Change the remount command to be more container-friendly
Passing in MS_REMOUNT | MS_BIND tells the kernel to only remove the
MS_RDONLY flag and keep all the other flags the same. This is also the
only kind of remount that is allowed in Linux containers (to be more
precise, within user namespaces).
This change also attempts to always run the remount command when dealing
with /, since containers will almost always run a loop device, and since
the number of the device changes, it is not convenient to put it in the
fstab. Plus, the container won't have permission to modify it, but might
be able to perform the remount.
Bug: 72178046
Test: `adb remount` works in both sailfish and Chrome OS
Change-Id: I9e8ec8afcd57f67875a312824667768b3aa89faa
diff --git a/adb/remount_service.cpp b/adb/remount_service.cpp
index 32ed28f..a4c7a5a 100644
--- a/adb/remount_service.cpp
+++ b/adb/remount_service.cpp
@@ -62,9 +62,9 @@
// The proc entry for / is full of lies, so check fstab instead.
// /proc/mounts lists rootfs and /dev/root, neither of which is what we want.
-static std::string find_mount(const char* dir) {
- if (strcmp(dir, "/") == 0) {
- return find_fstab_mount(dir);
+static std::string find_mount(const char* dir, bool is_root) {
+ if (is_root) {
+ return find_fstab_mount(dir);
} else {
return find_proc_mount(dir);
}
@@ -86,17 +86,29 @@
if (!directory_exists(dir)) {
return true;
}
- std::string dev = find_mount(dir);
- if (dev.empty()) {
+ bool is_root = strcmp(dir, "/") == 0;
+ std::string dev = find_mount(dir, is_root);
+ // Even if the device for the root is not found, we still try to remount it
+ // as rw. This typically only happens when running Android in a container:
+ // the root will almost always be in a loop device, which is dynamic, so
+ // it's not convenient to put in the fstab.
+ if (dev.empty() && !is_root) {
return true;
}
- if (!make_block_device_writable(dev)) {
+ if (!dev.empty() && !make_block_device_writable(dev)) {
WriteFdFmt(fd, "remount of %s failed; couldn't make block device %s writable: %s\n",
dir, dev.c_str(), strerror(errno));
return false;
}
+ if (mount(dev.c_str(), dir, "none", MS_REMOUNT | MS_BIND, nullptr) == -1) {
+ // This is useful for cases where the superblock is already marked as
+ // read-write, but the mount itself is read-only, such as containers
+ // where the remount with just MS_REMOUNT is forbidden by the kernel.
+ WriteFdFmt(fd, "remount of the %s mount failed: %s.\n", dir, strerror(errno));
+ return false;
+ }
if (mount(dev.c_str(), dir, "none", MS_REMOUNT, nullptr) == -1) {
- WriteFdFmt(fd, "remount of %s failed: %s\n", dir, strerror(errno));
+ WriteFdFmt(fd, "remount of the %s superblock failed: %s\n", dir, strerror(errno));
return false;
}
return true;