Attachment 8010 Details for Bug 9268 – git-am for 3.6.next.

[patch] git-am for 3.6.next.

look1 (text/plain), 8.26 KB, created by Jeremy Allison on 2012-10-08 19:26:07 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Jeremy Allison

Created: 2012-10-08 19:26:07 UTC

Size: 8.26 KB

patch

obsolete

>From e637d4311cf428a40c9133139a3dea8832ec10b1 Mon Sep 17 00:00:00 2001
>From: Rusty Russell <rusty@rustcorp.com.au>
>Date: Mon, 8 Oct 2012 11:56:47 -0700
>Subject: [PATCH 1/3] tdb: Make robust against shrinking tdbs
>
>When probing for a size change (eg. just before tdb_expand, tdb_check,
>tdb_rescue) we call tdb_oob(tdb, tdb->map_size, 1, 1).  Unfortunately
>this does nothing if the tdb has actually shrunk, which as Volker
>demonstrated, can actually happen if a "longlived" parent crashes.
>
>So move the map/update size/remap before the limit check.
>---
> lib/tdb/common/io.c |   15 ++++++++-------
> 1 files changed, 8 insertions(+), 7 deletions(-)
>
>diff --git a/lib/tdb/common/io.c b/lib/tdb/common/io.c
>index 78bbf2e..649b70f 100644
>--- a/lib/tdb/common/io.c
>+++ b/lib/tdb/common/io.c
>@@ -53,6 +53,14 @@ static int tdb_oob(struct tdb_context *tdb, tdb_off_t len, int probe)
> 		return -1;
> 	}
> 
>+	/* Unmap, update size, remap */
>+	if (tdb_munmap(tdb) == -1) {
>+		tdb->ecode = TDB_ERR_IO;
>+		return -1;
>+	}
>+	tdb->map_size = st.st_size;
>+	tdb_mmap(tdb);
>+
> 	if (st.st_size < (size_t)len) {
> 		if (!probe) {
> 			/* Ensure ecode is set for log fn. */
>@@ -63,13 +71,6 @@ static int tdb_oob(struct tdb_context *tdb, tdb_off_t len, int probe)
> 		return -1;
> 	}
> 
>-	/* Unmap, update size, remap */
>-	if (tdb_munmap(tdb) == -1) {
>-		tdb->ecode = TDB_ERR_IO;
>-		return -1;
>-	}
>-	tdb->map_size = st.st_size;
>-	tdb_mmap(tdb);
> 	return 0;
> }
> 
>-- 
>1.7.7.3
>
>
>From 14384988dea09230a4a3bb77dbb91bda3bd169a5 Mon Sep 17 00:00:00 2001
>From: Volker Lendecke <vl@samba.org>
>Date: Mon, 8 Oct 2012 12:02:43 -0700
>Subject: [PATCH 2/3] tdb: Make tdb robust against improper CLEAR_IF_FIRST
> restart
>
>When winbind is restarted, there is a potential crash in tdb. Following
>situation: We are in a cluster with ctdb. A winbind child hangs
>in a request to the DC. Cluster monitoring decides the node has a
>problem. Cluster monitoring decides to kill ctdbd. winbind child
>still hangs in a RPC request. winbind parent figures that ctdb is
>dead and immediately commits suicide. winbind parent is restarted by
>cluster management, overwriting gencache.tdb with CLEAR_IF_FIRST. The
>CLEAR_IF_FIRST logic as implemented now will not see that a child still
>has the tdb open, only the parent holds the ACTIVE_LOCK due to performance
>reasons. During the CLEAR_IF_FIRST logic is done, there is a very small
>window where we ftruncate(tfd, 0) the file and re-write a proper header
>without a lock. When during this small window the winbind child comes
>back, wanting to store something into gencache.tdb, that winbind child
>will crash with a SIGBUS.
>
>Sounds unlikely? See:
>
