Merge pull request #141 from behouba/search-memory-pages

memparse: add search functionality to memparse

cmd/memparse.go

@@ -49,6 +49,30 @@ func MemParse() *cobra.Command {
 		"Specify the output file to be written to",
 	)
 
+	flags.StringVarP(
+		searchPattern,
+		"search",
+		"s",
+		"",
+		"Search for a string pattern in memory pages",
+	)
+
+	flags.StringVarP(
+		searchRegexPattern,
+		"search-regex",
+		"r",
+		"",
+		"Search for a regex pattern in memory pages",
+	)
+
+	flags.IntVarP(
+		searchContext,
+		"context",
+		"c",
+		0,
+		"Print the specified number of bytes surrounding each match",
+	)
+
 	return cmd
 }
 
@@ -79,6 +103,10 @@ func memparse(cmd *cobra.Command, args []string) error {
 	}
 	defer internal.CleanupTasks(tasks)
 
+	if *searchPattern != "" || *searchRegexPattern != "" {
+		return printMemorySearchResultForPID(tasks[0])
+	}
+
 	if *pID != 0 {
 		return printProcessMemoryPages(tasks[0])
 	}
@@ -273,3 +301,68 @@ func generateHexAndAscii(data []byte) (string, string) {
 
 	return hex, ascii
 }
+
+// Searches for a pattern in the memory of a given PID and prints the results.
+func printMemorySearchResultForPID(task internal.Task) error {
+	c := crit.New(nil, nil, filepath.Join(task.OutputDir, metadata.CheckpointDirectory), false, false)
+	psTree, err := c.ExplorePs()
+	if err != nil {
+		return fmt.Errorf("failed to get process tree: %w", err)
+	}
+
+	// Check if PID exist within the checkpoint
+	ps := psTree.FindPs(*pID)
+	if ps == nil {
+		return fmt.Errorf("no process with PID %d (use `inspect --ps-tree` to view all PIDs)", *pID)
+	}
+
+	memReader, err := crit.NewMemoryReader(
+		filepath.Join(task.OutputDir, metadata.CheckpointDirectory),
+		*pID, pageSize,
+	)
+	if err != nil {
+		return fmt.Errorf("failed to create memory reader: %w", err)
+	}
+
+	if err := internal.UntarFiles(
+		task.CheckpointFilePath, task.OutputDir,
+		[]string{filepath.Join(metadata.CheckpointDirectory, fmt.Sprintf("pages-%d.img", memReader.GetPagesID()))},
+	); err != nil {
+		return fmt.Errorf("failed to extract pages file: %w", err)
+	}
+
+	pattern := *searchPattern
+	escapeRegExpCharacters := true
+	if pattern == "" {
+		pattern = *searchRegexPattern
+		escapeRegExpCharacters = false
+	}
+
+	results, err := memReader.SearchPattern(pattern, escapeRegExpCharacters, *searchContext, 0)
+	if err != nil {
+		return fmt.Errorf("failed to search pattern in memory: %w", err)
+	}
+
+	if len(results) == 0 {
+		fmt.Printf("No matches for pattern \"%s\" in the memory of PID %d\n", pattern, *pID)
+		return nil
+	}
+
+	table := tablewriter.NewWriter(os.Stdout)
+	table.SetHeader([]string{"Address", "Match", "Instance"})
+	table.SetAutoMergeCells(false)
+	table.SetRowLine(true)
+
+	for i, result := range results {
+		table.Append([]string{
+			fmt.Sprintf(
+				"%016x", result.Vaddr),
+			result.Match,
+			fmt.Sprintf("%d", i+1),
+		})
+	}
+
+	table.Render()
+
+	return nil
+}

cmd/options.go

@@ -3,15 +3,18 @@ package cmd
 import "github.com/checkpoint-restore/checkpointctl/internal"
 
 var (
-	format         *string = &internal.Format
-	stats          *bool   = &internal.Stats
-	mounts         *bool   = &internal.Mounts
-	outputFilePath *string = &internal.OutputFilePath
-	pID            *uint32 = &internal.PID
-	psTree         *bool   = &internal.PsTree
-	psTreeCmd      *bool   = &internal.PsTreeCmd
-	psTreeEnv      *bool   = &internal.PsTreeEnv
-	files          *bool   = &internal.Files
-	sockets        *bool   = &internal.Sockets
-	showAll        *bool   = &internal.ShowAll
+	format             *string = &internal.Format
+	stats              *bool   = &internal.Stats
+	mounts             *bool   = &internal.Mounts
+	outputFilePath     *string = &internal.OutputFilePath
+	pID                *uint32 = &internal.PID
+	psTree             *bool   = &internal.PsTree
+	psTreeCmd          *bool   = &internal.PsTreeCmd
+	psTreeEnv          *bool   = &internal.PsTreeEnv
+	files              *bool   = &internal.Files
+	sockets            *bool   = &internal.Sockets
+	showAll            *bool   = &internal.ShowAll
+	searchPattern      *string = &internal.SearchPattern
+	searchRegexPattern *string = &internal.SearchRegexPattern
+	searchContext      *int    = &internal.SearchContext
 )

docs/checkpointctl-memparse.adoc

@@ -20,6 +20,15 @@ include::footer.adoc[]
 *-p, --pid*=_PID_::
   Specify the PID of a process to analyze
 
