Microsoft Sentinel Analytic Rules
cloudbrothers.infoAzure Sentinel RepoToggle Dark/Light/Auto modeToggle Dark/Light/Auto modeToggle Dark/Light/Auto modeBack to homepage

A potentially malicious web request was executed against a web server

Back
Id46ac55ae-47b8-414a-8f94-89ccd1962178
RulenameA potentially malicious web request was executed against a web server
DescriptionDetects unobstructed Web Application Firewall (WAF) activity in sessions where the WAF blocked incoming requests by computing the ratio between blocked requests and unobstructed WAF requests in these sessions (BlockvsSuccessRatio metric).

A high ratio value for a given client IP and hostname calls for further investigation of the WAF data in that session, due to the significantly high number of blocked requests and a few unobstructed logs that may be malicious but have passed undetected through the WAF. The successCode variable defines what the detection thinks is a successful status code and should be altered to fit the environment.
SeverityMedium
TacticsInitialAccess
TechniquesT1190
Required data connectorsWAF
KindScheduled
Query frequency1d
Query period1d
Trigger threshold0
Trigger operatorgt
Source Urihttps://github.com/Azure/Azure-Sentinel/blob/master/Solutions/Azure Web Application Firewall (WAF)/Analytic Rules/MaliciousWAFSessions.yaml
Version1.0.5
Arm template46ac55ae-47b8-414a-8f94-89ccd1962178.json
Deploy To Azure
let queryperiod = 1d;
let mode = dynamic(['Blocked', 'Detected']);
let successCode = dynamic(['200', '101','204', '400','504','304','401','500']);
let sessionBin = 30m;
AzureDiagnostics
| where TimeGenerated > ago(queryperiod)
| where ResourceProvider == 'MICROSOFT.NETWORK' and Category =~ 'ApplicationGatewayFirewallLog' and action_s in (mode)
| sort by hostname_s asc, clientIp_s asc, TimeGenerated asc
| extend SessionBlockedStarted = row_window_session(TimeGenerated, queryperiod, 10m, ((clientIp_s != prev(clientIp_s)) or (hostname_s != prev(hostname_s))))
| summarize SessionBlockedEnded = max(TimeGenerated), SessionBlockedCount = count() by hostname_s, clientIp_s, SessionBlockedStarted
| extend TimeKey = range(bin(SessionBlockedStarted, sessionBin), bin(SessionBlockedEnded, sessionBin), sessionBin)
| mv-expand TimeKey to typeof(datetime)
| join kind = inner(
    AzureDiagnostics
    | where TimeGenerated > ago(queryperiod)
    | where Category =~ 'ApplicationGatewayAccessLog' and (isempty(httpStatus_d) or httpStatus_d in (successCode))
    | extend TimeKey = bin(TimeGenerated, sessionBin)
    | extend hostname_s = coalesce(hostname_s,host_s), clientIp_s = coalesce(clientIp_s,clientIP_s)
) on TimeKey, hostname_s , clientIp_s
| where TimeGenerated between (SessionBlockedStarted..SessionBlockedEnded)
| extend
    originalRequestUriWithArgs_s = column_ifexists("originalRequestUriWithArgs_s", ""),
    serverStatus_s = column_ifexists("serverStatus_s", "")
| summarize
    SuccessfulAccessCount = count(),
    UserAgents = make_set(userAgent_s, 250),
    RequestURIs = make_set(requestUri_s, 250),
    OriginalRequestURIs = make_set(originalRequestUriWithArgs_s, 250),
    SuccessCodes = make_set(httpStatus_d, 250),
    SuccessCodes_BackendServer = make_set(serverStatus_s, 250),
    take_any(SessionBlockedEnded, SessionBlockedCount)
    by hostname_s, clientIp_s, SessionBlockedStarted
| where SessionBlockedCount > SuccessfulAccessCount
| extend BlockvsSuccessRatio = SessionBlockedCount/toreal(SuccessfulAccessCount)
| sort by BlockvsSuccessRatio desc, SessionBlockedStarted asc
| project-reorder SessionBlockedStarted, SessionBlockedEnded, hostname_s, clientIp_s, SessionBlockedCount, SuccessfulAccessCount, BlockvsSuccessRatio, SuccessCodes, RequestURIs, OriginalRequestURIs, UserAgents
severity: Medium
relevantTechniques:
- T1190
queryFrequency: 1d
kind: Scheduled
version: 1.0.5
name: A potentially malicious web request was executed against a web server
triggerOperator: gt
description: |
  'Detects unobstructed Web Application Firewall (WAF) activity in sessions where the WAF blocked incoming requests by computing the ratio between blocked requests and unobstructed WAF requests in these sessions (BlockvsSuccessRatio metric).
