tar: postpone creation of symlinks with "suspicious" targets. Closes 8411

function                                             old     new   delta
data_extract_all                                     968    1038     +70
tar_main                                             952     986     +34
scan_tree                                            258     262      +4
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 3/0 up/down: 108/0)             Total: 108 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

3 files changed, 71 insertions, 24 deletions

diff --git a/archival/libarchive/data_extract_all.c b/archival/libarchive/data_extract_all.c
index 1830ffb8d..1ce927c2f 100644
--- a/archival/libarchive/data_extract_all.c
+++ b/archival/libarchive/data_extract_all.c
@@ -128,10 +128,11 @@ void FAST_FUNC data_extract_all(archive_handle_t *archive_handle)
 		res = link(hard_link, dst_name);
 		if (res != 0 && !(archive_handle->ah_flags & ARCHIVE_EXTRACT_QUIET)) {
 			/* shared message */
-			bb_perror_msg("can't create %slink "
-					"%s to %s", "hard",
+			bb_perror_msg("can't create %slink '%s' to '%s'",
+					"hard",
 					dst_name,
-					hard_link);
+					hard_link
+			);
 		}
 		/* Hardlinks have no separate mode/ownership, skip chown/chmod */
 		goto ret;
@@ -178,15 +179,44 @@ void FAST_FUNC data_extract_all(archive_handle_t *archive_handle)
 	case S_IFLNK:
 		/* Symlink */
 //TODO: what if file_header->link_target == NULL (say, corrupted tarball?)
+
+		/* To avoid a directory traversal attack via symlinks,
+		 * for certain link targets postpone creation of symlinks.
+		 *
+		 * For example, consider a .tar created via:
+		 *  $ tar cvf bug.tar anything.txt
+		 *  $ ln -s /tmp symlink
+		 *  $ tar --append -f bug.tar symlink
+		 *  $ rm symlink
+		 *  $ mkdir symlink
+		 *  $ tar --append -f bug.tar symlink/evil.py
+		 *
+		 * This will result in an archive that contains:
+		 *  $ tar --list -f bug.tar
+		 *  anything.txt
+		 *  symlink [-> /tmp]
+		 *  symlink/evil.py
+		 *
+		 * Untarring bug.tar would otherwise place evil.py in '/tmp'.
+		 */
+		if (file_header->link_target[0] == '/'
+		 || strstr(file_header->link_target, "..")
+		) {
+			llist_add_to(&archive_handle->symlink_placeholders,
+				xasprintf("%s%c%s", file_header->name, '\0', file_header->link_target)
+			);
+			break;
+		}
 		res = symlink(file_header->link_target, dst_name);
 		if (res != 0
 		 && !(archive_handle->ah_flags & ARCHIVE_EXTRACT_QUIET)
 		) {
 			/* shared message */
-			bb_perror_msg("can't create %slink "
-				"%s to %s", "sym",
+			bb_perror_msg("can't create %slink '%s' to '%s'",
+				"sym",
 				dst_name,
-				file_header->link_target);
+				file_header->link_target
+			);
 		}
 		break;
 	case S_IFSOCK:
diff --git a/archival/tar.c b/archival/tar.c
index 0fc574dfd..280ded4e1 100644
--- a/archival/tar.c
+++ b/archival/tar.c
@@ -22,24 +22,6 @@
  *
  * Licensed under GPLv2 or later, see file LICENSE in this source tree.
  */
-/* TODO: security with -C DESTDIR option can be enhanced.
- * Consider tar file created via:
- * $ tar cvf bug.tar anything.txt
- * $ ln -s /tmp symlink
- * $ tar --append -f bug.tar symlink
- * $ rm symlink
- * $ mkdir symlink
- * $ tar --append -f bug.tar symlink/evil.py
- *
- * This will result in an archive which contains:
- * $ tar --list -f bug.tar
- * anything.txt
- * symlink
- * symlink/evil.py
- *
- * Untarring it puts evil.py in '/tmp' even if the -C DESTDIR is given.
- * This doesn't feel right, and IIRC GNU tar doesn't do that.
- */
 
 //config:config TAR
 //config:	bool "tar (40 kb)"
@@ -296,6 +278,23 @@ static void chksum_and_xwrite(int fd, struct tar_header_t* hp)
 	xwrite(fd, hp, sizeof(*hp));
 }
 
+static void replace_symlink_placeholders(llist_t *list)
+{
+	while (list) {
+		char *target;
+
+		target = list->data + strlen(list->data) + 1;
+		if (symlink(target, list->data)) {
+			/* shared message */
+			bb_error_msg_and_die("can't create %slink '%s' to '%s'",
+				"sym",
+				list->data, target
+			);
+		}
+		list = list->link;
+	}
+}
+
 #if ENABLE_FEATURE_TAR_GNU_EXTENSIONS
 static void writeLongname(int fd, int type, const char *name, int dir)
 {
@@ -1252,6 +1251,8 @@ int tar_main(int argc UNUSED_PARAM, char **argv)
 	while (get_header_tar(tar_handle) == EXIT_SUCCESS)
 		bb_got_signal = EXIT_SUCCESS; /* saw at least one header, good */
 
+	replace_symlink_placeholders(tar_handle->symlink_placeholders);
+
 	/* Check that every file that should have been extracted was */
 	while (tar_handle->accept) {
 		if (!find_list_entry(tar_handle->reject, tar_handle->accept->data)
diff --git a/archival/tar_symlink_attack b/archival/tar_symlink_attack
new file mode 100755
index 000000000..35455f200
--- /dev/null
+++ b/archival/tar_symlink_attack
@@ -0,0 +1,16 @@
+#!/bin/sh
+# Makes "symlink attack" tarball (needs GNU tar for --append)
+
+true >anything.txt
+tar cvf tar_symlink_attack.tar anything.txt
+rm anything.txt
+
+ln -s /tmp symlink
+tar --append -f tar_symlink_attack.tar symlink
+rm symlink
+
+mkdir symlink
+echo BUG >symlink/bb_test_evilfile
+tar --append -f tar_symlink_attack.tar symlink/bb_test_evilfile
+rm symlink/bb_test_evilfile
+rmdir symlink