autofs-5.1.7 - fix nonstrict offset mount fail handling
From: Ian Kent <raven@themaw.net>
If a triggered offset mount fails automount is not handling nonstrict
mount failure correctly.
The nonstrict mount failure handling needs to convert an offset mount
failure to a success if the offset subtree below the failed mount is not
empty otherwise it must return the failure. The previous implementation
used -1 to indicate the subtree was empty and that was used to detect
when the mount should fail instead of converting the fail to a success.
Make the new implementation do the same.
Signed-off-by: Ian Kent <raven@themaw.net>
---
CHANGELOG | 1 +
lib/mounts.c | 2 +-
modules/parse_sun.c | 2 +-
3 files changed, 3 insertions(+), 2 deletions(-)
--- autofs-5.1.7.orig/CHANGELOG
+++ autofs-5.1.7/CHANGELOG
@@ -76,6 +76,7 @@
- fix direct mount deadlock.
- fix lookup_prune_one_cache() refactoring change.
- add missing description of null map option.
+- fix nonstrict offset mount fail handling.
25/01/2021 autofs-5.1.7
- make bind mounts propagation slave by default.
--- autofs-5.1.7.orig/lib/mounts.c
+++ autofs-5.1.7/lib/mounts.c
@@ -1616,7 +1616,7 @@ static int tree_mapent_traverse_subtree(
{
struct traverse_subtree_context *ctxt = ptr;
struct mapent *oe = MAPENT(n);
- int ret = 1;
+ int ret = -1;
if (n->left) {
ret = tree_mapent_traverse_subtree(n->left, work, ctxt);
--- autofs-5.1.7.orig/modules/parse_sun.c
+++ autofs-5.1.7/modules/parse_sun.c
@@ -1181,7 +1181,7 @@ static int mount_subtree(struct autofs_p
* offsets to be mounted.
*/
rv = sun_mount(ap, name, name, namelen, loc, loclen, options, ctxt);
- if (rv == 0) {
+ if (rv <= 0) {
ret = tree_mapent_mount_offsets(me, 1);
if (!ret) {
tree_mapent_cleanup_offsets(me);