cli/glusterd: improve rebalance fix-layout status reporting

Problem:
Currenly the CLI rebalance status command output does not indicate the
'type' of rebalance, i.e. whether a full rebalance or only a fix-layout
was carried out.

Fix: After the rebalance status of all peers is received by the
originator glusterd, alter it to reflect the type of rebalance
before passing it on to the CLI process.

Change-Id: I1940ffda0d36e25e5b33c84a0ea210394cc9e1d3
BUG: 1004744
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/5826
Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Anand Avati <avati@redhat.com>

10 files changed, 177 insertions, 9 deletions

diff --git a/cli/src/cli-rpc-ops.c b/cli/src/cli-rpc-ops.c
index c15e0e0862b..fe3db41b6bf 100644
--- a/cli/src/cli-rpc-ops.c
+++ b/cli/src/cli-rpc-ops.c
@@ -67,6 +67,10 @@ char *cli_vol_task_status_str[] = {"not started",
                                    "stopped",
                                    "completed",
                                    "failed",
+                                   "fix-layout in progress",
+                                   "fix-layout stopped",
+                                   "fix-layout completed",
+                                   "fix-layout failed",
 };
 
 int32_t
@@ -1334,10 +1338,10 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
                 goto out;
         }
 
-        cli_out ("%40s %16s %13s %13s %13s %13s %14s %s", "Node",
+        cli_out ("%40s %16s %13s %13s %13s %13s %20s %18s", "Node",
                  "Rebalanced-files", "size", "scanned", "failures", "skipped",
                  "status", "run time in secs");
-        cli_out ("%40s %16s %13s %13s %13s %13s %14s %16s", "---------",
+        cli_out ("%40s %16s %13s %13s %13s %13s %20s %18s", "---------",
                  "-----------", "-----------", "-----------", "-----------",
                  "-----------", "------------", "--------------");
         do {
@@ -1398,7 +1402,7 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
                 status = cli_vol_task_status_str[status_rcd];
                 size_str = gf_uint64_2human_readable(size);
                 cli_out ("%40s %16"PRIu64 " %13s" " %13"PRIu64 " %13"PRIu64
-                         " %13"PRIu64 " %14s %16.2f", node_uuid, files,
+                         " %13"PRIu64 " %20s %18.2f", node_uuid, files,
                          size_str, lookup, failures, skipped, status, elapsed);
                 GF_FREE(size_str);
 
@@ -1892,6 +1896,8 @@ xml_output:
                 case GF_DEFRAG_STATUS_FAILED:
                         status = "failed";
                         break;
+                default:
+                        break;
                 }
 
                 size_str = gf_uint64_2human_readable(size);
diff --git a/rpc/xdr/src/cli1-xdr.h b/rpc/xdr/src/cli1-xdr.h
index 20787c494a0..10453bbb084 100644
--- a/rpc/xdr/src/cli1-xdr.h
+++ b/rpc/xdr/src/cli1-xdr.h
@@ -48,6 +48,10 @@ enum gf_defrag_status_t {
 	GF_DEFRAG_STATUS_STOPPED = 2,
 	GF_DEFRAG_STATUS_COMPLETE = 3,
 	GF_DEFRAG_STATUS_FAILED = 4,
+	GF_DEFRAG_STATUS_LAYOUT_FIX_STARTED = 5,
+	GF_DEFRAG_STATUS_LAYOUT_FIX_STOPPED = 6,
+	GF_DEFRAG_STATUS_LAYOUT_FIX_COMPLETE = 7,
+	GF_DEFRAG_STATUS_LAYOUT_FIX_FAILED = 8,
 };
 typedef enum gf_defrag_status_t gf_defrag_status_t;
 
diff --git a/rpc/xdr/src/cli1-xdr.x b/rpc/xdr/src/cli1-xdr.x
index d674030ec78..8018e303412 100644
--- a/rpc/xdr/src/cli1-xdr.x
+++ b/rpc/xdr/src/cli1-xdr.x
@@ -11,7 +11,11 @@
         GF_DEFRAG_STATUS_STARTED,
         GF_DEFRAG_STATUS_STOPPED,
         GF_DEFRAG_STATUS_COMPLETE,
-        GF_DEFRAG_STATUS_FAILED
+        GF_DEFRAG_STATUS_FAILED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_STARTED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_STOPPED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_COMPLETE,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_FAILED
 } ;
 