+*-s, --search*=_STRING_::
+  Search for a string pattern in memory pages
+
+*-r, --search-regex*=_REGEX_::
+  Search for a regex pattern in memory pages
+
+*-c, --context*=_CONTEXT_::
+  Print the specified number of bytes surrounding each match
+
 == See also
 
-checkpointctl(1)
+checkpointctl(1)

go.mod

@@ -5,7 +5,7 @@ go 1.21
 toolchain go1.21.1
 
 require (
-	github.com/checkpoint-restore/go-criu/v7 v7.1.0
+	github.com/checkpoint-restore/go-criu/v7 v7.2.0
 	github.com/containers/storage v1.54.0
 	github.com/olekukonko/tablewriter v0.0.5
 	github.com/opencontainers/runtime-spec v1.2.0
@@ -25,6 +25,6 @@ require (
 	github.com/spf13/pflag v1.0.5 // indirect
 	github.com/syndtr/gocapability v0.0.0-20200815063812-42c35b437635 // indirect
 	github.com/ulikunitz/xz v0.5.12 // indirect
-	golang.org/x/sys v0.20.0 // indirect
-	google.golang.org/protobuf v1.33.0 // indirect
+	golang.org/x/sys v0.25.0 // indirect
+	google.golang.org/protobuf v1.34.2 // indirect
 )

go.sum

@@ -1,5 +1,5 @@
-github.com/checkpoint-restore/go-criu/v7 v7.1.0 h1:JbQyO4o+P8ycNTMLPiiDqXg49bAcy4WljWCzYQho35A=
-github.com/checkpoint-restore/go-criu/v7 v7.1.0/go.mod h1:1svAtmbtvX4BKI45OFzgoTTLG7oYFKdColv/Vcsb2A8=
+github.com/checkpoint-restore/go-criu/v7 v7.2.0 h1:qGiWA4App1gGlEfIJ68WR9jbezV9J7yZdjzglezcqKo=
+github.com/checkpoint-restore/go-criu/v7 v7.2.0/go.mod h1:u0LCWLg0w4yqqu14aXhiB4YD3a1qd8EcCEg7vda5dwo=
 github.com/containers/storage v1.54.0 h1:xwYAlf6n9OnIlURQLLg3FYHbO74fQ/2W2N6EtQEUM4I=
 github.com/containers/storage v1.54.0/go.mod h1:PlMOoinRrBSnhYODLxt4EXl0nmJt+X0kjG0Xdt9fMTw=
 github.com/cpuguy83/go-md2man/v2 v2.0.4/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
@@ -47,10 +47,10 @@ github.com/xlab/treeprint v1.2.0 h1:HzHnuAF1plUN2zGlAFHbSQP2qJ0ZAD3XF5XD7OesXRQ=
 github.com/xlab/treeprint v1.2.0/go.mod h1:gj5Gd3gPdKtR1ikdDK6fnFLdmIS0X30kTTuNd/WEJu0=
 golang.org/x/sys v0.0.0-20220520151302-bc2c85ada10a/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y=
-golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=
-google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
+golang.org/x/sys v0.25.0 h1:r+8e+loiHxRqhXVl6ML1nO3l1+oFoWbnlu2Ehimmi34=
+golang.org/x/sys v0.25.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+google.golang.org/protobuf v1.34.2 h1:6xV6lTsCfpGD21XK49h7MhtcApnLqkfYgPcdHftf6hg=
+google.golang.org/protobuf v1.34.2/go.mod h1:qYOHts0dSfpeUzUFpOMr/WGzszTmLH+DiWniOlNbLDw=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=

internal/options.go

@@ -1,15 +1,18 @@
 package internal
 
 var (
-	Format         string
-	Stats          bool
-	Mounts         bool
-	OutputFilePath string
-	PID            uint32
-	PsTree         bool
-	PsTreeCmd      bool
-	PsTreeEnv      bool
-	Files          bool
-	Sockets        bool
-	ShowAll        bool
+	Format             string
+	Stats              bool
+	Mounts             bool
+	OutputFilePath     string
+	PID                uint32
+	PsTree             bool
+	PsTreeCmd          bool
+	PsTreeEnv          bool
+	Files              bool
+	Sockets            bool
+	ShowAll            bool
+	SearchPattern      string
+	SearchRegexPattern string
+	SearchContext      int
 )

test/checkpointctl.bats

@@ -577,6 +577,76 @@ function teardown() {
 	[[ ${lines[3]} == *"....H...H.../..H"* ]]
 }
 
