[pkg/stanza][receiver/windowseventlog] Improve EventDataType support …

…in EventXML (Windows) (open-telemetry#28587) **Description:** The [XML schema for Windows events supports `Data` elements without the `Name` attribute](https://learn.microsoft.com/en-us/windows/win32/wes/eventschema-datafieldtype-complextype), however, the current implementation doesn't capture `Data` elements without the `Name` attribute. Capturing such elements is specially important for events for which the publisher metadata is invalid. These elements contain the data that will give a user a much better chance of actually understanding the event, see [here](open-telemetry#21491 (comment)) for an example. I'm adding also the optional `Binary` element. Although this element typically requires knowledge of the actual data type it is representing sometimes it can be useful together with the data elements. I consider this to be a breaking change because it modifies the layout of the event generated by the package. It isn't an addition, the old representation is changed, please refer to the changes in tests to see the difference. **Link to tracking Issue:** This is the last pending item to fix open-telemetry#24493, open-telemetry#21491 ([item 5](open-telemetry#21491)). **Testing:** - Local run of the affected receiver and package - "Run Windows" on my fork **Documentation:** N/A --------- Co-authored-by: Daniel Jaglowski <jaglows3@gmail.com>
jmsnll · Nov 12, 2023 · 992b1ea · 992b1ea
1 parent 8d9f25d
commit 992b1ea
Show file tree

Hide file tree

Showing 5 changed files with 201 additions and 55 deletions.
diff --git a/.chloggen/improve-event-data-support-for-windows-event-xml.yaml b/.chloggen/improve-event-data-support-for-windows-event-xml.yaml
@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: breaking
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: pkg/stanza
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Improve parsing of Windows Event XML by handling anonymous `Data` elements.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [21491]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext: This improves the contents of Windows log events for which the publisher manifest is unavailable. Previously, anonymous `Data` elements were ignored. This is a breaking change for users who were relying on the previous data format.
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [user]
diff --git a/pkg/stanza/operator/input/windows/testdata/xmlWithAnonymousEventDataEntries.xml b/pkg/stanza/operator/input/windows/testdata/xmlWithAnonymousEventDataEntries.xml
@@ -0,0 +1,23 @@
+<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
+  <System>
+    <Provider Name="VSS" /> 
+    <EventID Qualifiers="0">8194</EventID> 
+    <Version>0</Version> 
+    <Level>2</Level> 
+    <Task>0</Task> 
+    <Opcode>0</Opcode> 
+    <Keywords>0x80000000000000</Keywords> 
+    <TimeCreated SystemTime="2023-10-19T21:57:58.0685414Z" /> 
+    <EventRecordID>383972</EventRecordID> 
+    <Correlation /> 
+    <Execution ProcessID="0" ThreadID="0" /> 
+    <Channel>Application</Channel> 
+    <Computer>computer</Computer> 
+    <Security /> 
+  </System>
+  <EventData>
+    <Data>1st_value</Data> 
+    <Data>2nd_value</Data> 
+    <Binary>2D20</Binary> 
+  </EventData>
+</Event>
diff --git a/pkg/stanza/operator/input/windows/xml.go b/pkg/stanza/operator/input/windows/xml.go
@@ -13,24 +13,24 @@ import (
 
