summaryrefslogtreecommitdiffstats
path: root/heal
diff options
context:
space:
mode:
authorAnuradha Talur <atalur@redhat.com>2016-05-04 11:55:43 +0530
committerPranith Kumar Karampuri <pkarampu@redhat.com>2016-05-04 18:46:23 -0700
commit2326f8c2a775f6536d95e2e4676866f1b5dfc368 (patch)
tree04d94f7f6c3e7f0bd2995ee004c929741aac6c62 /heal
parent4c4624c9bad2edf27128cb122c64f15d7d63bbc8 (diff)
heal: Fix incorrect heal info output
Problem: In heterogenous clusters, heal info gives incorrect outout as "Failed to process entries completely. Number of entries so far: 0". This happens when the getxattr on virtual xattr for <brickpath>/.glusterfs/indices/dirty fails on older bricks as they do not recognize the xattr. Fix: Ignore that error so that heal info doesn't incorrectly report failure. Change-Id: I0d3541ed0594d67c3c3b8568a89a975a100bf6dd BUG: 1332798 Signed-off-by: Anuradha Talur <atalur@redhat.com> Reviewed-on: http://review.gluster.org/14199 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.com>
Diffstat (limited to 'heal')
-rw-r--r--heal/src/glfs-heal.c3
1 files changed, 2 insertions, 1 deletions
diff --git a/heal/src/glfs-heal.c b/heal/src/glfs-heal.c
index 4f636c6a54e..19bf4175cff 100644
--- a/heal/src/glfs-heal.c
+++ b/heal/src/glfs-heal.c
@@ -457,7 +457,8 @@ glfsh_print_pending_heals_type (glfs_t *fs, xlator_t *top_subvol, loc_t *rootloc
ret = glfsh_get_index_dir_loc (rootloc, xl, &dirloc, &op_errno,
vgfid);
if (ret < 0) {
- if (op_errno == ESTALE || op_errno == ENOENT)
+ if (op_errno == ESTALE || op_errno == ENOENT ||
+ op_errno == ENOTSUP)
ret = 0;
else
ret = -op_errno;