"id":"DET0337","name":"Detection Strategy for Modify Cloud Compute Infrastructure: Revert Cloud Instance","url":"https://attack.mitre.org/detectionstrategies/DET0337","stix_id":"x-mitre-detection-strategy--f5ee584b-bbbd-481a-af63-c49166b8b1a8","relationship_id":"relationship--e4b4bdd3-07b3-4d99-b054-330260bfec0a","description":"","analytics":[{"id":"AN0953","name":"Analytic 0953","guidance":"Defenders can detect suspicious reversion of cloud compute instances by monitoring for unusual snapshot restores, rollback actions, or ephemeral storage resets that occur outside expected administrative workflows. From a defender’s perspective, relevant detection chains include: a snapshot restore triggered by a new or rarely used account, a sequence of snapshot creation immediately followed by a restore and instance start, or rollbacks performed from anomalous geographic or network locations. These patterns may indicate attempts to remove forensic evidence or re-establish a clean execution state for persistence.","url":"https://attack.mitre.org/detectionstrategies/DET0337#AN0953","platforms":["IaaS"],"tuning":[{"field":"UserContext","description":"Identity of the user or service account performing rollback actions; tuned to exclude automation or approved workflows."},{"field":"TimeWindow","description":"Threshold for correlating snapshot creation followed by reversion within minutes; tuned to environment activity norms."},{"field":"GeoLocation","description":"Region or source IP where the revert request originated; tuned to align with enterprise cloud geography."},{"field":"ChangeTags","description":"Use of administrative tags or headers to distinguish legitimate restores from malicious activity."}],"log_sources":[{"component_id":"DC0073","component":"Instance Modification","name":"AWS:CloudTrail","channel":"RevertSnapshot","platforms":["IaaS"],"analytic_id":"AN0953","source_url":"https://attack.mitre.org/detectionstrategies/DET0337#AN0953","validation":"upstream_definition_not_locally_validated","component_page":"telemetry/DC0073/"},{"component_id":"DC0080","component":"Instance Start","name":"AWS:CloudTrail","channel":"StartInstances","platforms":["IaaS"],"analytic_id":"AN0953","source_url":"https://attack.mitre.org/detectionstrategies/DET0337#AN0953","validation":"upstream_definition_not_locally_validated","component_page":"telemetry/DC0080/"},{"component_id":"DC0089","component":"Instance Stop","name":"AWS:CloudTrail","channel":"StopInstances","platforms":["IaaS"],"analytic_id":"AN0953","source_url":"https://attack.mitre.org/detectionstrategies/DET0337#AN0953","validation":"upstream_definition_not_locally_validated","component_page":"telemetry/DC0089/"}]}]}],"telemetry_references":[{"id":"DC0073","name":"Instance Modification","tag":"Instance Modification","page":"telemetry/DC0073/","kind":"attack_component"},{"id":"DC0080","name":"Instance Start","tag":"Instance Start","page":"telemetry/DC0080/","kind":"attack_component"},{"id":"DC0089","name":"Instance Stop","tag":"Instance Stop","page":"telemetry/DC0089/","kind":"attack_component"}],"visibility":"source_mapped","visibility_note":null,"tool_references":[],"related_parent":{"key":"enterprise/T1578","id":"T1578","name":"Modify Cloud Compute Infrastructure","page":"detections/enterprise/T1578/","note":"Related parent research only. Not inherited rule or anomaly coverage for this sub-technique."},"article_url":"https://1200km.com/articles/read/2026/2026-04-20-malicious-activity-as-a-statistical-signal-a-detection-engineering-analysis-of-anomaly-bas-90df8b6dea12/","validation":{"source_join":"exact_ids","sigma_yaml":"parsed","backend_compilation":"not_run","live_telemetry":"not_run","detection_execution":"not_run"}},"tools":[],"software":[],"actors":[],"mitigations":[],"candidates":[],"simulation":{"atomic_status":"no_atomic_candidate","lab_execution":"not_run","browser_attack_runner":false,"prerequisites":"Use disposable lab hosts and test accounts, explicit target allowlists, a clean snapshot, bounded egress and a restore plan. Review every candidate and dependency before any execution.","scenario":null},"validation":{"source_identity":"checked_against_pinned_source","relationship_join":"exact_ids_no_inheritance","page_contract":"automated_checks_required","sensor_capture":"not_run","detection_backend":"not_run","adversary_simulation":"not_run","independent_incident_fact_check":"not_performed"},"technique_links":{"T1578.004":"/ttp-simulation/techniques/enterprise/T1578.004/"},"synthetic":{"schema":"1200km.ttp.collection-fixture.v1","synthetic":true,"technique_key":"enterprise/T1578.004","purpose":"Collection/parser contract exercise, NOT a positive example or detection of this TTP.","events":[{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0073","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","attribute":"instance_type","before":"small-lab","after":"medium-lab"}},{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0080","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","action":"start","state_before":"stopped","state_after":"pending"}},{"schema":"1200km.telemetry.example.v1","synthetic":true,"timestamp":"2026-09-27T12:00:00Z","telemetry_id":"DC0089","collector":"illustrative-lab-collector","observation":{"instance_id":"vm-lab-1","action":"stop","state_before":"running","state_after":"stopping"}}]}}
