diff mbox

ocfs2: force clean refmap when doing local recovery cleanup

Message ID 51FA4EEF.9050102@huawei.com
State New, archived
Headers show

Commit Message

Xue jiufei Aug. 1, 2013, 12:05 p.m. UTC
Function dlm_do_local_recovery_cleanup() should force clean refmap if
the owner of lockres is UNKNOWN. Otherwise node may hang when umounting
filesystems. 
Here's the situation:

	Node1                                    Node2
dlmlock()
  -> dlm_get_lock_resource()       
send DLM_MASTER_REQUEST_MSG to
other nodes.
                                    
                                       trying to master this lockres,
                                       return MAYBE.
							
selected as the master of lockresA,
set mle->master to Node1,
and do assert_master,
send DLM_ASSERT_MASTER_MSG to Node2.
                                       Node 2 has interest on lockresA
                                       and return
                                       DLM_ASSERT_RESPONSE_MASTERY_REF
                                       then something happened and
                                       Node2 crashed.
							           
receiving DLM_ASSERT_RESPONSE_MASTERY_REF,
set Node2 into refmap, and keep sending
DLM_ASSERT_MASTER_MSG to other nodes

o2hb found node2 down, calling
dlm_hb_node_down()
--> dlm_do_local_recovery_cleanup()
the master of lockresA is still UNKNOWN,
no need to call dlm_free_dead_locks().

set the master of lockresA to Node1, but
Node2 stills remains in refmap.

when Node1 umount, it found that the refmap of lockresA is not empty
and attempted to migrate it to Node2, But Node2 is already down,
so umount hang, trying to migrate lockresA again and again.

Signed-off-by: joyce <xuejiufei@huawei.com>
---
 fs/ocfs2/dlm/dlmrecovery.c |   18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

Comments

Sunil Mushran Aug. 2, 2013, 3:26 a.m. UTC | #1
I see no need for a separate function. Just do....

 } else if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) {
    if (test_bit(node, res->refmap))
          dlm_lockres_clear_refmap_bit(dlm, res, node);
}



On Thu, Aug 1, 2013 at 5:05 AM, Xue jiufei <xuejiufei@huawei.com> wrote:

> Function dlm_do_local_recovery_cleanup() should force clean refmap if
> the owner of lockres is UNKNOWN. Otherwise node may hang when umounting
> filesystems.
> Here's the situation:
>
>         Node1                                    Node2
> dlmlock()
>   -> dlm_get_lock_resource()
> send DLM_MASTER_REQUEST_MSG to
> other nodes.
>
>                                        trying to master this lockres,
>                                        return MAYBE.
>
> selected as the master of lockresA,
> set mle->master to Node1,
> and do assert_master,
> send DLM_ASSERT_MASTER_MSG to Node2.
>                                        Node 2 has interest on lockresA
>                                        and return
>                                        DLM_ASSERT_RESPONSE_MASTERY_REF
>                                        then something happened and
>                                        Node2 crashed.
>
> receiving DLM_ASSERT_RESPONSE_MASTERY_REF,
> set Node2 into refmap, and keep sending
> DLM_ASSERT_MASTER_MSG to other nodes
>
> o2hb found node2 down, calling
> dlm_hb_node_down()
> --> dlm_do_local_recovery_cleanup()
> the master of lockresA is still UNKNOWN,
> no need to call dlm_free_dead_locks().
>
> set the master of lockresA to Node1, but
> Node2 stills remains in refmap.
>
> when Node1 umount, it found that the refmap of lockresA is not empty
> and attempted to migrate it to Node2, But Node2 is already down,
> so umount hang, trying to migrate lockresA again and again.
>
> Signed-off-by: joyce <xuejiufei@huawei.com>
> ---
>  fs/ocfs2/dlm/dlmrecovery.c |   18 +++++++++++++++++-
>  1 file changed, 17 insertions(+), 1 deletion(-)
>
> diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
> index 773bd32..7b4413d 100644
> --- a/fs/ocfs2/dlm/dlmrecovery.c
> +++ b/fs/ocfs2/dlm/dlmrecovery.c
> @@ -2191,6 +2191,21 @@ static void dlm_revalidate_lvb(struct dlm_ctxt *dlm,
>         }
>  }
>
> +static void dlm_force_clean_refmap(struct dlm_ctxt *dlm,
> +               struct dlm_lock_resource *res, u16 dead_node)
> +{
> +       assert_spin_locked(&dlm->spinlock);
> +       assert_spin_locked(&res->spinlock);
> +
> +       if (test_bit(dead_node, res->refmap)) {
> +               mlog(0, "%s:%.*s: dead node %u had a ref, but had "
> +                               "no locks and had not purged before
> dying\n",
> +                               dlm->name, res->lockname.len,
> +                               res->lockname.name, dead_node);
> +               dlm_lockres_clear_refmap_bit(dlm, res, dead_node);
> +       }
> +}
> +
>  static void dlm_free_dead_locks(struct dlm_ctxt *dlm,
>                                 struct dlm_lock_resource *res, u8
> dead_node)
>  {
> @@ -2328,7 +2343,8 @@ static void dlm_do_local_recovery_cleanup(struct
> dlm_ctxt *dlm, u8 dead_node)
>                         } else if (res->owner == dlm->node_num) {
>                                 dlm_free_dead_locks(dlm, res, dead_node);
>                                 __dlm_lockres_calc_usage(dlm, res);
> -                       }
> +                       } else if (res->owner ==
> DLM_LOCK_RES_OWNER_UNKNOWN)
> +                               dlm_force_clean_refmap(dlm, res,
> dead_node);
>                         spin_unlock(&res->spinlock);
>                 }
>         }
> --
> 1.7.9.7
>
>
> _______________________________________________
> Ocfs2-devel mailing list
> Ocfs2-devel@oss.oracle.com
> https://oss.oracle.com/mailman/listinfo/ocfs2-devel
>
diff mbox

Patch

diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 773bd32..7b4413d 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -2191,6 +2191,21 @@  static void dlm_revalidate_lvb(struct dlm_ctxt *dlm,
 	}
 }
 
+static void dlm_force_clean_refmap(struct dlm_ctxt *dlm,
+		struct dlm_lock_resource *res, u16 dead_node)
+{
+	assert_spin_locked(&dlm->spinlock);
+	assert_spin_locked(&res->spinlock);
+
+	if (test_bit(dead_node, res->refmap)) {
+		mlog(0, "%s:%.*s: dead node %u had a ref, but had "
+				"no locks and had not purged before dying\n",
+				dlm->name, res->lockname.len,
+				res->lockname.name, dead_node);
+		dlm_lockres_clear_refmap_bit(dlm, res, dead_node);
+	}
+}
+
 static void dlm_free_dead_locks(struct dlm_ctxt *dlm,
 				struct dlm_lock_resource *res, u8 dead_node)
 {
@@ -2328,7 +2343,8 @@  static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node)
 			} else if (res->owner == dlm->node_num) {
 				dlm_free_dead_locks(dlm, res, dead_node);
 				__dlm_lockres_calc_usage(dlm, res);
-			}
+			} else if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN)
+				dlm_force_clean_refmap(dlm, res, dead_node);
 			spin_unlock(&res->spinlock);
 		}
 	}