diff options
| author | Srinivas Eeda <srinivas.eeda@oracle.com> | 2007-04-17 13:26:33 -0700 |
|---|---|---|
| committer | Mark Fasheh <mark.fasheh@oracle.com> | 2007-04-26 13:33:02 -0700 |
| commit | 756a1501ddbbe73098aa031939460930f6edc9cd (patch) | |
| tree | 571202550a811b937277e227b27c26a0fcdf8b7c | |
| parent | de46c33745f5e2ad594c72f2cf5f490861b16ce1 (diff) | |
ocfs2_dlm: fix race in dlm_remaster_locks
There is a possibility that dlm_remaster_locks could overwride node->state
with DLM_RECO_NODE_DATA_REQUESTED after dlm_reco_data_done_handler sets the
node->state to DLM_RECO_NODE_DATA_DONE. This could lead to recovery getting
stuck and requires a cluster reboot. Synchronize with dlm_reco_state_lock
spinlock.
Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
| -rw-r--r-- | fs/ocfs2/dlm/dlmrecovery.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 6d4a83d5015..c1807a42c49 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -611,6 +611,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) } } while (status != 0); + spin_lock(&dlm_reco_state_lock); switch (ndata->state) { case DLM_RECO_NODE_DATA_INIT: case DLM_RECO_NODE_DATA_FINALIZE_SENT: @@ -641,6 +642,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) ndata->node_num, dead_node); break; } + spin_unlock(&dlm_reco_state_lock); } mlog(0, "done requesting all lock info\n"); |