feat(admin): add profiling options for debugging high memory/CPU usage (#8923)

* feat(admin): add profiling options for debugging high memory/CPU usage

Add -debug, -debug.port, -cpuprofile, and -memprofile flags to the admin
command, matching the profiling support already available in master, volume,
and other server commands. This enables investigation of resource usage
issues like #8919.

* refactor(admin): move profiling flags into AdminOptions struct

Move cpuprofile and memprofile flags from global variables into the
AdminOptions struct and init() function for consistency with other flags.

* fix(debug): bind pprof server to localhost only and document profiling flags

StartDebugServer was binding to all interfaces (0.0.0.0), exposing
runtime profiling data to the network. Restrict to 127.0.0.1 since
this is a development/debugging tool.

Also add a "Debugging and Profiling" section to the admin command's
help text documenting the new flags.

weed/command/admin.go

@@ -30,6 +30,7 @@ import (
 	"github.com/seaweedfs/seaweedfs/weed/pb"
 	"github.com/seaweedfs/seaweedfs/weed/security"
 	"github.com/seaweedfs/seaweedfs/weed/util"
+	"github.com/seaweedfs/seaweedfs/weed/util/grace"
 )
 
 var (
@@ -48,6 +49,10 @@ type AdminOptions struct {
 	dataDir          *string
 	icebergPort      *int
 	urlPrefix        *string
+	debug            *bool
+	debugPort        *int
+	cpuProfile       *string
+	memProfile       *string
 }
 
 func init() {
@@ -64,6 +69,10 @@ func init() {
 	a.readOnlyPassword = cmdAdmin.Flag.String("readOnlyPassword", "", "read-only user password (optional, for view-only access; requires adminPassword to be set)")
 	a.icebergPort = cmdAdmin.Flag.Int("iceberg.port", 8181, "Iceberg REST Catalog port (0 to hide in UI)")
 	a.urlPrefix = cmdAdmin.Flag.String("urlPrefix", "", "URL path prefix when running behind a reverse proxy under a subdirectory (e.g. /seaweedfs)")
+	a.debug = cmdAdmin.Flag.Bool("debug", false, "serves runtime profiling data via pprof on the port specified by -debug.port")
+	a.debugPort = cmdAdmin.Flag.Int("debug.port", 6060, "http port for debugging")
+	a.cpuProfile = cmdAdmin.Flag.String("cpuprofile", "", "cpu profile output file")
+	a.memProfile = cmdAdmin.Flag.String("memprofile", "", "memory profile output file")
 }
 
 var cmdAdmin = &Command{
@@ -140,6 +149,16 @@ var cmdAdmin = &Command{
     - All static assets, API endpoints, and navigation links will use the prefix
     - Session cookies are scoped to the prefix path
 
+  Debugging and Profiling:
+    - Use -debug to start a pprof HTTP server for live profiling (localhost only)
+    - Set -debug.port to choose the pprof port (default 6060)
+    - Profiles are accessible at http://127.0.0.1:<debug.port>/debug/pprof/
+    - Use -cpuprofile and -memprofile to write profiles to files on shutdown
+    - WARNING: -debug exposes runtime internals; use only in trusted environments
+    - Examples:
+      weed admin -debug -debug.port=6060 -master="localhost:9333"
+      weed admin -cpuprofile=cpu.prof -memprofile=mem.prof -master="localhost:9333"
+
   Configuration File:
     - The security.toml file is read from ".", "$HOME/.seaweedfs/",
       "/usr/local/etc/seaweedfs/", or "/etc/seaweedfs/", in that order
@@ -149,6 +168,12 @@ var cmdAdmin = &Command{
 }
 
 func runAdmin(cmd *Command, args []string) bool {
+	if *a.debug {
+		grace.StartDebugServer(*a.debugPort)
+	}
+
+	grace.SetupProfiling(*a.cpuProfile, *a.memProfile)
+
 	// Load security configuration
 	util.LoadSecurityConfiguration()
 

weed/util/grace/pprof.go

@@ -15,8 +15,8 @@ import (
 // The server runs in a goroutine and serves pprof endpoints at /debug/pprof/*.
 func StartDebugServer(debugPort int) {
 	go func() {
-		addr := fmt.Sprintf(":%d", debugPort)
+		addr := fmt.Sprintf("127.0.0.1:%d", debugPort)
-		glog.V(0).Infof("Starting debug server for pprof at http://localhost%s/debug/pprof/", addr)
+		glog.V(0).Infof("Starting debug server for pprof at http://%s/debug/pprof/", addr)
 		if err := http.ListenAndServe(addr, nil); err != nil && err != http.ErrServerClosed {
 			glog.Errorf("Failed to start debug server on %s: %v", addr, err)
 		}