aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorDavid Teigland <teigland@redhat.com>2006-10-31 11:56:01 -0600
committerSteven Whitehouse <swhiteho@redhat.com>2006-11-30 10:35:13 -0500
commit91c0dc93a1a6bbdd79707ed311e48b4397df177f (patch)
treeec78620714a49fc41244afb667404b0ff825b1a2
parentd4400156d415540086c34a06e5d233122d6bf56a (diff)
[DLM] fix aborted recovery during node removal
Red Hat BZ 211914 With the new cluster infrastructure, dlm recovery for a node removal can be aborted and restarted for a node addition. When this happens, the restarted recovery isn't aware that it's doing recovery for the earlier removal as well as the addition. So, it then skips the recovery steps only required when nodes are removed. This can result in locks not being purged for failed/removed nodes. The fix is to check for removed nodes for which recovery has not been completed at the start of a new recovery sequence. Signed-off-by: David Teigland <teigland@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
-rw-r--r--fs/dlm/member.c8
-rw-r--r--fs/dlm/recoverd.c7
2 files changed, 15 insertions, 0 deletions
diff --git a/fs/dlm/member.c b/fs/dlm/member.c
index a3f7de7f3a8..85e2897bd74 100644
--- a/fs/dlm/member.c
+++ b/fs/dlm/member.c
@@ -186,6 +186,14 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
struct dlm_member *memb, *safe;
int i, error, found, pos = 0, neg = 0, low = -1;
+ /* previously removed members that we've not finished removing need to
+ count as a negative change so the "neg" recovery steps will happen */
+
+ list_for_each_entry(memb, &ls->ls_nodes_gone, list) {
+ log_debug(ls, "prev removed member %d", memb->nodeid);
+ neg++;
+ }
+
/* move departed members from ls_nodes to ls_nodes_gone */
list_for_each_entry_safe(memb, safe, &ls->ls_nodes, list) {
diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c
index 362e3eff4dc..4a1d6023fd9 100644
--- a/fs/dlm/recoverd.c
+++ b/fs/dlm/recoverd.c
@@ -164,6 +164,13 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv)
*/
dlm_recover_rsbs(ls);
+ } else {
+ /*
+ * Other lockspace members may be going through the "neg" steps
+ * while also adding us to the lockspace, in which case they'll
+ * be looking for this status bit during dlm_recover_locks().
+ */
+ dlm_set_recover_status(ls, DLM_RS_LOCKS);
}
dlm_release_root_list(ls);