  A high ratio value for a given client IP and hostname calls for further investigation of the WAF data in that session, due to the significantly high number of blocked requests and a few unobstructed logs that may be malicious but have passed undetected through the WAF. The successCode variable defines what the detection thinks is a successful status code and should be altered to fit the environment.'  
queryPeriod: 1d
query: |
  let queryperiod = 1d;
  let mode = dynamic(['Blocked', 'Detected']);
  let successCode = dynamic(['200', '101','204', '400','504','304','401','500']);
  let sessionBin = 30m;
  AzureDiagnostics
  | where TimeGenerated > ago(queryperiod)
  | where ResourceProvider == 'MICROSOFT.NETWORK' and Category =~ 'ApplicationGatewayFirewallLog' and action_s in (mode)
  | sort by hostname_s asc, clientIp_s asc, TimeGenerated asc
  | extend SessionBlockedStarted = row_window_session(TimeGenerated, queryperiod, 10m, ((clientIp_s != prev(clientIp_s)) or (hostname_s != prev(hostname_s))))
  | summarize SessionBlockedEnded = max(TimeGenerated), SessionBlockedCount = count() by hostname_s, clientIp_s, SessionBlockedStarted
  | extend TimeKey = range(bin(SessionBlockedStarted, sessionBin), bin(SessionBlockedEnded, sessionBin), sessionBin)
  | mv-expand TimeKey to typeof(datetime)
  | join kind = inner(
      AzureDiagnostics
      | where TimeGenerated > ago(queryperiod)
      | where Category =~ 'ApplicationGatewayAccessLog' and (isempty(httpStatus_d) or httpStatus_d in (successCode))
      | extend TimeKey = bin(TimeGenerated, sessionBin)
      | extend hostname_s = coalesce(hostname_s,host_s), clientIp_s = coalesce(clientIp_s,clientIP_s)
  ) on TimeKey, hostname_s , clientIp_s
  | where TimeGenerated between (SessionBlockedStarted..SessionBlockedEnded)
  | extend
      originalRequestUriWithArgs_s = column_ifexists("originalRequestUriWithArgs_s", ""),
      serverStatus_s = column_ifexists("serverStatus_s", "")
  | summarize
      SuccessfulAccessCount = count(),
      UserAgents = make_set(userAgent_s, 250),
      RequestURIs = make_set(requestUri_s, 250),
      OriginalRequestURIs = make_set(originalRequestUriWithArgs_s, 250),
      SuccessCodes = make_set(httpStatus_d, 250),
      SuccessCodes_BackendServer = make_set(serverStatus_s, 250),
      take_any(SessionBlockedEnded, SessionBlockedCount)
      by hostname_s, clientIp_s, SessionBlockedStarted
  | where SessionBlockedCount > SuccessfulAccessCount
  | extend BlockvsSuccessRatio = SessionBlockedCount/toreal(SuccessfulAccessCount)
  | sort by BlockvsSuccessRatio desc, SessionBlockedStarted asc
  | project-reorder SessionBlockedStarted, SessionBlockedEnded, hostname_s, clientIp_s, SessionBlockedCount, SuccessfulAccessCount, BlockvsSuccessRatio, SuccessCodes, RequestURIs, OriginalRequestURIs, UserAgents  
entityMappings:
- entityType: IP
  fieldMappings:
  - columnName: clientIp_s
    identifier: Address
tactics:
- InitialAccess
status: Available
triggerThreshold: 0
OriginalUri: https://github.com/Azure/Azure-Sentinel/blob/master/Solutions/Azure Web Application Firewall (WAF)/Analytic Rules/MaliciousWAFSessions.yaml
id: 46ac55ae-47b8-414a-8f94-89ccd1962178
requiredDataConnectors:
- dataTypes:
  - AzureDiagnostics
  connectorId: WAF
{
  "$schema": "https://schema.management.azure.com/schemas/2019-04-01/deploymentTemplate.json#",
  "contentVersion": "1.0.0.0",
  "parameters": {
    "workspace": {
      "type": "String"
    }
  },
  "resources": [
    {
      "apiVersion": "2023-02-01-preview",
      "id": "[concat(resourceId('Microsoft.OperationalInsights/workspaces/providers', parameters('workspace'), 'Microsoft.SecurityInsights'),'/alertRules/46ac55ae-47b8-414a-8f94-89ccd1962178')]",
      "kind": "Scheduled",
      "name": "[concat(parameters('workspace'),'/Microsoft.SecurityInsights/46ac55ae-47b8-414a-8f94-89ccd1962178')]",
      "properties": {
        "alertRuleTemplateName": "46ac55ae-47b8-414a-8f94-89ccd1962178",
        "customDetails": null,