>    [2012/09/29 07:02:31.871607,  0] lib/util.c:1183(smb_panic)
>      PANIC (pid 1814517): internal error
>    [2012/09/29 07:02:31.877596,  0] lib/util.c:1287(log_stack_trace)
>      BACKTRACE: 35 stack frames:
>       #0 winbindd(log_stack_trace+0x1a) [0x7feb7d4ca18a]
>       #1 winbindd(smb_panic+0x2b) [0x7feb7d4ca25b]
>       #2 winbindd(+0x1a3cc4) [0x7feb7d4bacc4]
>       #3 /lib64/libc.so.6(+0x32900) [0x7feb7a929900]
>       #4 /lib64/libc.so.6(memcpy+0x35) [0x7feb7a97f355]
>       #5 /usr/lib64/libtdb.so.1(+0x6e76) [0x7feb7b0b0e76]
>       #6 /usr/lib64/libtdb.so.1(+0x3d37) [0x7feb7b0add37]
>       #7 /usr/lib64/libtdb.so.1(+0x863d) [0x7feb7b0b263d]
>       #8 /usr/lib64/libtdb.so.1(+0x8700) [0x7feb7b0b2700]
>       #9 /usr/lib64/libtdb.so.1(+0x2505) [0x7feb7b0ac505]
>       #10 /usr/lib64/libtdb.so.1(+0x25b7) [0x7feb7b0ac5b7]
>       #11 /usr/lib64/libtdb.so.1(tdb_fetch+0x13) [0x7feb7b0ac633]
>       #12 winbindd(gencache_set_data_blob+0x259) [0x7feb7d4d8449]
>       #13 winbindd(gencache_set+0x53) [0x7feb7d4d85b3]
>       #14 winbindd(gencache_del+0x5e) [0x7feb7d4d879e]
>       #15 winbindd(saf_delete+0x93) [0x7feb7d54b693]
>       #16 winbindd(+0xe507e) [0x7feb7d3fc07e]
>       #17 winbindd(+0xe85e5) [0x7feb7d3ff5e5]
>       #18 winbindd(+0xe65be) [0x7feb7d3fd5be]
>       #19 winbindd(+0xe7562) [0x7feb7d3fe562]
>       #20 winbindd(init_dc_connection+0x2e) [0x7feb7d3fe5be]
>       #21 winbindd(+0xe75d9) [0x7feb7d3fe5d9]
>       #22 winbindd(cm_connect_netlogon+0x58) [0x7feb7d3fe658]
>       #23 winbindd(_wbint_PingDc+0x61) [0x7feb7d410991]
>       #24 winbindd(+0x103175) [0x7feb7d41a175]
>       #25 winbindd(winbindd_dual_ndrcmd+0xb7) [0x7feb7d4107d7]
>       #26 winbindd(+0xf8609) [0x7feb7d40f609]
>       #27 winbindd(+0xf9075) [0x7feb7d410075]
>       #28 winbindd(tevent_common_loop_immediate+0xe8) [0x7feb7d4db198]
>       #29 winbindd(run_events_poll+0x3c) [0x7feb7d4d93fc]
>       #30 winbindd(+0x1c2b52) [0x7feb7d4d9b52]
>       #31 winbindd(_tevent_loop_once+0x90) [0x7feb7d4d9f60]
>       #32 winbindd(main+0x7b3) [0x7feb7d3e7aa3]
>       #33 /lib64/libc.so.6(__libc_start_main+0xfd) [0x7feb7a915cdd]
>       #34 winbindd(+0xce2a9) [0x7feb7d3e52a9]
>
>This is in a winbind child, logfiles surrounding indicate the parent
>was restarted.
>
>This patch takes all chain locks around the CLEAR_IF_FIRST introduced
>tdb_new_database.
>---
> lib/tdb/common/open.c |   34 +++++++++++++++++++++++++++++-----
> 1 files changed, 29 insertions(+), 5 deletions(-)
>
>diff --git a/lib/tdb/common/open.c b/lib/tdb/common/open.c
>index ec45689..eeea8c7 100644
>--- a/lib/tdb/common/open.c
>+++ b/lib/tdb/common/open.c
>@@ -313,12 +313,36 @@ _PUBLIC_ struct tdb_context *tdb_open_ex(const char *name, int hash_size, int td
> 	if ((tdb_flags & TDB_CLEAR_IF_FIRST) &&
> 	    (!tdb->read_only) &&
> 	    (locked = (tdb_nest_lock(tdb, ACTIVE_LOCK, F_WRLCK, TDB_LOCK_NOWAIT|TDB_LOCK_PROBE) == 0))) {
>-		open_flags |= O_CREAT;
>-		if (ftruncate(tdb->fd, 0) == -1) {
>+		int ret;
>+		ret = tdb_brlock(tdb, F_WRLCK, FREELIST_TOP, 0,