  enum gf1_cluster_type {
diff --git a/tests/bugs/bug-1004744.t b/tests/bugs/bug-1004744.t
new file mode 100644
index 00000000000..0290119ef89
--- /dev/null
+++ b/tests/bugs/bug-1004744.t
@@ -0,0 +1,48 @@
+#!/bin/bash
+
+#Test case: After a rebalance fix-layout, check if the rebalance status command
+#displays the appropriate message at the CLI.
+
+. $(dirname $0)/../include.rc
+. $(dirname $0)/../volume.rc
+
+cleanup;
+
+#Basic checks
+TEST glusterd
+TEST pidof glusterd
+TEST $CLI volume info
+
+#Create a 2x1 distributed volume
+TEST $CLI volume create $V0 $H0:$B0/${V0}{1,2};
+TEST $CLI volume start $V0
+
+# Mount FUSE and create file/directory
+TEST glusterfs -s $H0 --volfile-id $V0 $M0
+for i in `seq 1 10`;
+do
+       mkdir $M0/dir_$i
+       echo file>$M0/dir_$i/file_$i
+       for j in `seq 1 100`;
+       do
+                mkdir $M0/dir_$i/dir_$j
+                echo file>$M0/dir_$i/dir_$j/file_$j
+       done
+done
+
+#add 2 bricks
+TEST $CLI volume add-brick $V0 $H0:$B0/${V0}{3,4};
+
+#perform rebalance fix-layout
+TEST $CLI volume rebalance $V0 fix-layout start
+
+EXPECT_WITHIN 1 "fix-layout in progress" rebalance_status_field $V0;
+
+EXPECT_WITHIN 20 "fix-layout completed" rebalance_status_field $V0;
+
+TEST umount $M0
+TEST $CLI volume stop $V0
+TEST $CLI volume delete $V0;
+TEST ! $CLI volume info $V0;
+
+cleanup;
diff --git a/tests/bugs/bug-853258.t b/tests/bugs/bug-853258.t
index 79cb88099f2..0ca99555107 100755
--- a/tests/bugs/bug-853258.t
+++ b/tests/bugs/bug-853258.t
@@ -20,7 +20,7 @@ EXPECT_WITHIN 15 'Started' volinfo_field $V0 'Status';
 
 # Force assignment of initial ranges.
 TEST $CLI volume rebalance $V0 fix-layout start
-EXPECT_WITHIN 15 "completed" rebalance_status_field $V0
+EXPECT_WITHIN 15 "fix-layout completed" rebalance_status_field $V0
 
 # Get the original values.
 xattrs=""
@@ -32,7 +32,7 @@ done
 TEST $CLI volume add-brick $V0 $H0:$B0/${V0}3
 # Force assignment of initial ranges.
 TEST $CLI volume rebalance $V0 fix-layout start
-EXPECT_WITHIN 15 "completed" rebalance_status_field $V0
+EXPECT_WITHIN 15 "fix-layout completed" rebalance_status_field $V0
 
 for i in $(seq 0 3); do
 	xattrs="$xattrs $(dht_get_layout $B0/${V0}$i)"
diff --git a/tests/volume.rc b/tests/volume.rc
index 470fe9a7c3b..98712242020 100644
--- a/tests/volume.rc
+++ b/tests/volume.rc
@@ -27,7 +27,12 @@ function volume_option()
 }
 
 function rebalance_status_field {
-        $CLI volume rebalance $1 status | awk '{print $7}' | sed -n 3p
+        #The rebalance status can be upto 3 words, (ex:'fix-layout in progress'), hence the awk-print $7 thru $9.
+        #But if the status is less than 3 words, it also prints the next field i.e the run_time_in_secs.(ex:'completed 3.00').
+        #So we trim the numbers out with  `tr`. Finally remove the trailing white spaces with sed. What we get is one of the
+        #strings in the 'cli_vol_task_status_str' char array of cli-rpc-ops.c
+
+        $CLI volume rebalance $1 status | awk '{print $7,$8,$9}' |sed -n 3p |tr -d '[^0-9+\.]'|sed 's/ *$//g'
 }
 
 function remove_brick_status_completed_field {
diff --git a/xlators/cluster/dht/src/dht-common.h b/xlators/cluster/dht/src/dht-common.h
index 6ba18339b00..bdaa8d64d84 100644
--- a/xlators/cluster/dht/src/dht-common.h
+++ b/xlators/cluster/dht/src/dht-common.h
@@ -212,6 +212,10 @@ enum gf_defrag_status_t {
         GF_DEFRAG_STATUS_STOPPED,
         GF_DEFRAG_STATUS_COMPLETE,
         GF_DEFRAG_STATUS_FAILED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_STARTED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_STOPPED,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_COMPLETE,
+        GF_DEFRAG_STATUS_LAYOUT_FIX_FAILED,
 };
 typedef enum gf_defrag_status_t gf_defrag_status_t;
 
diff --git a/xlators/cluster/dht/src/dht-rebalance.c b/xlators/cluster/dht/src/dht-rebalance.c
index 5300ff2fe7a..8c36b7a090b 100644
--- a/xlators/cluster/dht/src/dht-rebalance.c
+++ b/xlators/cluster/dht/src/dht-rebalance.c
@@ -1760,6 +1760,8 @@ log:
         case GF_DEFRAG_STATUS_FAILED:
                 status = "failed";
                 break;
+        default:
+                break;
         }
 
         gf_log (THIS->name, GF_LOG_INFO, "Rebalance is %s. Time taken is %.2f "
diff --git a/xlators/mgmt/glusterd/src/glusterd-op-sm.c b/xlators/mgmt/glusterd/src/glusterd-op-sm.c
index 01f19164041..c1c9ff75e30 100644
--- a/xlators/mgmt/glusterd/src/glusterd-op-sm.c
+++ b/xlators/mgmt/glusterd/src/glusterd-op-sm.c
@@ -3072,6 +3072,97 @@ out:
         return ret;
 }
 
