cluster/ec: Automate heal for replace brick

Problem:
After a replace brick command, newly added
brick does not contain data which existed
on old brick.

Solution:
Do getxattr after initialization of all the
bricks. This will trigger heal for brick root
as soon as it finds the version mismatch on
newly added brick.

Removing tests from ec-new-entry.t which were
required to simulate automation of heal after
replace brick.

Change-Id: I08e3dfa565374097f6c08856325ea77727437e11
BUG: 1304686
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-on: http://review.gluster.org/13353
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Smoke: Gluster Build System <jenkins@build.gluster.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.com>

3 files changed, 68 insertions, 0 deletions

diff --git a/xlators/cluster/ec/src/ec-heal.c b/xlators/cluster/ec/src/ec-heal.c
index d80cdabc5ab..94ff4757b4d 100644
--- a/xlators/cluster/ec/src/ec-heal.c
+++ b/xlators/cluster/ec/src/ec-heal.c
@@ -2554,3 +2554,63 @@ fail:
     if (func)
             func (frame, NULL, this, -1, err, 0, 0, 0, NULL);
 }
+
+int
+ec_replace_heal_done (int ret, call_frame_t *heal, void *opaque)
+{
+        ec_t *ec = opaque;
+
+        gf_msg_debug (ec->xl->name, 0,
+                "getxattr on bricks is done ret %d", ret);
+        return 0;
+}
+
+int32_t
+ec_replace_heal (ec_t *ec, inode_t *inode)
+{
+        loc_t        loc = {0};
+        int          ret = 0;
+
+        loc.inode = inode_ref (inode);
+        gf_uuid_copy (loc.gfid, inode->gfid);
+        ret = syncop_getxattr (ec->xl, &loc, NULL, EC_XATTR_HEAL,
+                               NULL, NULL);
+        if (ret < 0)
+                gf_msg_debug (ec->xl->name, 0,
+                        "Heal failed for replace brick ret = %d", ret);
+
+        loc_wipe (&loc);
+        return ret;
+}
+
+int32_t
+ec_replace_brick_heal_wrap (void  *opaque)
+{
+         ec_t *ec = opaque;
+         inode_table_t   *itable = NULL;
+         int32_t         ret     = -1;
+
+         if (ec->xl->itable)
+                 itable = ec->xl->itable;
+         else
+                 goto out;
+         ret = ec_replace_heal (ec, itable->root);
+out:
+         return ret;
+}
+
+int32_t
+ec_launch_replace_heal (ec_t *ec)
+{
+	int ret = -1;
+
+        if (!ec)
+                return ret;
+        ret = synctask_new (ec->xl->ctx->env, ec_replace_brick_heal_wrap,
+                            ec_replace_heal_done, NULL, ec);
+        if (ret < 0) {
+                gf_msg_debug (ec->xl->name, 0,
+                        "Heal failed for replace brick ret = %d", ret);
+        }
+        return ret;
+}
diff --git a/xlators/cluster/ec/src/ec-helpers.h b/xlators/cluster/ec/src/ec-helpers.h
index 14243df54f3..1f39da2c09f 100644
--- a/xlators/cluster/ec/src/ec-helpers.h
+++ b/xlators/cluster/ec/src/ec-helpers.h
@@ -62,6 +62,9 @@ ec_filter_internal_xattrs (dict_t *xattr);
 
 gf_boolean_t
 ec_is_data_fop (glusterfs_fop_t fop);
+
+int32_t
+ec_launch_replace_heal (ec_t *ec);
 /*
 gf_boolean_t
 ec_is_metadata_fop (glusterfs_fop_t fop);
diff --git a/xlators/cluster/ec/src/ec.c b/xlators/cluster/ec/src/ec.c
index aad24e4dab7..1f15af10dab 100644
--- a/xlators/cluster/ec/src/ec.c
+++ b/xlators/cluster/ec/src/ec.c
@@ -515,6 +515,11 @@ unlock:
         if (propagate) {
                 error = default_notify (this, event, data);
         }
+
+        if (ec->shd.iamshd &&
+            ec->xl_notify_count == ec->nodes) {
+                ec_launch_replace_heal (ec);
+        }
 out:
         return error;
 }