+@test "Run checkpointctl memparse --search=PATH with invalid PID" {
+	cp data/config.dump \
+		data/spec.dump "$TEST_TMP_DIR1"
+	mkdir "$TEST_TMP_DIR1"/checkpoint
+	cp test-imgs/pstree.img \
+		test-imgs/core-*.img \
+		test-imgs/pagemap-*.img \
+		test-imgs/pages-*.img "$TEST_TMP_DIR1"/checkpoint
+	( cd "$TEST_TMP_DIR1" && tar cf "$TEST_TMP_DIR2"/test.tar . )
+	checkpointctl memparse --search=PATH "$TEST_TMP_DIR2"/test.tar --pid=999
+	[ "$status" -eq 1 ]
+	[[ ${lines[0]} == *"no process with PID 999"* ]]
+}
+
+@test "Run checkpointctl memparse with --search=PATH and --context=-1" {
+	cp data/config.dump \
+		data/spec.dump "$TEST_TMP_DIR1"
+	mkdir "$TEST_TMP_DIR1"/checkpoint
+	cp test-imgs/pstree.img \
+		test-imgs/core-*.img \
+		test-imgs/pagemap-*.img \
+		test-imgs/pages-*.img "$TEST_TMP_DIR1"/checkpoint
+	( cd "$TEST_TMP_DIR1" && tar cf "$TEST_TMP_DIR2"/test.tar . )
+	checkpointctl memparse --search=PATH --context=-1 "$TEST_TMP_DIR2"/test.tar --pid=1
+	[ "$status" -eq 1 ]
+	[[ ${lines[0]} == *"context size cannot be negative"* ]]
+}
+
+@test "Run checkpointctl memparse with --search=NON_EXISTING_PATTERN" {
+	cp data/config.dump \
+		data/spec.dump "$TEST_TMP_DIR1"
+	mkdir "$TEST_TMP_DIR1"/checkpoint
+	cp test-imgs/pstree.img \
+		test-imgs/core-*.img \
+		test-imgs/pagemap-*.img \
+		test-imgs/pages-*.img "$TEST_TMP_DIR1"/checkpoint
+	( cd "$TEST_TMP_DIR1" && tar cf "$TEST_TMP_DIR2"/test.tar . )
+	checkpointctl memparse --search=NON_EXISTING_PATTERN "$TEST_TMP_DIR2"/test.tar --pid=1
+	[ "$status" -eq 0 ]
+	[[ ${lines[0]} == *"No matches"* ]]
+}
+
+@test "Run checkpointctl memparse with --search=PATH and --context=10 flags" {
+	cp data/config.dump \
+		data/spec.dump "$TEST_TMP_DIR1"
+	mkdir "$TEST_TMP_DIR1"/checkpoint
+	cp test-imgs/pstree.img \
+		test-imgs/core-*.img \
+		test-imgs/pagemap-*.img \
+		test-imgs/pages-*.img "$TEST_TMP_DIR1"/checkpoint
+	( cd "$TEST_TMP_DIR1" && tar cf "$TEST_TMP_DIR2"/test.tar . )
+	checkpointctl memparse --search=PATH --context=10 "$TEST_TMP_DIR2"/test.tar --pid=1
+	[ "$status" -eq 0 ]
+	[[ ${lines[3]} == *"PATH"* ]]
+}
+
+@test "Run checkpointctl memparse with --search-regex='HOME=([^?]+)' " {
+	cp data/config.dump \
+		data/spec.dump "$TEST_TMP_DIR1"
+	mkdir "$TEST_TMP_DIR1"/checkpoint
+	cp test-imgs/pstree.img \
+		test-imgs/core-*.img \
+		test-imgs/pagemap-*.img \
+		test-imgs/pages-*.img "$TEST_TMP_DIR1"/checkpoint
+	( cd "$TEST_TMP_DIR1" && tar cf "$TEST_TMP_DIR2"/test.tar . )
+	checkpointctl memparse --search-regex='HOME=([^?]+)' "$TEST_TMP_DIR2"/test.tar --pid=1
+	[ "$status" -eq 0 ]
+	[[ ${lines[3]} == *"HOME"* ]]
+}
+
 @test "Run checkpointctl memparse with tar file and invalid PID" {
 	cp data/config.dump \
 		data/spec.dump "$TEST_TMP_DIR1"