afr: allow I/O when favorite-child-policy is enabled

Problem:
Currently, I/O on a split-brained file fails even when the
favorite-child-policy is set until the self-heal is complete.

Fix:
If a valid 'source' is found using the set favorite-child-policy, inspect
and reset the afr pending xattrs on the 'sinks' (inside appropriate locks),
refresh the inode and then proceed with the read or write transaction.

The resetting itself happens in the self-heal code and hence can also
happen in the client side background-heal or by the shd's index-heal in
addition to the txn code path explained above. When it happens in via
heal, we also add checks in undo-pending to not reset the sink xattrs
again.

> Reviewed-on: http://review.gluster.org/15673
> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
> Smoke: Gluster Build System <jenkins@build.gluster.org>
> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>

Change-Id: Ic8c1317720cb26bd114b6fe6af4e58c73b864626
BUG: 1378547
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reported-by: Simon Turcotte-Langevin <simon.turcotte-langevin@ubisoft.com>
Reviewed-on: http://review.gluster.org/16091
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Niels de Vos <ndevos@redhat.com>

1 files changed, 14 insertions, 1 deletions

diff --git a/xlators/cluster/afr/src/afr-self-heal-metadata.c b/xlators/cluster/afr/src/afr-self-heal-metadata.c
index 85aaca7cefd..a67717bf9d4 100644
--- a/xlators/cluster/afr/src/afr-self-heal-metadata.c
+++ b/xlators/cluster/afr/src/afr-self-heal-metadata.c
@@ -202,6 +202,7 @@ __afr_selfheal_metadata_finalize_source (call_frame_t *frame, xlator_t *this,
                                          unsigned char *sources,
                                          unsigned char *sinks,
 					 unsigned char *healed_sinks,
+					 unsigned char *undid_pending,
 					 unsigned char *locked_on,
 					 struct afr_reply *replies)
 {
@@ -223,8 +224,14 @@ __afr_selfheal_metadata_finalize_source (call_frame_t *frame, xlator_t *this,
                                                             healed_sinks,
                                                             locked_on, replies,
                                                       AFR_METADATA_TRANSACTION);
-                if (source >= 0)
+                if (source >= 0) {
+                        _afr_fav_child_reset_sink_xattrs (frame, this, inode,
+                                                         source, healed_sinks,
+                                                         undid_pending,
+                                                       AFR_METADATA_TRANSACTION,
+                                                         locked_on, replies);
                         return source;
+                }
 
 		/* If this is a directory mtime/ctime only split brain
 		   use the most recent */
@@ -304,6 +311,7 @@ int
 __afr_selfheal_metadata_prepare (call_frame_t *frame, xlator_t *this, inode_t *inode,
 				 unsigned char *locked_on, unsigned char *sources,
 				 unsigned char *sinks, unsigned char *healed_sinks,
+                                 unsigned char *undid_pending,
 				 struct afr_reply *replies, gf_boolean_t *pflag)
 {
 	int ret = -1;
@@ -358,6 +366,7 @@ __afr_selfheal_metadata_prepare (call_frame_t *frame, xlator_t *this, inode_t *i
 	source = __afr_selfheal_metadata_finalize_source (frame, this, inode,
                                                           sources, sinks,
                                                           healed_sinks,
+                                                          undid_pending,
                                                           locked_on, replies);
 
 	if (source < 0)
@@ -375,6 +384,7 @@ afr_selfheal_metadata (call_frame_t *frame, xlator_t *this, inode_t *inode)
 	unsigned char *sinks = NULL;
 	unsigned char *data_lock = NULL;
 	unsigned char *healed_sinks = NULL;
+	unsigned char *undid_pending = NULL;
 	struct afr_reply *locked_replies = NULL;
         gf_boolean_t did_sh = _gf_true;
 	int source = -1;
@@ -384,6 +394,7 @@ afr_selfheal_metadata (call_frame_t *frame, xlator_t *this, inode_t *inode)
 	sources = alloca0 (priv->child_count);
 	sinks = alloca0 (priv->child_count);
 	healed_sinks = alloca0 (priv->child_count);
+	undid_pending = alloca0 (priv->child_count);
 	data_lock = alloca0 (priv->child_count);
 
 	locked_replies = alloca0 (sizeof (*locked_replies) * priv->child_count);
@@ -399,6 +410,7 @@ afr_selfheal_metadata (call_frame_t *frame, xlator_t *this, inode_t *inode)
 		ret = __afr_selfheal_metadata_prepare (frame, this, inode,
                                                        data_lock, sources,
                                                        sinks, healed_sinks,
+                                                       undid_pending,
 						       locked_replies, NULL);
 		if (ret < 0)
 			goto unlock;
@@ -417,6 +429,7 @@ afr_selfheal_metadata (call_frame_t *frame, xlator_t *this, inode_t *inode)
 
                 ret = afr_selfheal_undo_pending (frame, this, inode, sources,
                                                  sinks, healed_sinks,
+                                                 undid_pending,
                                                  AFR_METADATA_TRANSACTION,
                                                  locked_replies, data_lock);
 	}