        "description": "'Detects unobstructed Web Application Firewall (WAF) activity in sessions where the WAF blocked incoming requests by computing the ratio between blocked requests and unobstructed WAF requests in these sessions (BlockvsSuccessRatio metric).\nA high ratio value for a given client IP and hostname calls for further investigation of the WAF data in that session, due to the significantly high number of blocked requests and a few unobstructed logs that may be malicious but have passed undetected through the WAF. The successCode variable defines what the detection thinks is a successful status code and should be altered to fit the environment.'\n",
        "displayName": "A potentially malicious web request was executed against a web server",
        "enabled": true,
        "entityMappings": [
          {
            "entityType": "IP",
            "fieldMappings": [
              {
                "columnName": "clientIp_s",
                "identifier": "Address"
              }
            ]
          }
        ],
        "OriginalUri": "https://github.com/Azure/Azure-Sentinel/blob/master/Solutions/Azure Web Application Firewall (WAF)/Analytic Rules/MaliciousWAFSessions.yaml",
        "query": "let queryperiod = 1d;\nlet mode = dynamic(['Blocked', 'Detected']);\nlet successCode = dynamic(['200', '101','204', '400','504','304','401','500']);\nlet sessionBin = 30m;\nAzureDiagnostics\n| where TimeGenerated > ago(queryperiod)\n| where ResourceProvider == 'MICROSOFT.NETWORK' and Category =~ 'ApplicationGatewayFirewallLog' and action_s in (mode)\n| sort by hostname_s asc, clientIp_s asc, TimeGenerated asc\n| extend SessionBlockedStarted = row_window_session(TimeGenerated, queryperiod, 10m, ((clientIp_s != prev(clientIp_s)) or (hostname_s != prev(hostname_s))))\n| summarize SessionBlockedEnded = max(TimeGenerated), SessionBlockedCount = count() by hostname_s, clientIp_s, SessionBlockedStarted\n| extend TimeKey = range(bin(SessionBlockedStarted, sessionBin), bin(SessionBlockedEnded, sessionBin), sessionBin)\n| mv-expand TimeKey to typeof(datetime)\n| join kind = inner(\n    AzureDiagnostics\n    | where TimeGenerated > ago(queryperiod)\n    | where Category =~ 'ApplicationGatewayAccessLog' and (isempty(httpStatus_d) or httpStatus_d in (successCode))\n    | extend TimeKey = bin(TimeGenerated, sessionBin)\n    | extend hostname_s = coalesce(hostname_s,host_s), clientIp_s = coalesce(clientIp_s,clientIP_s)\n) on TimeKey, hostname_s , clientIp_s\n| where TimeGenerated between (SessionBlockedStarted..SessionBlockedEnded)\n| extend\n    originalRequestUriWithArgs_s = column_ifexists(\"originalRequestUriWithArgs_s\", \"\"),\n    serverStatus_s = column_ifexists(\"serverStatus_s\", \"\")\n| summarize\n    SuccessfulAccessCount = count(),\n    UserAgents = make_set(userAgent_s, 250),\n    RequestURIs = make_set(requestUri_s, 250),\n    OriginalRequestURIs = make_set(originalRequestUriWithArgs_s, 250),\n    SuccessCodes = make_set(httpStatus_d, 250),\n    SuccessCodes_BackendServer = make_set(serverStatus_s, 250),\n    take_any(SessionBlockedEnded, SessionBlockedCount)\n    by hostname_s, clientIp_s, SessionBlockedStarted\n| where SessionBlockedCount > SuccessfulAccessCount\n| extend BlockvsSuccessRatio = SessionBlockedCount/toreal(SuccessfulAccessCount)\n| sort by BlockvsSuccessRatio desc, SessionBlockedStarted asc\n| project-reorder SessionBlockedStarted, SessionBlockedEnded, hostname_s, clientIp_s, SessionBlockedCount, SuccessfulAccessCount, BlockvsSuccessRatio, SuccessCodes, RequestURIs, OriginalRequestURIs, UserAgents\n",
        "queryFrequency": "P1D",
        "queryPeriod": "P1D",
        "severity": "Medium",
        "status": "Available",
        "suppressionDuration": "PT1H",
        "suppressionEnabled": false,
        "tactics": [
          "InitialAccess"
        ],
        "techniques": [
          "T1190"
        ],
        "templateVersion": "1.0.5",
        "triggerOperator": "GreaterThan",
        "triggerThreshold": 0
      },
      "type": "Microsoft.OperationalInsights/workspaces/providers/alertRules"
    }
  ]
}