 // EventXML is the rendered xml of an event.
 type EventXML struct {
-	EventID          EventID          `xml:"System>EventID"`
-	Provider         Provider         `xml:"System>Provider"`
-	Computer         string           `xml:"System>Computer"`
-	Channel          string           `xml:"System>Channel"`
-	RecordID         uint64           `xml:"System>EventRecordID"`
-	TimeCreated      TimeCreated      `xml:"System>TimeCreated"`
-	Message          string           `xml:"RenderingInfo>Message"`
-	RenderedLevel    string           `xml:"RenderingInfo>Level"`
-	Level            string           `xml:"System>Level"`
-	RenderedTask     string           `xml:"RenderingInfo>Task"`
-	Task             string           `xml:"System>Task"`
-	RenderedOpcode   string           `xml:"RenderingInfo>Opcode"`
-	Opcode           string           `xml:"System>Opcode"`
-	RenderedKeywords []string         `xml:"RenderingInfo>Keywords>Keyword"`
-	Keywords         []string         `xml:"System>Keywords"`
-	Security         *Security        `xml:"System>Security"`
-	Execution        *Execution       `xml:"System>Execution"`
-	EventData        []EventDataEntry `xml:"EventData>Data"`
+	EventID          EventID     `xml:"System>EventID"`
+	Provider         Provider    `xml:"System>Provider"`
+	Computer         string      `xml:"System>Computer"`
+	Channel          string      `xml:"System>Channel"`
+	RecordID         uint64      `xml:"System>EventRecordID"`
+	TimeCreated      TimeCreated `xml:"System>TimeCreated"`
+	Message          string      `xml:"RenderingInfo>Message"`
+	RenderedLevel    string      `xml:"RenderingInfo>Level"`
+	Level            string      `xml:"System>Level"`
+	RenderedTask     string      `xml:"RenderingInfo>Task"`
+	Task             string      `xml:"System>Task"`
+	RenderedOpcode   string      `xml:"RenderingInfo>Opcode"`
+	Opcode           string      `xml:"System>Opcode"`
+	RenderedKeywords []string    `xml:"RenderingInfo>Keywords>Keyword"`
+	Keywords         []string    `xml:"System>Keywords"`
+	Security         *Security   `xml:"System>Security"`
+	Execution        *Execution  `xml:"System>Execution"`
+	EventData        EventData   `xml:"EventData"`
 }
 
 // parseTimestamp will parse the timestamp of the event.
@@ -148,19 +148,30 @@ func (e *EventXML) parseMessage() (string, map[string]interface{}) {
 	}
 }
 
-// parse event data entries into a map[string]interface
-// where the key is the Name attribute, and value is the element value
-// entries without Name are ignored
+// parse event data into a map[string]interface
 // see: https://learn.microsoft.com/en-us/windows/win32/wes/eventschema-datafieldtype-complextype
-func parseEventData(entries []EventDataEntry) map[string]interface{} {
-	outputMap := make(map[string]interface{}, len(entries))
+func parseEventData(eventData EventData) map[string]interface{} {
+	outputMap := make(map[string]interface{}, 3)
+	if eventData.Name != "" {
+		outputMap["name"] = eventData.Name
+	}
+	if eventData.Binary != "" {
+		outputMap["binary"] = eventData.Binary
+	}
+
+	if len(eventData.Data) == 0 {
+		return outputMap
+	}
 
-	for _, entry := range entries {
-		if entry.Name != "" {
-			outputMap[entry.Name] = entry.Value
+	dataMaps := make([]interface{}, len(eventData.Data))
+	for i, data := range eventData.Data {
+		dataMaps[i] = map[string]interface{}{
+			data.Name: data.Value,
 		}
 	}
 
+	outputMap["data"] = dataMaps
+
 	return outputMap
 }
 
@@ -191,7 +202,16 @@ type Provider struct {
 	EventSourceName string `xml:"EventSourceName,attr"`
 }
 
-type EventDataEntry struct {
+type EventData struct {
+	// https://learn.microsoft.com/en-us/windows/win32/wes/eventschema-eventdatatype-complextype
+	// ComplexData is not supported.
+	Name   string `xml:"Name,attr"`
+	Data   []Data `xml:"Data"`
+	Binary string `xml:"Binary"`
+}
+
+type Data struct {
+	// https://learn.microsoft.com/en-us/windows/win32/wes/eventschema-datafieldtype-complextype
 	Name  string `xml:"Name,attr"`
 	Value string `xml:",chardata"`
 }

diff --git a/pkg/stanza/operator/input/windows/xml_test.go b/pkg/stanza/operator/input/windows/xml_test.go
@@ -82,8 +82,8 @@ func TestParseBody(t *testing.T) {
 		Task:     "task",
 		Opcode:   "opcode",
 		Keywords: []string{"keyword"},
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"},
+		EventData: EventData{
+			Data: []Data{{Name: "1st_name", Value: "value"}, {Name: "2nd_name", Value: "another_value"}},
 		},
 		RenderedLevel:    "rendered_level",
 		RenderedTask:     "rendered_task",
@@ -110,7 +110,12 @@ func TestParseBody(t *testing.T) {
 		"task":        "rendered_task",
 		"opcode":      "rendered_opcode",
 		"keywords":    []string{"RenderedKeywords"},
-		"event_data":  map[string]interface{}{"name": "value", "another_name": "another_value"},
+		"event_data": map[string]interface{}{
+			"data": []interface{}{
+				map[string]interface{}{"1st_name": "value"},
+				map[string]interface{}{"2nd_name": "another_value"},
+			},
+		},
 	}
 
