distribute: Respect end-of-dir on readdir only for last subvol

We need to ensure that only the last subvolume's end-of-directory
notification is respected so that directory reading does not stop
before all subvolumes have been read. That could happen because the
posix for each subvolume sends a ENOENT on end-of-directory but in
distribute we're not concerned only with a posix's view of the
directory but the aggregated namespace' view of the directory,
which is maintained by distribute.

Signed-off-by: Shehjar Tikoo <shehjart@gluster.com>
Signed-off-by: Anand V. Avati <avati@dev.gluster.com>

BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator)
URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597

3 files changed, 40 insertions, 0 deletions

diff --git a/xlators/cluster/dht/src/dht-common.c b/xlators/cluster/dht/src/dht-common.c
index 523a62429f6..7dcc7205ad0 100644
--- a/xlators/cluster/dht/src/dht-common.c
+++ b/xlators/cluster/dht/src/dht-common.c
@@ -2168,6 +2168,15 @@ dht_readdirp_cbk (call_frame_t *frame, void *cookie, xlator_t *this, int op_ret,
                 count++;
 	}
 	op_ret = count;
+        /* We need to ensure that only the last subvolume's end-of-directory
+         * notification is respected so that directory reading does not stop
+         * before all subvolumes have been read. That could happen because the
+         * posix for each subvolume sends a ENOENT on end-of-directory but in
+         * distribute we're not concerned only with a posix's view of the
+         * directory but the aggregated namespace' view of the directory.
+         */
+        if (prev->this != dht_last_up_subvol (this))
+                op_errno = 0;
 
 done:
 	if (count == 0) {
@@ -2258,6 +2267,15 @@ dht_readdir_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
                 }
         }
 	op_ret = count;
+        /* We need to ensure that only the last subvolume's end-of-directory
+         * notification is respected so that directory reading does not stop
+         * before all subvolumes have been read. That could happen because the
+         * posix for each subvolume sends a ENOENT on end-of-directory but in
+         * distribute we're not concerned only with a posix's view of the
+         * directory but the aggregated namespace' view of the directory.
+         */
+        if (prev->this != dht_last_up_subvol (this))
+                op_errno = 0;
 
 done:
 	if (count == 0) {
diff --git a/xlators/cluster/dht/src/dht-common.h b/xlators/cluster/dht/src/dht-common.h
index 42a75adfbde..d739f852dc4 100644
--- a/xlators/cluster/dht/src/dht-common.h
+++ b/xlators/cluster/dht/src/dht-common.h
@@ -269,6 +269,7 @@ int dht_layout_set (xlator_t *this, inode_t *inode, dht_layout_t *layout);
 void dht_layout_unref (xlator_t *this, dht_layout_t *layout);
 dht_layout_t *dht_layout_ref (xlator_t *this, dht_layout_t *layout);
 xlator_t *dht_first_up_subvol (xlator_t *this);
+xlator_t *dht_last_up_subvol (xlator_t *this);
 
 int dht_build_child_loc (xlator_t *this, loc_t *child, loc_t *parent, char *name);
 
diff --git a/xlators/cluster/dht/src/dht-helper.c b/xlators/cluster/dht/src/dht-helper.c
index 5a89ba58489..ebc0e5e014a 100644
--- a/xlators/cluster/dht/src/dht-helper.c
+++ b/xlators/cluster/dht/src/dht-helper.c
@@ -208,6 +208,27 @@ dht_first_up_subvol (xlator_t *this)
 	return child;
 }
 
+xlator_t *
+dht_last_up_subvol (xlator_t *this)
+{
+        dht_conf_t *conf = NULL;
+        xlator_t   *child = NULL;
+        int         i = 0;
+
+        conf = this->private;
+        LOCK (&conf->subvolume_lock);
+        {
+                for (i = conf->subvolume_cnt-1; i >= 0; i--) {
+                        if (conf->subvolume_status[i]) {
+                                child = conf->subvolumes[i];
+                                break;
+                        }
+                }
+        }
+        UNLOCK (&conf->subvolume_lock);
+
+        return child;
+}
 
 xlator_t *
 dht_subvol_get_hashed (xlator_t *this, loc_t *loc)