glusterd : Fix for error mesage while detaching peers

When peer is detached from a cluster, an error log is being
generated in glusterd.log -"Failed to reconfigure all daemon
services". This log is seen in the originator node where the
detach is issued.

This happens in two cases.
Case 1: Detach peer with no volume been created in cluster.
Case 2: Detach peer after deleting all the volumes which were
created but never started.
In any one of the above two cases, in glusterd_check_files_identical()
GlusterD fails to retrieve nfs-server.vol file from /var/lib/glusterd/nfs
which gets created only when a volume is in place and and is started.

With this fix both the above cases have been handled by added
validation to skip reconfigure if there is no volume in started
state.

> Reviewed-on: https://review.gluster.org/16607
> Smoke: Gluster Build System <jenkins@build.gluster.org>
> Tested-by: Atin Mukherjee <amukherj@redhat.com>
> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>

(cherry picked from commit be44a1bd519af69b21acf682b0908d4d695f868e)

Change-Id: I039c0840e3d61ab54575e1e00c6a6a00874d84c0
BUG: 1423406
Signed-off-by: Gaurav Yadav <gyadav@redhat.com>
Reviewed-on: https://review.gluster.org/16645
Smoke: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Reviewed-by: Samikshan Bairagya <samikshan@gmail.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>

1 files changed, 15 insertions, 0 deletions

diff --git a/xlators/mgmt/glusterd/src/glusterd-nfs-svc.c b/xlators/mgmt/glusterd/src/glusterd-nfs-svc.c
index c6ab0c5d521..da343420790 100644
--- a/xlators/mgmt/glusterd/src/glusterd-nfs-svc.c
+++ b/xlators/mgmt/glusterd/src/glusterd-nfs-svc.c
@@ -145,17 +145,32 @@ glusterd_nfssvc_reconfigure ()
         xlator_t        *this            = NULL;
         glusterd_conf_t *priv            = NULL;
         gf_boolean_t     identical       = _gf_false;
+        gf_boolean_t     vol_started     = _gf_false;
+        glusterd_volinfo_t *volinfo      = NULL;
 
         this = THIS;
         GF_VALIDATE_OR_GOTO (this->name, this, out);
 
         priv = this->private;
         GF_VALIDATE_OR_GOTO (this->name, priv, out);
+
+        cds_list_for_each_entry (volinfo, &priv->volumes, vol_list) {
+                if (GLUSTERD_STATUS_STARTED == volinfo->status) {
+                        vol_started = _gf_true;
+                        break;
+                }
+        }
+        if (!vol_started) {
+                ret = 0;
+                goto out;
+        }
+
         /*
          * Check both OLD and NEW volfiles, if they are SAME by size
          * and cksum i.e. "character-by-character". If YES, then
          * NOTHING has been changed, just return.
          */
+
         ret = glusterd_svc_check_volfile_identical (priv->nfs_svc.name,
                                                     build_nfs_graph,
                                                     &identical);