cluster/tier: account for reordered layouts

For a tiered volume the cold subvolume is always at a fixed
position in the graph. DHT's layout array, on the other hand,
may have the cold subvolume in either the first or second
index, therefore code cannot make any assumptions. The fix
searches the layout for the correct position dynamically
rather than statically.

The bug manifested itself in NFS, in which a newly attached
subvolume had not received an existing directory. This case
is a "stale entry" and marked as such in the layout for
that directory.  The code did not see this, because it
looked at the wrong index in the layout array.

The fix also adds the check for decomissioned bricks, and
fixes a problem in detach tier related to starting the
rebalance process: we never received the right defrag
command and it did not get directed to the tier translator.

Change-Id: I77cdf9fbb0a777640c98003188565a79be9d0b56
BUG: 1214289
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Tested-by: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
Reviewed-by: Joseph Fernandes <josferna@redhat.com>
Reviewed-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: http://review.gluster.org/11092

2 files changed, 32 insertions, 14 deletions

diff --git a/xlators/cluster/dht/src/dht-common.c b/xlators/cluster/dht/src/dht-common.c
index 6dc64eb13ab..0e020213470 100644
--- a/xlators/cluster/dht/src/dht-common.c
+++ b/xlators/cluster/dht/src/dht-common.c
@@ -7097,7 +7097,8 @@ int32_t dht_migration_needed(xlator_t *this)
 
         defrag = conf->defrag;
 
-        if (defrag->cmd != GF_DEFRAG_CMD_START_TIER)
+        if ((defrag->cmd != GF_DEFRAG_CMD_START_TIER) &&
+            (defrag->cmd != GF_DEFRAG_CMD_START_DETACH_TIER))
                 ret = 1;
 
 out:
diff --git a/xlators/cluster/dht/src/tier.c b/xlators/cluster/dht/src/tier.c
index 2279e060442..4cd7dfdd9b4 100644
--- a/xlators/cluster/dht/src/tier.c
+++ b/xlators/cluster/dht/src/tier.c
@@ -920,7 +920,8 @@ tier_migration_needed (xlator_t *this)
 
         defrag = conf->defrag;
 
-        if (defrag->cmd == GF_DEFRAG_CMD_START_TIER)
+        if ((defrag->cmd == GF_DEFRAG_CMD_START_TIER) ||
+            (defrag->cmd == GF_DEFRAG_CMD_START_DETACH_TIER))
                 ret = 1;
 out:
         return ret;
@@ -962,9 +963,11 @@ tier_search (xlator_t *this, dht_layout_t *layout, const char *name)
 {
         xlator_t                *subvol = NULL;
         void                    *value;
-        int                      search_first_subvol = 0;
+        int                      search_subvol = 0;
         dht_conf_t              *conf   = NULL;
         gf_defrag_info_t        *defrag = NULL;
+        int                      layout_cold = 0;
+        int                      layout_hot = 1;
 
         GF_VALIDATE_OR_GOTO("tier", this, out);
         GF_VALIDATE_OR_GOTO(this->name, layout, out);
@@ -973,28 +976,42 @@ tier_search (xlator_t *this, dht_layout_t *layout, const char *name)
 
         conf = this->private;
 
+        /* The first subvolume in the graph is always cold. */
+        /* Find the position of the cold subvolume in the layout. */
+        layout_cold = 0;
+        layout_hot = 1;
+        if (conf->subvolumes[0] != layout->list[0].xlator) {
+                layout_cold = 1;
+                layout_hot = 0;
+        }
+
+        search_subvol = layout_hot;
+
         defrag = conf->defrag;
         if (defrag && defrag->cmd == GF_DEFRAG_CMD_START_DETACH_TIER)
-                search_first_subvol = 1;
+                search_subvol = layout_cold;
 
+        /* "decommission_subvols_cnt" can only be non-zero on detach. */
+        /* This will change once brick add/remove is supported for */
+        /* tiered volumes. */
+        else if (conf->decommission_subvols_cnt) {
+                search_subvol = layout_cold;
+        }
         else if (!dict_get_ptr (this->options, "rule", &value) &&
-                 !strcmp(layout->list[0].xlator->name, value)) {
-                search_first_subvol = 1;
+                 !strcmp(layout->list[layout_cold].xlator->name, value)) {
+                search_subvol = layout_cold;
         }
 
-        if ((layout->list[0].err > 0) && (layout->list[0].err != ENOTCONN))
-                search_first_subvol = 0;
+        if ((layout->list[search_subvol].err > 0) &&
+            (layout->list[search_subvol].err != ENOTCONN))
+                search_subvol = layout_cold;
 
-        if (search_first_subvol)
-                subvol = layout->list[0].xlator;
-        else
-                subvol = layout->list[1].xlator;
+        subvol = layout->list[search_subvol].xlator;
+ out:
 
-out:
         return subvol;
 }
 
-
 dht_methods_t tier_methods = {
         .migration_get_dst_subvol = tier_migration_get_dst,
         .migration_other = tier_start,