ms_exchange_mailbox_replication_service_writing_active_server_pages.yml

name: MS Exchange Mailbox Replication service writing Active Server Pages
id: 985f322c-57a5-11ec-b9ac-acde48001122
version: 1
date: '2023-07-10'
author: Michael Haag, Splunk
status: experimental
type: TTP
description: 'The following query identifies suspicious .aspx created in 3 paths identified
  by Microsoft as known drop locations for Exchange exploitation related to HAFNIUM
  group and recently disclosed vulnerablity named ProxyShell. Paths include: `\HttpProxy\owa\auth\`,
  `\inetpub\wwwroot\aspnet_client\`, and `\HttpProxy\OAB\`. The analytic is limited
  to process name MSExchangeMailboxReplication.exe, which typically does not write
  .aspx files to disk. Upon triage, the suspicious .aspx file will likely look obvious
  on the surface. inspect the contents for script code inside. Identify additional
  log sources, IIS included, to review source and other potential exploitation. It
  is often the case that a particular threat is only applicable to a specific subset
  of systems in your environment. Typically analytics to detect those threats are
  written without the benefit of being able to only target those systems as well.
  Writing analytics against all systems when those behaviors are limited to identifiable
  subsets of those systems is suboptimal. Consider the case ProxyShell vulnerability
  on Microsoft Exchange Servers. With asset information, a hunter can limit their
  analytics to systems that have been identified as Exchange servers. A hunter may
  start with the theory that the exchange server is communicating with new systems
  that it has not previously. If this theory is run against all publicly facing systems,
  the amount of noise it will generate will likely render this theory untenable. However,
  using the asset information to limit this analytic to just the Exchange servers
  will reduce the noise allowing the hunter to focus only on the systems where this
  behavioral change is relevant.'
data_source:
- Sysmon EventID 1
- Sysmon EventID 11
search: '| tstats `security_content_summariesonly` count FROM datamodel=Endpoint.Processes
  where Processes.process_name=MSExchangeMailboxReplication.exe  by _time span=1h
  Processes.process_id Processes.process_name Processes.process_guid Processes.dest
  | `drop_dm_object_name(Processes)` | join process_guid, _time [| tstats `security_content_summariesonly`
  count min(_time) as firstTime max(_time) as lastTime FROM datamodel=Endpoint.Filesystem
  where Filesystem.file_path IN ("*\\HttpProxy\\owa\\auth\\*", "*\\inetpub\\wwwroot\\aspnet_client\\*",
  "*\\HttpProxy\\OAB\\*") Filesystem.file_name="*.aspx" by _time span=1h Filesystem.dest
  Filesystem.file_create_time Filesystem.file_name Filesystem.file_path | `drop_dm_object_name(Filesystem)`
  | fields _time dest file_create_time file_name file_path process_name process_path
  process process_guid] | dedup file_create_time | table dest file_create_time, file_name,
  file_path, process_name | `ms_exchange_mailbox_replication_service_writing_active_server_pages_filter`'
how_to_implement: To successfully implement this search you need to be ingesting information
  on process that include the name of the process responsible for the changes from
  your endpoints into the `Endpoint` datamodel in the `Processes` node and `Filesystem`
  node.
known_false_positives: The query is structured in a way that `action` (read, create)
  is not defined. Review the results of this query, filter, and tune as necessary.
  It may be necessary to generate this query specific to your endpoint product.
references:
- https://redcanary.com/blog/blackbyte-ransomware/
tags:
  analytic_story:
  - ProxyShell
  - Ransomware
  - BlackByte Ransomware
  asset_type: Endpoint
  confidence: 90
  impact: 90
  message: A file - $file_name$ was written to disk that is related to IIS exploitation
    related to ProxyShell. Review further file modifications on endpoint $dest$ by
    user $user$.
  mitre_attack_id:
  - T1505
  - T1505.003
  - T1190
  - T1133
  observable:
  - name: user
    type: User
    role:
    - Victim
  - name: dest
    type: Hostname
    role:
    - Victim
  - name: file_name
    type: File Name
    role:
    - Victim
  product:
  - Splunk Enterprise
  - Splunk Enterprise Security
  - Splunk Cloud
  required_fields:
  - _time
  - Filesystem.file_path
  - Filesystem.process_id
  - Filesystem.file_name
  - Filesystem.file_hash
  - Filesystem.user
  - Filesystem.process_guid
  - Processes.process_name
  - Processes.process_id
  - Processes.process_name
  - Processes.process_guid
  risk_score: 81
  security_domain: endpoint