Hi Ian,
Here is another trivial patch. The connectathon test sweet does some
negative testing. Here is a malformed entry that we will attempt to mount:
b1c /server.example.com
It turns out that we think this is a multimount entry, and the what string
is actually NULL! The attached patch catches this before handing off to
the mount module. This was the least intrusive fix I could think of, but
this could warrant further debate as to where in the parse code this should
be handled.
-Jeff
--- autofs-4.1.4/modules/parse_sun.c.orig 2005-04-29 16:34:40.133808256
-0400
+++ autofs-4.1.4/modules/parse_sun.c 2005-04-29 16:36:27.975413840 -0400
@@ -678,6 +678,10 @@ static int sun_mount(const char *root, c
"mounting root %s, mountpoint %s, what %s, fstype %s, options %s\n",
root, mountpoint, what, fstype, options);
+ /* A malformed entry of the form key /xyz will trigger this case */
+ if (!what || *what == '\0')
+ return 1;
+
if (!strcmp(fstype, "nfs")) {
rv = mount_nfs->mount_mount(root, mountpoint,
strlen(mountpoint),
what, fstype, options,
mount_nfs->context);
_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs
