I found lots of places where we assume error will be set when calling

one of our virtualised functions, such as db_open(), but error is only
set when a system call fails, and it is not uncommon for us to fail a
function internally without ever making a system call. That led to us
passing back success when a function had in fact failed.

I found two places where we relied on map_nt_error_from_unix()
returning success when errno==0, but lots and lots of places where we
relied on the reverse, so I fixed those two places.

map_nt_error_from_unix() will now always return an error, returning
NT_STATUS_UNSUCCESSFUL if errno is 0
(cherry picked from commit 69d40ca4c1af925d4b0e59ddc69ef8c26e6501d1)
(This used to be commit 834684a524a24bb4eb46b4af583d39947dc87d95)

1 files changed, 10 insertions, 2 deletions

diff --git a/source3/lib/errmap_unix.c b/source3/lib/errmap_unix.c
index 8194cf80cc..2cd2386c5c 100644
--- a/source3/lib/errmap_unix.c
+++ b/source3/lib/errmap_unix.c
@@ -107,8 +107,16 @@ NTSTATUS map_nt_error_from_unix(int unix_error)
 {
 	int i = 0;
 
-	if (unix_error == 0)
-		return NT_STATUS_OK;
+	if (unix_error == 0) {
+		/* we map this to an error, not success, as this
+		   function is only called in an error path. Lots of
+		   our virtualised functions may fail without making a
+		   unix system call that fails (such as when they are
+		   checking for some handle existing), so unix_error
+		   may be unset
+		*/
+		return NT_STATUS_UNSUCCESSFUL;
+	}
 
 	/* Look through list */
 	while(unix_dos_nt_errmap[i].unix_error != 0) {