dlm: add new flag DLM_RELEASE_RECOVER for dlm_lockspace_release

When dlm_lockspace_release() is passed DLM_RELEASE_RECOVER, it
tells the dlm to handle the release/leave as if the node had failed,
i.e. perform recovery steps for a failed node, like recover_slot().

When DLM_RELEASE_RECOVER is set:
- dlm_release_lockspace() includes RELEASE_RECOVER=1 in the OFFLINE
  uevent sent to userspace.
- userspace/dlm_controld sends a message to all lockspace members
  indicating that the subsequent node removal should be handled as
  if the node had failed.
- when dlm_controld on all nodes receives the new message, it sets
  the release_recover configfs entry to 1 for the node.
- when the dlm/kernel next performs recovery and removes the node,
  it will see that release_recover has been set, and will perform
  recovery steps for the node as if it had failed, e.g. the
  recover_slot() callback is called to notify the fs.

Signed-off-by: Alexander Aring <aahringo@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>

diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 6ff666a511c7ae0f3b0b0823b188d47fc5b9b733..d986b7ef153dcf214dae6533c450a91fb0cd9e38 100644 (file)
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -738,7 +738,7 @@ static int release_lockspace(struct dlm_ls *ls, int release_option)
 
        if (release_option != DLM_RELEASE_NO_EVENT &&
            dlm_user_daemon_available())
-               do_uevent(ls, 0, 0);
+               do_uevent(ls, 0, (release_option == DLM_RELEASE_RECOVER));
 
        dlm_recoverd_stop(ls);
 

diff --git a/fs/dlm/member.c b/fs/dlm/member.c
index 152d2cb16f59143c55968afc11e58d82fec1c156..356337102015ec97acca02083a07c294aef97692 100644 (file)
--- a/fs/dlm/member.c
+++ b/fs/dlm/member.c
@@ -478,7 +478,8 @@ static void dlm_lsop_recover_prep(struct dlm_ls *ls)
        ls->ls_ops->recover_prep(ls->ls_ops_arg);
 }
 
-static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb)
+static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb,
+                                 unsigned int release_recover)
 {
        struct dlm_slot slot;
        uint32_t seq;
@@ -495,7 +496,7 @@ static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb)
 
        error = dlm_comm_seq(memb->nodeid, &seq, false);
 
-       if (!error && seq == memb->comm_seq)
+       if (!release_recover && !error && seq == memb->comm_seq)
                return;
 
        slot.nodeid = memb->nodeid;
@@ -552,6 +553,7 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
        struct dlm_member *memb, *safe;
        struct dlm_config_node *node;
        int i, error, neg = 0, low = -1;
+       unsigned int release_recover;
 
        /* previously removed members that we've not finished removing need to
         * count as a negative change so the "neg" recovery steps will happen
@@ -572,8 +574,12 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
                if (node && !node->new && !node->gone)
                        continue;
 
+               release_recover = 0;
+
                if (node->gone) {
-                       log_rinfo(ls, "remove member %d", memb->nodeid);
+                       release_recover = node->release_recover;
+                       log_rinfo(ls, "remove member %d%s", memb->nodeid,
+                                 release_recover ? " (release_recover)" : "");
                } else {
                        /* removed and re-added */
                        log_rinfo(ls, "remove member %d comm_seq %u %u",
@@ -584,7 +590,7 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
                list_move(&memb->list, &ls->ls_nodes_gone);
                remove_remote_member(memb->nodeid);
                ls->ls_num_nodes--;
-               dlm_lsop_recover_slot(ls, memb);
+               dlm_lsop_recover_slot(ls, memb, release_recover);
        }
 
        /* add new members to ls_nodes */

diff --git a/include/linux/dlm.h b/include/linux/dlm.h
index cc7a36244893d0baeb4baa864df31b7edd5039c0..108eb953eb18110ad3f78103266df2d5ab1aec90 100644 (file)
--- a/include/linux/dlm.h
+++ b/include/linux/dlm.h
@@ -103,11 +103,16 @@ int dlm_new_lockspace(const char *name, const char *cluster,
  *   a leave event to the cluster manager, so other nodes will
  *   not be notified that the node should be removed from the
  *   list of lockspace members.
+ *
+ * DLM_RELEASE_RECOVER like DLM_RELEASE_NORMAL, but the remaining
+ *   nodes will handle the removal of the node as if the node
+ *   had failed, e.g. the recover_slot() callback would be used.
  */
 #define DLM_RELEASE_NO_LOCKS           0
 #define DLM_RELEASE_UNUSED             1
 #define DLM_RELEASE_NORMAL             2
 #define DLM_RELEASE_NO_EVENT           3
+#define DLM_RELEASE_RECOVER            4
 
 /*
  * dlm_release_lockspace