cluster/afr: Remove local from owners_list on failure of lock-acquisition

When eager-lock lock acquisition fails because of say network failures, the
local is not being removed from owners_list, this leads to accumulation of
waiting frames and the application will hang because the waiting frames are
under the assumption that another transaction is in the process of acquiring
lock because owner-list is not empty. Handled this case as well in this patch.
Added asserts to make it easier to find these problems in future.

fixes bz#1699731
Change-Id: I3101393265e9827755725b1f2d94a93d8709e923
Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>

1 files changed, 4 insertions, 4 deletions

diff --git a/xlators/cluster/afr/src/afr-common.c b/xlators/cluster/afr/src/afr-common.c
index 56f4833b36c..8e8d7eb7473 100644
--- a/xlators/cluster/afr/src/afr-common.c
+++ b/xlators/cluster/afr/src/afr-common.c
@@ -5773,6 +5773,10 @@ afr_transaction_local_init(afr_local_t *local, xlator_t *this)
     afr_private_t *priv = NULL;
 
     priv = this->private;
+    INIT_LIST_HEAD(&local->transaction.wait_list);
+    INIT_LIST_HEAD(&local->transaction.owner_list);
+    INIT_LIST_HEAD(&local->ta_waitq);
+    INIT_LIST_HEAD(&local->ta_onwireq);
     ret = afr_internal_lock_init(&local->internal_lock, priv->child_count);
     if (ret < 0)
         goto out;
@@ -5810,10 +5814,6 @@ afr_transaction_local_init(afr_local_t *local, xlator_t *this)
         goto out;
 
     ret = 0;
-    INIT_LIST_HEAD(&local->transaction.wait_list);
-    INIT_LIST_HEAD(&local->transaction.owner_list);
-    INIT_LIST_HEAD(&local->ta_waitq);
-    INIT_LIST_HEAD(&local->ta_onwireq);
 out:
     return ret;
 }