 	require.Equal(t, expected, xml.parseBody())
@@ -138,8 +143,8 @@ func TestParseBodySecurityExecution(t *testing.T) {
 		Task:     "task",
 		Opcode:   "opcode",
 		Keywords: []string{"keyword"},
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"},
+		EventData: EventData{
+			Data: []Data{{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"}},
 		},
 		Execution: &Execution{
 			ProcessID: 13,
@@ -180,7 +185,12 @@ func TestParseBodySecurityExecution(t *testing.T) {
 		"security": map[string]any{
 			"user_id": "my-user-id",
 		},
-		"event_data": map[string]interface{}{"name": "value", "another_name": "another_value"},
+		"event_data": map[string]interface{}{
+			"data": []interface{}{
+				map[string]interface{}{"name": "value"},
+				map[string]interface{}{"another_name": "another_value"},
+			},
+		},
 	}
 
 	require.Equal(t, expected, xml.parseBody())
@@ -214,8 +224,8 @@ func TestParseBodyFullExecution(t *testing.T) {
 		Task:     "task",
 		Opcode:   "opcode",
 		Keywords: []string{"keyword"},
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"},
+		EventData: EventData{
+			Data: []Data{{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"}},
 		},
 		Execution: &Execution{
 			ProcessID:     13,
@@ -266,7 +276,12 @@ func TestParseBodyFullExecution(t *testing.T) {
 		"security": map[string]any{
 			"user_id": "my-user-id",
 		},
-		"event_data": map[string]interface{}{"name": "value", "another_name": "another_value"},
+		"event_data": map[string]interface{}{
+			"data": []interface{}{
+				map[string]interface{}{"name": "value"},
+				map[string]interface{}{"another_name": "another_value"},
+			},
+		},
 	}
 
 	require.Equal(t, expected, xml.parseBody())
@@ -294,8 +309,8 @@ func TestParseNoRendered(t *testing.T) {
 		Task:     "task",
 		Opcode:   "opcode",
 		Keywords: []string{"keyword"},
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"},
+		EventData: EventData{
+			Data: []Data{{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"}},
 		},
 	}
 
@@ -318,7 +333,12 @@ func TestParseNoRendered(t *testing.T) {
 		"task":        "task",
 		"opcode":      "opcode",
 		"keywords":    []string{"keyword"},
-		"event_data":  map[string]interface{}{"name": "value", "another_name": "another_value"},
+		"event_data": map[string]interface{}{
+			"data": []interface{}{
+				map[string]interface{}{"name": "value"},
+				map[string]interface{}{"another_name": "another_value"},
+			},
+		},
 	}
 
 	require.Equal(t, expected, xml.parseBody())
@@ -346,8 +366,8 @@ func TestParseBodySecurity(t *testing.T) {
 		Task:     "task",
 		Opcode:   "opcode",
 		Keywords: []string{"keyword"},
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"},
+		EventData: EventData{
+			Data: []Data{{Name: "name", Value: "value"}, {Name: "another_name", Value: "another_value"}},
 		},
 		RenderedLevel:    "rendered_level",
 		RenderedTask:     "rendered_task",
@@ -374,32 +394,48 @@ func TestParseBodySecurity(t *testing.T) {
 		"task":        "rendered_task",
 		"opcode":      "rendered_opcode",
 		"keywords":    []string{"RenderedKeywords"},
-		"event_data":  map[string]interface{}{"name": "value", "another_name": "another_value"},
+		"event_data": map[string]interface{}{
+			"data": []interface{}{
+				map[string]interface{}{"name": "value"},
+				map[string]interface{}{"another_name": "another_value"},
+			},
+		},
 	}
 
