s3:smbprofile: Replace sysv shmem with tdb

What?

This patch gets rid of the central shared memory segment referenced by
"profile_p". Instead, every smbd gets a static profile_area where it collects
profiling data. Once a second, every smbd writes this profiling data into a
record of its own in a "smbprofile.tdb". smbstatus -P does a tdb_traverse on this
database and sums up what it finds.

Why?

At least in my perception sysv IPC has not the best reputation on earth. The
code before this patch uses shmat(). Samba ages ago has developed a good
abstraction of shared memory: It's called tdb.

The main reason why I started this is that I have a request to become
more flexible with profiling data. Samba should be able to collect data
per share or per user, something which is almost impossible to do with
a fixed structure. My idea is to for example install a profile area per
share and every second marshall this into one tdb record indexed by share
name. smbstatus -P would then also collect the data and either aggregate
them or put them into individual per-share statistics. This flexibility
in the data model is not really possible with one fixed structure.

But isn't it slow?

Well, I don't think so. I can't really prove it, but I do believe that on large
boxes atomically incrementing a shared memory value for every SMB does show up
due to NUMA effects. With this patch the hot code path is completely
process-local. Once a second every smbd writes into a central tdb, this of
course does atomic operations. But it's once a second, not on every SMB2 read.

There's two places where I would like to improve things: With the current code
all smbds wake up once a second. With 10,000 potentially idle smbds this will
become noticable. That's why the current only starts the timer when something has
changed.

The second place is the tdb traverse: Right now traverse is blocking in the
sense that when it has to switch hash chains it will block. With mutexes, this
means a syscall. I have a traverse light in mind that works as follows: It
assumes a locked hash chain and then walks the complete chain in one run
without unlocking in between. This way the caller can do nonblocking locks in
the first round and only do blocking locks in a second round. Also, a lot of
syscall overhead will vanish. This way smbstatus -P will have almost zero
impact on normal operations.

Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>

Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>

1 files changed, 301 insertions, 80 deletions

diff --git a/source3/profile/profile.c b/source3/profile/profile.c
index c7206385027..7002537354c 100644
--- a/source3/profile/profile.c
+++ b/source3/profile/profile.c
@@ -24,54 +24,42 @@
 #include "system/filesys.h"
 #include "messages.h"
 #include "smbprofile.h"
+#include "lib/tdb_wrap/tdb_wrap.h"
+#include <tevent.h>
+#include "../lib/crypto/crypto.h"
 
-#define PROF_SHMEM_KEY ((key_t)0x07021999)
-#define PROF_SHM_MAGIC 0x6349985
-#define PROF_SHM_VERSION 15
-
-#define IPC_PERMS ((S_IRUSR | S_IWUSR) | S_IRGRP | S_IROTH)
-
-static int shm_id;
-static bool read_only;
-
-struct profile_header {
-	int prof_shm_magic;
-	int prof_shm_version;
-	struct profile_stats stats;
-};
-
-static struct profile_header *profile_h;
 struct profile_stats *profile_p;
-
-bool do_profile_flag = False;
-bool do_profile_times = False;
+struct smbprofile_global_state smbprofile_state;
 
