[Cluster-devel] cluster/cmirror-kernel/src dm-cmirror-client.c ...

[jbrassow@sourceware.org <jbrassow@sourceware.org>]

CVSROOT:	/cvs/cluster
Module name:	cluster
Branch: 	RHEL4
Changes by:	jbrassow at sourceware.org	2007-10-03 19:02:52

Modified files:
	cmirror-kernel/src: dm-cmirror-client.c dm-cmirror-server.c 

Log message:
	Bug 316031: dm-mirror: incorrect order of mirror presuspend ops caus...
	
	With kernel (dm-raid1.c) presuspend changes, we can now tell when
	recovery has been shutdown and when we can allow blocked writes.
	
	This should fix a hang issue when converting from one mirror type
	to another.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/cmirror-kernel/src/dm-cmirror-client.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.1.2.54&r2=1.1.2.55
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/cmirror-kernel/src/dm-cmirror-server.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.1.2.40&r2=1.1.2.41

--- cluster/cmirror-kernel/src/Attic/dm-cmirror-client.c	2007/09/27 20:31:18	1.1.2.54
+++ cluster/cmirror-kernel/src/Attic/dm-cmirror-client.c	2007/10/03 19:02:51	1.1.2.55
@@ -592,6 +592,8 @@
 			 * should start out suspended.
 			 */
 			atomic_set(&lc->suspended, 1);
+			lc->recovery_halted = 1;
+			DMDEBUG("Secondary log... suspended and recovery_halted");
 		}
 	}
 
@@ -764,6 +766,11 @@
 
 static int cluster_presuspend(struct dirty_log *log)
 {
+	struct log_c *lc = (struct log_c *) log->context;
+
+	DMDEBUG("cluster_presuspend: recovery halted on %s(%d)",
+		lc->uuid + (strlen(lc->uuid) - 8), lc->uuid_ref);
+	lc->recovery_halted = 1;
 	return 0;
 }
 
@@ -834,8 +841,9 @@
 	struct log_c *lc = (struct log_c *) log->context;
 
 	lc->sync_search = 0;
-	lc->recovery_halted = 0;
 	resume_server_requests();
+	DMDEBUG("cluster_resume: Setting recovery_halted = 0");
+	lc->recovery_halted = 0;
 	atomic_set(&lc->suspended, 0);
 
 	return 0;
@@ -1253,6 +1261,8 @@
 						(atomic_read(&tmp_lc->in_sync)) ? "YES" : "NO");
 					DMDEBUG("  suspended   : %s",
 						(atomic_read(&tmp_lc->suspended)) ? "YES" : "NO");
+					DMDEBUG("  recovery_halted : %s",
+						(tmp_lc->recovery_halted) ? "YES" : "NO");
 					DMDEBUG("  server_id   : %u", tmp_lc->server_id);
 					DMDEBUG("  server_valid: %s",
 						((tmp_lc->server_id != 0xDEAD) &&
--- cluster/cmirror-kernel/src/Attic/dm-cmirror-server.c	2007/09/27 20:31:18	1.1.2.40
+++ cluster/cmirror-kernel/src/Attic/dm-cmirror-server.c	2007/10/03 19:02:51	1.1.2.41
@@ -468,9 +468,15 @@
 		lr->u.lr_int_rtn = 1;
 
 		/* Try to make this region a priority */
+		/*
 		if ((lr->u.lr_region != lc->recovering_region) &&
 		    (lc->recovering_next == (uint64_t)-1))
 			lc->recovering_next = lr->u.lr_region;
+		*/
+		if ((lr->u.lr_region != lc->recovering_region) &&
+		    ((lc->recovering_next == (uint64_t)-1) ||
+		     (lc->recovering_next > lr->u.lr_region)))
+			lc->recovering_next = lr->u.lr_region;		
 		return 0;
 	}
 
@@ -728,6 +734,7 @@
 	 * failed.  In this case, there will not be a record for
 	 * the region.
 	 */
+	DMDEBUG("server_complete_resync_work - Setting recovery_halted = 1");
 	lc->recovery_halted = 1;
 
 	ru = find_ru(lc, who, lr->u.lr_region);