 	require.Equal(t, expected, xml.parseBody())
 }
 
 func TestParseEventData(t *testing.T) {
 	xmlMap := EventXML{
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"},
+		EventData: EventData{
+			Name:   "EVENT_DATA",
+			Data:   []Data{{Name: "name", Value: "value"}},
+			Binary: "2D20",
 		},
 	}
 
 	parsed := xmlMap.parseBody()
-	expectedMap := map[string]interface{}{"name": "value"}
+	expectedMap := map[string]interface{}{
+		"name": "EVENT_DATA",
+		"data": []interface{}{
+			map[string]interface{}{"name": "value"},
+		},
+		"binary": "2D20",
+	}
 	require.Equal(t, expectedMap, parsed["event_data"])
 
 	xmlMixed := EventXML{
-		EventData: []EventDataEntry{
-			{Name: "name", Value: "value"},
-			{Value: "noname"},
+		EventData: EventData{
+			Data: []Data{{Name: "name", Value: "value"}, {Value: "no_name"}},
 		},
 	}
 
 	parsed = xmlMixed.parseBody()
-	expectedSlice := map[string]interface{}{"name": "value"}
+	expectedSlice := map[string]interface{}{
+		"data": []interface{}{
+			map[string]interface{}{"name": "value"},
+			map[string]interface{}{"": "no_name"}},
+	}
 	require.Equal(t, expectedSlice, parsed["event_data"])
 }
 
@@ -428,16 +464,54 @@ func TestUnmarshalWithEventData(t *testing.T) {
 		TimeCreated: TimeCreated{
 			SystemTime: "2022-04-22T10:20:52.3778625Z",
 		},
+		Computer:  "computer",
+		Channel:   "Application",
+		RecordID:  23401,
+		Level:     "4",
+		Message:   "",
+		Task:      "0",
+		Opcode:    "0",
+		Execution: &Execution{},
+		Security:  &Security{},
+		EventData: EventData{
+			Data: []Data{
+				{Name: "Time", Value: "2022-04-28T19:48:52Z"},
+				{Name: "Source", Value: "RulesEngine"}},
+		},
+		Keywords: []string{"0x80000000000000"},
+	}
+
+	require.Equal(t, xml, event)
+}
+
+func TestUnmarshalWithAnonymousEventDataEntries(t *testing.T) {
+	data, err := os.ReadFile(filepath.Join("testdata", "xmlWithAnonymousEventDataEntries.xml"))
+	require.NoError(t, err)
+
+	event, err := unmarshalEventXML(data)
+	require.NoError(t, err)
+
+	xml := EventXML{
+		EventID: EventID{
+			ID:         8194,
+			Qualifiers: 0,
+		},
+		Provider: Provider{
+			Name: "VSS",
+		},
+		TimeCreated: TimeCreated{
+			SystemTime: "2023-10-19T21:57:58.0685414Z",
+		},
 		Computer: "computer",
 		Channel:  "Application",
-		RecordID: 23401,
-		Level:    "4",
+		RecordID: 383972,
+		Level:    "2",
 		Message:  "",
 		Task:     "0",
 		Opcode:   "0",
-		EventData: []EventDataEntry{
-			{Name: "Time", Value: "2022-04-28T19:48:52Z"},
-			{Name: "Source", Value: "RulesEngine"},
+		EventData: EventData{
+			Data:   []Data{{Name: "", Value: "1st_value"}, {Name: "", Value: "2nd_value"}},
+			Binary: "2D20",
 		},
 		Keywords:  []string{"0x80000000000000"},
 		Security:  &Security{},

diff --git a/receiver/windowseventlogreceiver/receiver_windows_test.go b/receiver/windowseventlogreceiver/receiver_windows_test.go
@@ -117,7 +117,9 @@ func TestReadWindowsEventLogger(t *testing.T) {
 	eventData := body["event_data"]
 	eventDataMap, ok := eventData.(map[string]interface{})
 	require.True(t, ok)
-	require.Equal(t, map[string]interface{}{}, eventDataMap)
+	require.Equal(t, map[string]interface{}{
+		"data": []interface{}{map[string]interface{}{"": "Test log"}},
+	}, eventDataMap)
 
 	eventID := body["event_id"]
 	require.NotNil(t, eventID)