 /****************************************************************************
 Set a profiling level.
 ****************************************************************************/
 void set_profile_level(int level, struct server_id src)
 {
+	SMB_ASSERT(smbprofile_state.internal.db != NULL);
+
 	switch (level) {
 	case 0:		/* turn off profiling */
-		do_profile_flag = False;
-		do_profile_times = False;
+		smbprofile_state.config.do_count = false;
+		smbprofile_state.config.do_times = false;
 		DEBUG(1,("INFO: Profiling turned OFF from pid %d\n",
 			 (int)procid_to_pid(&src)));
 		break;
 	case 1:		/* turn on counter profiling only */
-		do_profile_flag = True;
-		do_profile_times = False;
+		smbprofile_state.config.do_count = true;
+		smbprofile_state.config.do_times = false;
 		DEBUG(1,("INFO: Profiling counts turned ON from pid %d\n",
 			 (int)procid_to_pid(&src)));
 		break;
 	case 2:		/* turn on complete profiling */
-		do_profile_flag = True;
-		do_profile_times = True;
+		smbprofile_state.config.do_count = true;
+		smbprofile_state.config.do_times = true;
 		DEBUG(1,("INFO: Full profiling turned ON from pid %d\n",
 			 (int)procid_to_pid(&src)));
 		break;
 	case 3:		/* reset profile values */
-		memset((char *)profile_p, 0, sizeof(*profile_p));
+		ZERO_STRUCT(profile_p->values);
+		tdb_wipe_all(smbprofile_state.internal.db->tdb);
 		DEBUG(1,("INFO: Profiling values cleared from pid %d\n",
 			 (int)procid_to_pid(&src)));
 		break;
@@ -109,7 +97,13 @@ static void reqprofile_message(struct messaging_context *msg_ctx,
 {
         int level;
 
-	level = 1 + (do_profile_flag?2:0) + (do_profile_times?4:0);
+	level = 1;
+	if (smbprofile_state.config.do_count) {
+		level += 2;
+	}
+	if (smbprofile_state.config.do_times) {
+		level += 4;
+	}
 
 	DEBUG(1,("INFO: Received REQ_PROFILELEVEL message from PID %u\n",
 		 (unsigned int)procid_to_pid(&src)));
@@ -122,75 +116,302 @@ static void reqprofile_message(struct messaging_context *msg_ctx,
   ******************************************************************/
 bool profile_setup(struct messaging_context *msg_ctx, bool rdonly)
 {
-	struct shmid_ds shm_ds;
+	unsigned char tmp[16] = {};
+	MD5_CTX md5;
+	char *db_name;
 
-	read_only = rdonly;
+	if (smbprofile_state.internal.db != NULL) {
+		return true;
+	}
 
- again:
-	/* try to use an existing key */
-	shm_id = shmget(PROF_SHMEM_KEY, 0, 0);
+	db_name = cache_path("smbprofile.tdb");
+	if (db_name == NULL) {
+		return false;
+	}
 
-	/* if that failed then create one. There is a race condition here
-	   if we are running from inetd. Bad luck. */
-	if (shm_id == -1) {
-		if (read_only) return False;
-		shm_id = shmget(PROF_SHMEM_KEY, sizeof(*profile_h),
-				IPC_CREAT | IPC_EXCL | IPC_PERMS);
+	smbprofile_state.internal.db = tdb_wrap_open(
+		NULL, db_name, 0,
+		rdonly ? 0 : TDB_CLEAR_IF_FIRST|TDB_MUTEX_LOCKING,
+		O_CREAT | (rdonly ? O_RDONLY : O_RDWR), 0644);
+	if (smbprofile_state.internal.db == NULL) {
+		return false;
 	}
 
-	if (shm_id == -1) {
-		DEBUG(0,("Can't create or use IPC area. Error was %s\n",
-			 strerror(errno)));
-		return False;
+	if (msg_ctx != NULL) {
+		messaging_register(msg_ctx, NULL, MSG_PROFILE,
+				   profile_message);
+		messaging_register(msg_ctx, NULL, MSG_REQ_PROFILELEVEL,
+				   reqprofile_message);
 	}
 
