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.

Change-Id: Ic8c1317720cb26bd114b6fe6af4e58c73b864626
BUG: 1386188
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reported-by: Simon Turcotte-Langevin <simon.turcotte-langevin@ubisoft.com>
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>

1 files changed, 14 insertions, 6 deletions

diff --git a/xlators/cluster/afr/src/afr-self-heal-data.c b/xlators/cluster/afr/src/afr-self-heal-data.c
index fbbbd192323..d032284926c 100644
--- a/xlators/cluster/afr/src/afr-self-heal-data.c
+++ b/xlators/cluster/afr/src/afr-self-heal-data.c
@@ -580,6 +580,7 @@ __afr_selfheal_data_finalize_source (call_frame_t *frame, xlator_t *this,
                                      unsigned char *sinks,
 				     unsigned char *healed_sinks,
 				     unsigned char *locked_on,
+				     unsigned char *undid_pending,
 				     struct afr_reply *replies,
                                      uint64_t *witness)
 {
@@ -603,6 +604,11 @@ __afr_selfheal_data_finalize_source (call_frame_t *frame, xlator_t *this,
                                  "file=%s", this->name, uuid_utoa(inode->gfid));
                         return -EIO;
                 }
+
+                _afr_fav_child_reset_sink_xattrs (frame, this, inode, source,
+                                                 healed_sinks, undid_pending,
+                                                 AFR_DATA_TRANSACTION,
+                                                 locked_on, replies);
                 return source;
 	}
 
@@ -642,6 +648,7 @@ __afr_selfheal_data_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;
@@ -677,8 +684,8 @@ __afr_selfheal_data_prepare (call_frame_t *frame, xlator_t *this,
 	source = __afr_selfheal_data_finalize_source (frame, this, inode,
                                                       sources, sinks,
                                                       healed_sinks,
-                                                      locked_on, replies,
-                                                      witness);
+                                                      locked_on, undid_pending,
+                                                      replies, witness);
 	if (source < 0)
 		return -EIO;
 
@@ -696,6 +703,7 @@ __afr_selfheal_data (call_frame_t *frame, xlator_t *this, fd_t *fd,
 	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;
 	int source = -1;
         gf_boolean_t did_sh = _gf_true;
@@ -707,6 +715,7 @@ __afr_selfheal_data (call_frame_t *frame, xlator_t *this, fd_t *fd,
 	sinks = alloca0 (priv->child_count);
 	healed_sinks = alloca0 (priv->child_count);
 	data_lock = alloca0 (priv->child_count);
+        undid_pending = alloca0 (priv->child_count);
 
 	locked_replies = alloca0 (sizeof (*locked_replies) * priv->child_count);
 
@@ -726,9 +735,8 @@ __afr_selfheal_data (call_frame_t *frame, xlator_t *this, fd_t *fd,
 
 		ret = __afr_selfheal_data_prepare (frame, this, fd->inode,
                                                    data_lock, sources, sinks,
-                                                   healed_sinks,
-						   locked_replies,
-                                                   NULL);
+                                                   healed_sinks, undid_pending,
+                                                   locked_replies, NULL);
 		if (ret < 0)
 			goto unlock;
 
@@ -787,7 +795,7 @@ restore_time:
         }
         ret = afr_selfheal_undo_pending (frame, this, fd->inode,
                                          sources, sinks, healed_sinks,
-                                         AFR_DATA_TRANSACTION,
+                                         undid_pending, AFR_DATA_TRANSACTION,
                                          locked_replies, data_lock);
 skip_undo_pending:
 	afr_selfheal_uninodelk (frame, this, fd->inode, this->name, 0, 0,