cluster/dht: Treat migration failures due to space constraints as skipped

Currently rebalance/remove-brick op's display migration failed count even
for files which failed due to space issues (not enough space for file, or
migration leading to cluster imbalance)

These will now be counted as skipped, and rebalance/remove-brick status
will display the additional counter

Change-Id: I674904d380b5f8300e9ca9e6af557c3d30d6cff4
BUG: 989846
Signed-off-by: shishir gowda <sgowda@redhat.com>
Reviewed-on: http://review.gluster.org/5399
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>

4 files changed, 58 insertions, 5 deletions

diff --git a/xlators/cluster/dht/src/dht-common.h b/xlators/cluster/dht/src/dht-common.h
index 2be762994..8a62b9eec 100644
--- a/xlators/cluster/dht/src/dht-common.h
+++ b/xlators/cluster/dht/src/dht-common.h
@@ -227,6 +227,7 @@ struct gf_defrag_info_ {
         uint64_t                     total_data;
         uint64_t                     num_files_lookedup;
         uint64_t                     total_failures;
+        uint64_t                     skipped;
         gf_lock_t                    lock;
         int                          cmd;
         pthread_t                    th;
diff --git a/xlators/cluster/dht/src/dht-rebalance.c b/xlators/cluster/dht/src/dht-rebalance.c
index 8510a9b3b..dab6de609 100644
--- a/xlators/cluster/dht/src/dht-rebalance.c
+++ b/xlators/cluster/dht/src/dht-rebalance.c
@@ -1101,6 +1101,7 @@ gf_defrag_migrate_data (xlator_t *this, gf_defrag_info_t *defrag, loc_t *loc,
         struct timeval           end            = {0,};
         double                   elapsed        = {0,};
         struct timeval           start          = {0,};
+        int32_t                  err            = 0;
 
         gf_log (this->name, GF_LOG_INFO, "migrate data called on %s",
                 loc->path);
@@ -1258,9 +1259,21 @@ gf_defrag_migrate_data (xlator_t *this, gf_defrag_info_t *defrag, loc_t *loc,
                         ret = syncop_setxattr (this, &entry_loc, migrate_data,
                                                0);
                         if (ret) {
-                                gf_log (this->name, GF_LOG_ERROR, "migrate-data"
-                                        " failed for %s", entry_loc.path);
-                                defrag->total_failures +=1;
+                                err = op_errno;
+                                /* errno is overloaded. See
+                                 * rebalance_task_completion () */
+                                if (err != ENOSPC) {
+                                        gf_log (this->name, GF_LOG_DEBUG,
+                                                "migrate-data skipped for %s"
+                                                " due to space constraints",
+                                                entry_loc.path);
+                                        defrag->skipped +=1;
+                                } else{
+                                        gf_log (this->name, GF_LOG_ERROR,
+                                                "migrate-data failed for %s",
+                                                entry_loc.path);
+                                        defrag->total_failures +=1;
+                                }
                         }
 
                         if (ret == -1) {
@@ -1659,6 +1672,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
         uint64_t size   = 0;
         uint64_t lookup = 0;
         uint64_t failures = 0;
+        uint64_t skipped = 0;
         char     *status = "";
         double   elapsed = 0;
         struct timeval end = {0,};
@@ -1675,6 +1689,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
         size   = defrag->total_data;
         lookup = defrag->num_files_lookedup;
         failures = defrag->total_failures;
+        skipped = defrag->skipped;
 
         gettimeofday (&end, NULL);
 
@@ -1698,6 +1713,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
                 gf_log (THIS->name, GF_LOG_WARNING,
                         "failed to set lookedup file count");
 
+
         ret = dict_set_int32 (dict, "status", defrag->defrag_status);
         if (ret)
                 gf_log (THIS->name, GF_LOG_WARNING,
@@ -1710,6 +1726,14 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
         }
 