-	profile_h = (struct profile_header *)shmat(shm_id, 0,
-						   read_only?SHM_RDONLY:0);
-	if ((long)profile_h == -1) {
-		DEBUG(0,("Can't attach to IPC area. Error was %s\n",
-			 strerror(errno)));
-		return False;
+	MD5Init(&md5);
+
+	MD5Update(&md5,
+		  (const uint8_t *)&smbprofile_state.stats.global,
+		  sizeof(smbprofile_state.stats.global));
+
+#define __UPDATE(str) do { \
+	MD5Update(&md5, (const uint8_t *)str, strlen(str)); \
+} while(0)
+#define SMBPROFILE_STATS_START
+#define SMBPROFILE_STATS_SECTION_START(name, display) do { \
+	__UPDATE(#name "+" #display); \
+} while(0);
+#define SMBPROFILE_STATS_COUNT(name) do { \
+	__UPDATE(#name "+count"); \
+} while(0);
+#define SMBPROFILE_STATS_TIME(name) do { \
+	__UPDATE(#name "+time"); \
+} while(0);
+#define SMBPROFILE_STATS_BASIC(name) do { \
+	__UPDATE(#name "+count"); \
+	__UPDATE(#name "+time"); \
+} while(0);
+#define SMBPROFILE_STATS_BYTES(name) do { \
+	__UPDATE(#name "+count"); \
+	__UPDATE(#name "+time"); \
+	__UPDATE(#name "+idle"); \
+	__UPDATE(#name "+bytes"); \
+} while(0);
+#define SMBPROFILE_STATS_IOBYTES(name) do { \
+	__UPDATE(#name "+count"); \
+	__UPDATE(#name "+time"); \
+	__UPDATE(#name "+idle"); \
+	__UPDATE(#name "+inbytes"); \
+	__UPDATE(#name "+outbytes"); \
+} while(0);
+#define SMBPROFILE_STATS_SECTION_END
+#define SMBPROFILE_STATS_END
+	SMBPROFILE_STATS_ALL_SECTIONS
+#undef __UPDATE
+#undef SMBPROFILE_STATS_START
+#undef SMBPROFILE_STATS_SECTION_START
+#undef SMBPROFILE_STATS_COUNT
+#undef SMBPROFILE_STATS_TIME
+#undef SMBPROFILE_STATS_BASIC
+#undef SMBPROFILE_STATS_BYTES
+#undef SMBPROFILE_STATS_IOBYTES
+#undef SMBPROFILE_STATS_SECTION_END
+#undef SMBPROFILE_STATS_END
+
+	MD5Final(tmp, &md5);
+
+	profile_p = &smbprofile_state.stats.global;
+
+	profile_p->magic = BVAL(tmp, 0);
+	if (profile_p->magic == 0) {
+		profile_p->magic = BVAL(tmp, 8);
 	}
 
-	/* find out who created this memory area */
-	if (shmctl(shm_id, IPC_STAT, &shm_ds) != 0) {
-		DEBUG(0,("ERROR shmctl : can't IPC_STAT. Error was %s\n",
-			 strerror(errno)));
-		return False;
+	return True;
+}
+
+void smbprofile_dump_setup(struct tevent_context *ev)
+{
+	TALLOC_FREE(smbprofile_state.internal.te);
+	smbprofile_state.internal.ev = ev;
+}
+
+static void smbprofile_dump_timer(struct tevent_context *ev,
+				  struct tevent_timer *te,
+				  struct timeval current_time,
+				  void *private_data)
+{
+	smbprofile_dump();
+}
+
+void smbprofile_dump_schedule_timer(void)
+{
+	struct timeval tv;
+
+	GetTimeOfDay(&tv);
+	tv.tv_sec += 1;
+
+	smbprofile_state.internal.te = tevent_add_timer(
+				smbprofile_state.internal.ev,
+				smbprofile_state.internal.ev,
+				tv,
+				smbprofile_dump_timer,
+				NULL);
+}
+
+static int profile_stats_parser(TDB_DATA key, TDB_DATA value,
+				void *private_data)
+{
+	struct profile_stats *s = private_data;
+
+	if (value.dsize != sizeof(struct profile_stats)) {
+		*s = (struct profile_stats) {};
+		return 0;
 	}
 
