ocfs2: special case recovery lock in dlmlock_remote()
If the previous master of the recovery lock dies, let calc_usage take it
down completely and let the caller completely redo the dlmlock() call.
Otherwise, there will never be an opportunity to re-master the lockres and
recovery wont be able to progress.
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index 0ff9348..20b38dc 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -227,7 +227,16 @@
res->state &= ~DLM_LOCK_RES_IN_PROGRESS;
lock->lock_pending = 0;
if (status != DLM_NORMAL) {
- if (status != DLM_NOTQUEUED) {
+ if (status == DLM_RECOVERING &&
+ dlm_is_recovery_lock(res->lockname.name,
+ res->lockname.len)) {
+ /* recovery lock was mastered by dead node.
+ * we need to have calc_usage shoot down this
+ * lockres and completely remaster it. */
+ mlog(0, "%s: recovery lock was owned by "
+ "dead node %u, remaster it now.\n",
+ dlm->name, res->owner);
+ } else if (status != DLM_NOTQUEUED) {
/*
* DO NOT call calc_usage, as this would unhash
* the remote lockres before we ever get to use
@@ -691,18 +700,22 @@
msleep(100);
/* no waiting for dlm_reco_thread */
if (recovery) {
- if (status == DLM_RECOVERING) {
- mlog(0, "%s: got RECOVERING "
- "for $REOCVERY lock, master "
- "was %u\n", dlm->name,
- res->owner);
- dlm_wait_for_node_death(dlm, res->owner,
- DLM_NODE_DEATH_WAIT_MAX);
- }
+ if (status != DLM_RECOVERING)
+ goto retry_lock;
+
+ mlog(0, "%s: got RECOVERING "
+ "for $RECOVERY lock, master "
+ "was %u\n", dlm->name,
+ res->owner);
+ /* wait to see the node go down, then
+ * drop down and allow the lockres to
+ * get cleaned up. need to remaster. */
+ dlm_wait_for_node_death(dlm, res->owner,
+ DLM_NODE_DEATH_WAIT_MAX);
} else {
dlm_wait_for_recovery(dlm);
+ goto retry_lock;
}
- goto retry_lock;
}
if (status != DLM_NORMAL) {
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 86199f6..00209f4 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -2314,6 +2314,10 @@
mlog(0, "%s: reco master %u is ready to recover %u\n",
dlm->name, dlm->reco.new_master, dlm->reco.dead_node);
status = -EEXIST;
+ } else if (ret == DLM_RECOVERING) {
+ mlog(0, "dlm=%s dlmlock says master node died (this=%u)\n",
+ dlm->name, dlm->node_num);
+ goto again;
} else {
struct dlm_lock_resource *res;