         ret = dict_set_uint64 (dict, "failures", failures);
+        if (ret)
+                gf_log (THIS->name, GF_LOG_WARNING,
+                        "failed to set failure count");
+
+        ret = dict_set_uint64 (dict, "skipped", skipped);
+        if (ret)
+                gf_log (THIS->name, GF_LOG_WARNING,
+                        "failed to set skipped file count");
 log:
         switch (defrag->defrag_status) {
         case GF_DEFRAG_STATUS_NOT_STARTED:
@@ -1732,8 +1756,8 @@ log:
         gf_log (THIS->name, GF_LOG_INFO, "Rebalance is %s. Time taken is %.2f "
                 "secs", status, elapsed);
         gf_log (THIS->name, GF_LOG_INFO, "Files migrated: %"PRIu64", size: %"
-                PRIu64", lookups: %"PRIu64", failures: %"PRIu64, files, size,
-                lookup, failures);
+                PRIu64", lookups: %"PRIu64", failures: %"PRIu64", skipped: "
+                "%"PRIu64, files, size, lookup, failures, skipped);
 
 
 out:
diff --git a/xlators/mgmt/glusterd/src/glusterd-utils.c b/xlators/mgmt/glusterd/src/glusterd-utils.c
index 85c17cd51..e894a9cd3 100644
--- a/xlators/mgmt/glusterd/src/glusterd-utils.c
+++ b/xlators/mgmt/glusterd/src/glusterd-utils.c
@@ -6154,6 +6154,7 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
         uint64_t                        lookup = 0;
         gf_defrag_status_t              status = GF_DEFRAG_STATUS_NOT_STARTED;
         uint64_t                        failures = 0;
+        uint64_t                        skipped = 0;
         xlator_t                       *this = NULL;
         double                          run_time = 0;
 
@@ -6184,6 +6185,11 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
                 gf_log (this->name, GF_LOG_TRACE,
                         "failed to get failure count");
 
+        ret = dict_get_uint64 (rsp_dict, "skipped", &skipped);
+        if (ret)
+                gf_log (this->name, GF_LOG_TRACE,
+                        "failed to get skipped count");
+
         ret = dict_get_double (rsp_dict, "run-time", &run_time);
         if (ret)
                 gf_log (this->name, GF_LOG_TRACE,
@@ -6199,6 +6205,8 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
                 volinfo->rebal.defrag_status = status;
         if (failures)
                 volinfo->rebal.rebalance_failures = failures;
+        if (skipped)
+                volinfo->rebal.skipped_files = skipped;
         if (run_time)
                 volinfo->rebal.rebalance_time = run_time;
 
@@ -6930,6 +6938,18 @@ glusterd_volume_rebalance_use_rsp_dict (dict_t *aggr, dict_t *rsp_dict)
         }
 
         memset (key, 0, 256);
+        snprintf (key, 256, "skipped-%d", index);
+        ret = dict_get_uint64 (rsp_dict, key, &value);
+        if (!ret) {
+                memset (key, 0, 256);
+                snprintf (key, 256, "skipped-%d", current_index);
+                ret = dict_set_uint64 (ctx_dict, key, value);
+                if (ret) {
+                        gf_log (THIS->name, GF_LOG_DEBUG,
+                                "failed to set skipped count");
+                }
+        }
+        memset (key, 0, 256);
         snprintf (key, 256, "run-time-%d", index);
         ret = dict_get_double (rsp_dict, key, &elapsed_time);
         if (!ret) {
@@ -7347,6 +7367,13 @@ glusterd_defrag_volume_node_rsp (dict_t *req_dict, dict_t *rsp_dict,
                 gf_log (THIS->name, GF_LOG_ERROR,
                         "failed to set failure count");
 
+        memset (key, 0 , 256);
+        snprintf (key, 256, "skipped-%d", i);
+        ret = dict_set_uint64 (op_ctx, key, volinfo->rebal.skipped_files);
+        if (ret)
+                gf_log (THIS->name, GF_LOG_ERROR,
+                        "failed to set skipped count");
+
         memset (key, 0, 256);
         snprintf (key, 256, "run-time-%d", i);
         ret = dict_set_double (op_ctx, key, volinfo->rebal.rebalance_time);
diff --git a/xlators/mgmt/glusterd/src/glusterd.h b/xlators/mgmt/glusterd/src/glusterd.h
index ccdf1c435..8c043a9a4 100644
--- a/xlators/mgmt/glusterd/src/glusterd.h
+++ b/xlators/mgmt/glusterd/src/glusterd.h
@@ -237,6 +237,7 @@ struct glusterd_rebalance_ {
         uint64_t                rebalance_files;
         uint64_t                rebalance_data;
         uint64_t                lookedup_files;
+        uint64_t                skipped_files;
         glusterd_defrag_info_t  *defrag;
         gf_cli_defrag_type      defrag_cmd;
         uint64_t                rebalance_failures;