-	if (shm_ds.shm_perm.cuid != sec_initial_uid() ||
-	    shm_ds.shm_perm.cgid != sec_initial_gid()) {
-		DEBUG(0,("ERROR: we did not create the shmem "
-			 "(owned by another user, uid %u, gid %u)\n",
-			 shm_ds.shm_perm.cuid,
-			 shm_ds.shm_perm.cgid));
-		return False;
+	memcpy(s, value.dptr, value.dsize);
+	if (s->magic != profile_p->magic) {
+		*s = (struct profile_stats) {};
+		return 0;
 	}
 
-	if (shm_ds.shm_segsz != sizeof(*profile_h)) {
-		DEBUG(0,("WARNING: profile size is %d (expected %d). Deleting\n",
-			 (int)shm_ds.shm_segsz, (int)sizeof(*profile_h)));
-		if (shmctl(shm_id, IPC_RMID, &shm_ds) == 0) {
-			goto again;
-		} else {
-			return False;
-		}
+	return 0;
+}
+
+void smbprofile_dump(void)
+{
+	pid_t pid = getpid();
+	TDB_DATA key = { .dptr = (uint8_t *)&pid, .dsize = sizeof(pid) };
+	struct profile_stats s = {};
+	int ret;
+
+	TALLOC_FREE(smbprofile_state.internal.te);
+
+	if (smbprofile_state.internal.db == NULL) {
+		return;
 	}
 
-	if (!read_only && (shm_ds.shm_nattch == 1)) {
-		memset((char *)profile_h, 0, sizeof(*profile_h));
-		profile_h->prof_shm_magic = PROF_SHM_MAGIC;
-		profile_h->prof_shm_version = PROF_SHM_VERSION;
-		DEBUG(3,("Initialised profile area\n"));
+	ret = tdb_chainlock(smbprofile_state.internal.db->tdb, key);
+	if (ret != 0) {
+		return;
 	}
 