>+				TDB_LOCK_WAIT);
>+		if (ret == -1) {
> 			TDB_LOG((tdb, TDB_DEBUG_FATAL, "tdb_open_ex: "
>-				 "failed to truncate %s: %s\n",
>-				 name, strerror(errno)));
>-			goto fail; /* errno set by ftruncate */
>+				"tdb_brlock failed for %s: %s\n",
>+				name, strerror(errno)));
>+			goto fail;
>+		}
>+		ret = tdb_new_database(tdb, hash_size);
>+		if (ret == -1) {
>+			TDB_LOG((tdb, TDB_DEBUG_FATAL, "tdb_open_ex: "
>+				"tdb_new_database failed for %s: %s\n",
>+				name, strerror(errno)));
>+			tdb_unlockall(tdb);
>+			goto fail;
>+		}
>+		ret = tdb_brunlock(tdb, F_WRLCK, FREELIST_TOP, 0);
>+		if (ret == -1) {
>+			TDB_LOG((tdb, TDB_DEBUG_FATAL, "tdb_open_ex: "
>+				"tdb_unlockall failed for %s: %s\n",
>+				name, strerror(errno)));
>+			goto fail;
>+		}
>+		ret = lseek(tdb->fd, 0, SEEK_SET);
>+		if (ret == -1) {
>+			TDB_LOG((tdb, TDB_DEBUG_FATAL, "tdb_open_ex: "
>+				"lseek failed for %s: %s\n",
>+				name, strerror(errno)));
>+			goto fail;
> 		}
> 	}
> 
>-- 
>1.7.7.3
>
>
>From 387efb6d5679a729051baddcd03032d6c789b6a3 Mon Sep 17 00:00:00 2001
>From: Volker Lendecke <vl@samba.org>
>Date: Mon, 8 Oct 2012 12:25:49 -0700
>Subject: [PATCH 3/3] s3: Add two tests a CLEAR_IF_FIRST crash
>
>---
> source3/torture/torture.c |   56 +++++++++++++++++++++++++++++++++++++++++++++
> 1 files changed, 56 insertions(+), 0 deletions(-)
>
>diff --git a/source3/torture/torture.c b/source3/torture/torture.c
>index 658224a..d37d83c 100644
>--- a/source3/torture/torture.c
>+++ b/source3/torture/torture.c
>@@ -8102,6 +8102,60 @@ fail:
> 	return result;
> }
> 
>+static bool run_local_tdb_opener(int dummy)
>+{
>+	TDB_CONTEXT *t;
>+	unsigned v = 0;
>+
>+	while (1) {
>+		t = tdb_open("test.tdb", 1000, TDB_CLEAR_IF_FIRST,
>+			     O_RDWR|O_CREAT, 0755);
>+		if (t == NULL) {
>+			perror("tdb_open failed");
>+			return false;
>+		}
>+		tdb_close(t);
>+
>+		v += 1;
>+		printf("\r%u", v);
>+	}
>+	return true;
>+}
>+
>+static bool run_local_tdb_writer(int dummy)
>+{
>+	TDB_CONTEXT *t;
>+	unsigned v = 0;
>+	TDB_DATA val;
>+
>+	t = tdb_open("test.tdb", 1000, 0, O_RDWR|O_CREAT, 0755);
>+	if (t == 0) {
>+		perror("tdb_open failed");
>+		return 1;
>+	}
>+
>+	val.dptr = (uint8_t *)&v;
>+	val.dsize = sizeof(v);
>+
>+	while (1) {
>+		TDB_DATA data;
>+		int ret;
>+
>+		ret = tdb_store(t, val, val, 0);
>+		if (ret != 0) {
>+			printf("%s\n", tdb_errorstr(t));
>+		}
>+		v += 1;
>+		printf("\r%u", v);
>+
>+		data = tdb_fetch(t, val);
>+		if (data.dptr != NULL) {
>+			SAFE_FREE(data.dptr);
>+		}
>+	}
>+	return true;
>+}
>+
> static double create_procs(bool (*fn)(int), bool *result)
> {
> 	int i, status;
>@@ -8290,6 +8344,8 @@ static struct {
> 	{ "LOCAL-binary_to_sid", run_local_binary_to_sid, 0},
> 	{ "LOCAL-DBTRANS", run_local_dbtrans, 0},
> 	{ "LOCAL-TEVENT-SELECT", run_local_tevent_select, 0},
>+	{ "LOCAL-TDB-OPENER", run_local_tdb_opener, 0 },
>+	{ "LOCAL-TDB-WRITER", run_local_tdb_writer, 0 },
> 	{NULL, NULL, 0}};
> 
> 
>-- 
>1.7.7.3
>

Flags:

vl: review+

Actions: View

Attachments on bug 9268: 8009 | 8010 | 8011