[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	2007-02-02 17:22:55

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

Log message:
	- Fix for bug #225337
	
	Reset 'sync_search' if (lc->sync_search >= lc->region_count) &&
	(lc->sync_count < lc->region_count).  It indicates that a failure
	during recovery has taken place, and we are likely able to handle
	it.
	
	Also, do not issue clear/mark region requests if it is already
	known that the log device has failed.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/cmirror-kernel/src/dm-cmirror-client.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.1.2.34&r2=1.1.2.35
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/cmirror-kernel/src/dm-cmirror-server.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.1.2.19&r2=1.1.2.20

--- cluster/cmirror-kernel/src/Attic/dm-cmirror-client.c	2007/01/08 19:28:26	1.1.2.34
+++ cluster/cmirror-kernel/src/Attic/dm-cmirror-client.c	2007/02/02 17:22:55	1.1.2.35
@@ -568,7 +568,7 @@
 		** while we are here.  If the clear region request fails, it**
 		** would be re-added - perhaps prematurely clearing the bit */
 		
-		if(rs){
+		if(rs && !rs->rs_lc->log_dev_failed){
 			_consult_server(rs->rs_lc, rs->rs_region,
 					LRT_CLEAR_REGION, NULL, &retry);
 
@@ -951,33 +951,35 @@
 
 	spin_unlock(&region_state_lock);
 
-	while((error = consult_server(lc, region, LRT_MARK_REGION, NULL))){
-		if (error == -EBUSY) {
-			/* Remote recovering delay and try again */
-			DMDEBUG("Delaying mark to region %Lu, due to recovery",
-				region);
-			set_current_state(TASK_INTERRUPTIBLE);
-			schedule_timeout(HZ/2);
-			continue;
-		}
+	if (!lc->log_dev_failed) {
+		while((error = consult_server(lc, region, LRT_MARK_REGION, NULL))){
+			if (error == -EBUSY) {
+				/* Remote recovering delay and try again */
+				DMDEBUG("Delaying mark to region %Lu, due to recovery",
+					region);
+				set_current_state(TASK_INTERRUPTIBLE);
+				schedule_timeout(HZ/2);
+				continue;
+			}
 
-		if (error == -EIO) {
-			lc->log_dev_failed = 1;
-			break;
+			if (error == -EIO) {
+				lc->log_dev_failed = 1;
+				break;
+			}
+			DMWARN("unable to get server (%u) to mark region (%Lu)",
+			       lc->server_id, region);
+			DMWARN("Reason :: %d", error);
 		}
-		DMWARN("unable to get server (%u) to mark region (%Lu)",
-		       lc->server_id, region);
-		DMWARN("Reason :: %d", error);
-	}
 
-	if (lc->log_dev_failed) {
-		dm_table_event(lc->ti->table);
-		/*
-		DMERR("Write failed on mirror log device, %s",
-		      lc->log_dev->name);
-		if (!atomic_read(&lc->suspended))
-			wait_for_completion(&lc->failure_completion);
-		*/
+		if (lc->log_dev_failed) {
+			dm_table_event(lc->ti->table);
+			/*
+			  DMERR("Write failed on mirror log device, %s",
+			  lc->log_dev->name);
+			  if (!atomic_read(&lc->suspended))
+			  wait_for_completion(&lc->failure_completion);
+			*/
+		}
 	}
 	return;
 }
--- cluster/cmirror-kernel/src/Attic/dm-cmirror-server.c	2007/01/08 19:28:26	1.1.2.19
+++ cluster/cmirror-kernel/src/Attic/dm-cmirror-server.c	2007/02/02 17:22:55	1.1.2.20
@@ -224,8 +224,16 @@
 
 static int _core_get_resync_work(struct log_c *lc, region_t *region)
 {
-	if (lc->sync_search >= lc->region_count){
-		return 0;
+	if (lc->sync_search >= lc->region_count) {
+		/*
+		 * FIXME: pvmove is not supported yet, but when it is,
+		 * an audit of sync_count changes will need to be made
+		 */
+		if (lc->sync_count < lc->region_count) {
+			lc->sync_search = 0;
+		} else {
+			return 0;
+		}
 	}
 	do {
 		*region = ext2_find_next_zero_bit((unsigned long *) lc->sync_bits,
@@ -557,9 +565,12 @@
 
 	if(!find_ru_by_region(lc, lr->u.lr_region)){
 		log_set_bit(lc, lc->clean_bits, lr->u.lr_region);
+		write_bits(lc);
+		/*
 		if (write_bits(lc))
 			DMERR("Write bits failed on mirror log device, %s",
 			      lc->log_dev->name);
+		*/
 	}
 	return 0;
 }