support for de-commissioning a node using 'remove-brick'

to achieve this, we now create volume-file with
'decommissioned-nodes' option in distribute volume, then just
perform the rebalance set of operations (with 'force' flag set).

now onwards, the 'remove-brick' (with 'start' option) operation tries
to migrate data from removed bricks to existing bricks.

'remove-brick' also supports similar options as of replace-brick.

* (no options) -> works as 'force', will have the current behavior
         of remove-brick, ie., no data-migration, volume changes.

* start  (starts remove-brick with data-migration/draining process,
          which takes care of migrating data and once complete, will
          commit the changes to volume file)
* pause  (stop data migration, but keep the volume file intact with
          extra options whatever is set)
* abort  (stop data-migration, and fall back to old configuration)
* commit (if volume is stopped, commits the changes to volumefile)
* force  (stops the data-migration and commits the changes to
          volume file)

Change-Id: I3952bcfbe604a0952e68b6accace7014d5e401d3
BUG: 1952
Reviewed-on: http://review.gluster.com/118
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vijay@gluster.com>

1 files changed, 62 insertions, 0 deletions

diff --git a/xlators/cluster/dht/src/dht-common.c b/xlators/cluster/dht/src/dht-common.c
index 6f8594e30..e221e10ab 100644
--- a/xlators/cluster/dht/src/dht-common.c
+++ b/xlators/cluster/dht/src/dht-common.c
@@ -1690,6 +1690,46 @@ dht_common_setxattr_cbk (call_frame_t *frame, void *cookie,
 }
 
 int
+dht_checking_pathinfo_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
+                           int op_ret, int op_errno, dict_t *xattr)
+{
+        int           i     = -1;
+        int           ret   = -1;
+        char         *value = NULL;
+        dht_local_t  *local = NULL;
+        dht_conf_t   *conf  = NULL;
+        call_frame_t *prev  = NULL;
+        int           this_call_cnt = 0;
+
+        local = frame->local;
+        prev = cookie;
+        conf = this->private;
+
+        if (op_ret == -1)
+                goto out;
+
+
+        ret = dict_get_str (xattr, GF_XATTR_PATHINFO_KEY, &value);
+        if (ret)
+                goto out;
+
+        if (!strcmp (value, local->key)) {
+                for (i = 0; i < conf->subvolume_cnt; i++) {
+                        if (conf->subvolumes[i] == prev->this)
+                                conf->decommissioned_bricks[i] = prev->this;
+                }
+        }
+
+out:
+        this_call_cnt = dht_frame_return (frame);
+        if (is_last_call (this_call_cnt)) {
+                DHT_STACK_UNWIND (setxattr, frame, local->op_ret, ENOTSUP);
+        }
+        return 0;
+
+}
+
+int
 dht_setxattr (call_frame_t *frame, xlator_t *this,
               loc_t *loc, dict_t *xattr, int flags)
 {
@@ -1771,6 +1811,28 @@ dht_setxattr (call_frame_t *frame, xlator_t *this,
 
         }
 
+        tmp = dict_get (xattr, "decommission-brick");
+        if (tmp) {
+                /* This operation should happen only on '/' */
+                if (__is_root_gfid (loc->inode->gfid) != 0) {
+                        op_errno = ENOTSUP;
+                        goto err;
+                }
+
+                memcpy (value, tmp->data, ((tmp->len < 4095) ? tmp->len : 4095));
+                local->key = gf_strdup (value);
+                local->call_cnt = conf->subvolume_cnt;
+
+                for (i = 0 ; i < conf->subvolume_cnt; i++) {
+                        /* Get the pathinfo, and then compare */
+                        STACK_WIND (frame, dht_checking_pathinfo_cbk,
+                                    conf->subvolumes[i],
+                                    conf->subvolumes[i]->fops->getxattr,
+                                    loc, GF_XATTR_PATHINFO_KEY);
+                }
+                return 0;
+        }
+
         tmp = dict_get (xattr, GF_XATTR_FIX_LAYOUT_KEY);
         if (tmp) {
                 gf_log (this->name, GF_LOG_INFO,