-	profile_p = &profile_h->stats;
-	if (msg_ctx != NULL) {
-		messaging_register(msg_ctx, NULL, MSG_PROFILE,
-				   profile_message);
-		messaging_register(msg_ctx, NULL, MSG_REQ_PROFILELEVEL,
-				   reqprofile_message);
+	tdb_parse_record(smbprofile_state.internal.db->tdb,
+			 key, profile_stats_parser, &s);
+
+	smbprofile_stats_accumulate(profile_p, &s);
+
+	tdb_store(smbprofile_state.internal.db->tdb, key,
+		  (TDB_DATA) {
+			.dptr = (uint8_t *)profile_p,
+			.dsize = sizeof(*profile_p)
+		  },
+		  0);
+
+	tdb_chainunlock(smbprofile_state.internal.db->tdb, key);
+	ZERO_STRUCT(profile_p->values);
+
+	return;
+}
+
+void smbprofile_cleanup(pid_t pid)
+{
+	TDB_DATA key = { .dptr = (uint8_t *)&pid, .dsize = sizeof(pid) };
+	struct profile_stats s = {};
+	struct profile_stats acc = {};
+	int ret;
+
+	if (smbprofile_state.internal.db == NULL) {
+		return;
 	}
-	return True;
+
+	ret = tdb_chainlock(smbprofile_state.internal.db->tdb, key);
+	if (ret != 0) {
+		return;
+	}
+	ret = tdb_parse_record(smbprofile_state.internal.db->tdb,
+			       key, profile_stats_parser, &s);
+	if (ret == -1) {
+		tdb_chainunlock(smbprofile_state.internal.db->tdb, key);
+		return;
+	}
+	tdb_delete(smbprofile_state.internal.db->tdb, key);
+	tdb_chainunlock(smbprofile_state.internal.db->tdb, key);
+
+	pid = getpid();
+	ret = tdb_chainlock(smbprofile_state.internal.db->tdb, key);
+	if (ret != 0) {
+		return;
+	}
+	tdb_parse_record(smbprofile_state.internal.db->tdb,
+			 key, profile_stats_parser, &acc);
+
+	/*
+	 * We may have to fix the disconnect count
+	 * in case the process died
+	 */
+	s.values.disconnect_stats.count = s.values.connect_stats.count;
+
+	smbprofile_stats_accumulate(&acc, &s);
+
+	acc.magic = profile_p->magic;
+	tdb_store(smbprofile_state.internal.db->tdb, key,
+		  (TDB_DATA) {
+			.dptr = (uint8_t *)&acc,
+			.dsize = sizeof(acc)
+		  },
+		  0);
+
+	tdb_chainunlock(smbprofile_state.internal.db->tdb, key);
+}
+
+void smbprofile_stats_accumulate(struct profile_stats *acc,
+				 const struct profile_stats *add)
+{
+#define SMBPROFILE_STATS_START
+#define SMBPROFILE_STATS_SECTION_START(name, display)
+#define SMBPROFILE_STATS_COUNT(name) do { \
+	acc->values.name##_stats.count += add->values.name##_stats.count; \
+} while(0);
+#define SMBPROFILE_STATS_TIME(name) do { \
+	acc->values.name##_stats.time += add->values.name##_stats.time; \
+} while(0);
+#define SMBPROFILE_STATS_BASIC(name) do { \
+	acc->values.name##_stats.count += add->values.name##_stats.count; \
+	acc->values.name##_stats.time += add->values.name##_stats.time; \
+} while(0);
+#define SMBPROFILE_STATS_BYTES(name) do { \
+	acc->values.name##_stats.count += add->values.name##_stats.count; \
+	acc->values.name##_stats.time += add->values.name##_stats.time; \
+	acc->values.name##_stats.idle += add->values.name##_stats.idle; \
+	acc->values.name##_stats.bytes += add->values.name##_stats.bytes; \
+} while(0);
+#define SMBPROFILE_STATS_IOBYTES(name) do { \
+	acc->values.name##_stats.count += add->values.name##_stats.count; \
+	acc->values.name##_stats.time += add->values.name##_stats.time; \
+	acc->values.name##_stats.idle += add->values.name##_stats.idle; \
+	acc->values.name##_stats.inbytes += add->values.name##_stats.inbytes; \
+	acc->values.name##_stats.outbytes += add->values.name##_stats.outbytes; \
+} while(0);
+#define SMBPROFILE_STATS_SECTION_END
+#define SMBPROFILE_STATS_END
+	SMBPROFILE_STATS_ALL_SECTIONS
+#undef SMBPROFILE_STATS_START
+#undef SMBPROFILE_STATS_SECTION_START
+#undef SMBPROFILE_STATS_COUNT
+#undef SMBPROFILE_STATS_TIME
+#undef SMBPROFILE_STATS_BASIC
+#undef SMBPROFILE_STATS_BYTES
+#undef SMBPROFILE_STATS_IOBYTES
+#undef SMBPROFILE_STATS_SECTION_END
+#undef SMBPROFILE_STATS_END
+}
+
+static int smbprofile_collect_fn(struct tdb_context *tdb,
+				 TDB_DATA key, TDB_DATA value,
+				 void *private_data)
+{
+	struct profile_stats *acc = (struct profile_stats *)private_data;
+	const struct profile_stats *v;
+
+	if (value.dsize != sizeof(struct profile_stats)) {
+		return 0;
+	}
+
+	v = (const struct profile_stats *)value.dptr;
+
+	if (v->magic != profile_p->magic) {
+		return 0;
+	}
+
+	smbprofile_stats_accumulate(acc, v);
+	return 0;
+}
+
+void smbprofile_collect(struct profile_stats *stats)
+{
+	*stats = (struct profile_stats) {};
+
+	if (smbprofile_state.internal.db == NULL) {
+		return;
+	}
+
+	tdb_traverse_read(smbprofile_state.internal.db->tdb,
+			  smbprofile_collect_fn, stats);
 }