From: NeilBrown <neilb@suse.de>
Date: Mon, 17 Jun 2013 06:55:31 +0000 (+1000)
Subject: Assemble/Incr: Don't include spares with too-high event count.
X-Git-Tag: mdadm-3.3-rc1~33
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=f80057aec5d314798251e318555cb8ac92e4c06f;p=thirdparty%2Fmdadm.git

Assemble/Incr: Don't include spares with too-high event count.

Some failure scenarios can leave a spare with a higher event count
than an in-sync device.  Assembling an array like this will confuse
the kernel.
So detect spares with event counts higher than the best non-spare
event count and exclude them from the array.

Reported-by: Alexander Lyakas <alex.bolshoy@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
---

diff --git a/Assemble.c b/Assemble.c
index b45abc2f..c927c209 100644
--- a/Assemble.c
+++ b/Assemble.c
@@ -690,11 +690,12 @@ static int load_devices(struct devs *devices, char *devmap,
 		devices[devcnt].i = *content;
 		devices[devcnt].i.disk.major = major(stb.st_rdev);
 		devices[devcnt].i.disk.minor = minor(stb.st_rdev);
-		if (most_recent < devcnt) {
-			if (devices[devcnt].i.events
-			    > devices[most_recent].i.events)
+
+		if (devices[devcnt].i.events
+		    > devices[most_recent].i.events &&
+		    devices[devcnt].i.disk.state == 6)
 				most_recent = devcnt;
-		}
+
 		if (content->array.level == LEVEL_MULTIPATH)
 			/* with multipath, the raid_disk from the superblock is meaningless */
 			i = devcnt;
@@ -1456,8 +1457,15 @@ try_again:
 			best[i] = -1;
 			continue;
 		}
+		/* Require event counter to be same as, or just less than,
+		 * most recent.  If it is bigger, it must be a stray spare and
+		 * should be ignored.
+		 */
 		if (devices[j].i.events+event_margin >=
-		    devices[most_recent].i.events) {
+		    devices[most_recent].i.events &&
+		    devices[j].i.events <=
+		    devices[most_recent].i.events
+			) {
 			devices[j].uptodate = 1;
 			if (i < content->array.raid_disks * 2) {
 				if (devices[j].i.recovery_start == MaxSector ||
diff --git a/Incremental.c b/Incremental.c
index f9c0feb9..ce89702f 100644
--- a/Incremental.c
+++ b/Incremental.c
@@ -760,6 +760,14 @@ static int count_active(struct supertype *st, struct mdinfo *sra,
 		if (avail[i])
 			cnt++;
 	}
+	/* Also need to reject any spare device with an event count that
+	 * is too high
+	 */
+	for (d = sra->devs; d; d = d->next) {
+		if (!(d->disk.state & (1<<MD_DISK_SYNC)) &&
+		    d->events > max_events)
+			d->disk.state |= (1 << MD_DISK_REMOVED);
+	}
 	free(best);
 	free(devmap);
 	return cnt + replcnt;