+static int
+reassign_defrag_status (dict_t *dict, char *key, gf_defrag_status_t *status)
+{
+        int ret = 0;
+
+        if (!*status)
+                return ret;
+
+        switch (*status) {
+        case GF_DEFRAG_STATUS_STARTED:
+                *status = GF_DEFRAG_STATUS_LAYOUT_FIX_STARTED;
+                break;
+
+        case GF_DEFRAG_STATUS_STOPPED:
+                *status = GF_DEFRAG_STATUS_LAYOUT_FIX_STOPPED;
+                break;
+
+        case GF_DEFRAG_STATUS_COMPLETE:
+                *status = GF_DEFRAG_STATUS_LAYOUT_FIX_COMPLETE;
+                break;
+
+        case GF_DEFRAG_STATUS_FAILED:
+                *status = GF_DEFRAG_STATUS_LAYOUT_FIX_FAILED;
+                break;
+        default:
+                break;
+         }
+
+        ret = dict_set_int32(dict, key, *status);
+        if (ret)
+                gf_log (THIS->name, GF_LOG_WARNING,
+                        "failed to reset defrag %s in dict", key);
+
+        return ret;
+}
+
+/* Check and reassign the defrag_status enum got from the rebalance process
+ * of all peers so that the rebalance-status CLI command can display if a
+ * full-rebalance or just a fix-layout was carried out.
+ */
+static int
+glusterd_op_check_peer_defrag_status (dict_t *dict, int count)
+{
+        glusterd_volinfo_t *volinfo  = NULL;
+        gf_defrag_status_t status    = GF_DEFRAG_STATUS_NOT_STARTED;
+        char               key[256]  = {0,};
+        char               *volname  = NULL;
+        int                ret       = -1;
+        int                i         = 1;
+
+        ret = dict_get_str (dict, "volname", &volname);
+        if (ret) {
+                gf_log (THIS->name, GF_LOG_WARNING, "Unable to get volume name");
+                goto out;
+        }
+
+        ret = glusterd_volinfo_find (volname, &volinfo);
+        if (ret) {
+                gf_log (THIS->name, GF_LOG_WARNING, FMTSTR_CHECK_VOL_EXISTS,
+                        volname);
+                goto out;
+        }
+
+        if (volinfo->rebal.defrag_cmd != GF_DEFRAG_CMD_START_LAYOUT_FIX) {
+                /* Fix layout was not issued; we don't need to reassign
+                   the status */
+                ret = 0;
+                goto out;
+        }
+
+        do {
+                memset (key, 0, 256);
+                snprintf (key, 256, "status-%d", i);
+                ret = dict_get_int32 (dict, key, (int32_t *)&status);
+                if (ret) {
+                        gf_log (THIS->name, GF_LOG_WARNING,
+                                "failed to get defrag %s", key);
+                        goto out;
+                }
+                ret = reassign_defrag_status (dict, key, &status);
+                if (ret)
+                        goto out;
+                i++;
+        } while (i <= count);
+
+        ret = 0;
+out:
+        return ret;
+
+}
+
 /* This function is used to modify the op_ctx dict before sending it back
  * to cli. This is useful in situations like changing the peer uuids to
  * hostnames etc.
@@ -3181,6 +3272,11 @@ glusterd_op_modify_op_ctx (glusterd_op_t op, void *ctx)
                 if (ret)
                         gf_log (this->name, GF_LOG_WARNING,
                                 "Failed uuid to hostname conversion");
+
+                ret = glusterd_op_check_peer_defrag_status (op_ctx, count);
+                if (ret)
+                        gf_log (this->name, GF_LOG_ERROR,
+                                "Failed to reset defrag status for fix-layout");
                 break;
 
         default:
diff --git a/xlators/mgmt/glusterd/src/glusterd-rebalance.c b/xlators/mgmt/glusterd/src/glusterd-rebalance.c
index 60dd9c19bde..031de2c9b0e 100644
--- a/xlators/mgmt/glusterd/src/glusterd-rebalance.c
+++ b/xlators/mgmt/glusterd/src/glusterd-rebalance.c
@@ -146,8 +146,6 @@ __glusterd_defrag_notify (struct rpc_clnt *rpc, void *mydata,
                                                 GF_DEFRAG_STATUS_STARTED) {
                                 volinfo->rebal.defrag_status =
                                                    GF_DEFRAG_STATUS_FAILED;
-                        } else {
-                                volinfo->rebal.defrag_cmd = 0;
                         }
                  }
 
@@ -219,6 +217,7 @@ glusterd_handle_defrag_start (glusterd_volinfo_t *volinfo, char *op_errstr,
 
         defrag->cmd = cmd;
 
+        volinfo->rebal.defrag_cmd = cmd;
         volinfo->rebal.op = op;
 
         LOCK_INIT (&defrag->lock);