@@ -760,8 +760,61 @@ static noinline int device_list_add(const char *path,
rcu_string_free(device->name);
rcu_assign_pointer(device->name, name);
if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state)) {
- fs_devices->missing_devices--;
- clear_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state);
+ int ret;
+ struct btrfs_fs_info *fs_info = fs_devices->fs_info;
+ fmode_t fmode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
+
+ if (btrfs_super_flags(disk_super) &
+ BTRFS_SUPER_FLAG_SEEDING)
+ fmode &= ~FMODE_WRITE;
+
+ /*
+ * Missing can be set only when FS is mounted.
+ * So here its always fs_devices->opened > 0 and most
+ * of the struct device members are already updated by
+ * the mount process even if this device was missing, so
+ * now follow the normal open device procedure for this
+ * device. The scrub will take care of filling the
+ * missing stripes for raid56 and balance for raid1 and
+ * raid10.
+ */
+ ASSERT(fs_devices->opened);
+ mutex_lock(&fs_devices->device_list_mutex);
+ mutex_lock(&fs_info->chunk_mutex);
+ /*
+ * As of now do not fail the dev scan thread for the
+ * reason that btrfs_open_one_device() fails and keep
+ * the legacy dev scan requisites as it is.
+ * And reset missing only if open is successful, as
+ * user can rerun dev scan after fixing the device
+ * for which the device open (below) failed.
+ */
+ ret = btrfs_open_one_device(fs_devices, device, fmode,
+ fs_info->bdev_holder);
+ if (!ret) {
+ fs_devices->missing_devices--;
+ clear_bit(BTRFS_DEV_STATE_MISSING,
+ &device->dev_state);
+ btrfs_clear_opt(fs_info->mount_opt, DEGRADED);
+ btrfs_warn(fs_info,
+ "BTRFS: device %s devid %llu joined\n",
+ path, devid);
+ }
+
+ if (test_bit(BTRFS_DEV_STATE_WRITEABLE,
+ &device->dev_state) &&
+ !test_bit(BTRFS_DEV_STATE_REPLACE_TGT,
+ &device->dev_state)) {
+ fs_devices->total_rw_bytes +=
+ device->total_bytes;
+ atomic64_add(device->total_bytes -
+ device->bytes_used,
+ &fs_info->free_chunk_space);
+ }
+ set_bit(BTRFS_DEV_STATE_IN_FS_METADATA,
+ &device->dev_state);
+ mutex_unlock(&fs_info->chunk_mutex);
+ mutex_unlock(&fs_devices->device_list_mutex);
}
}
If the device is not present at the time of (-o degrade) mount, the mount context will create a dummy missing struct btrfs_device. Later this device may reappear after the FS is mounted and then device is included in the device list but it missed the open_device part. So this patch handles that case by going through the open_device steps which this device missed and finally adds to the device alloc list. So now with this patch, to bring back the missing device user can run, btrfs dev scan <path-of-missing-device> Without this kernel patch, even though 'btrfs fi show' and 'btrfs dev ready' would tell you that missing device has reappeared successfully but actually in kernel FS layer it didn't. Signed-off-by: Anand Jain <anand.jain@oracle.com> --- This patch needs: [PATCH 0/4] factor __btrfs_open_devices() v2: Add more comments. Add more change log. Add to check if device missing is set, to handle the case dev open fail and user will rerun the dev scan v3: Reword comments in the code. The device missing check added in v2, is sent as a separate patch [patch] btrfs: fix inconsistency during missing device rejoin fs/btrfs/volumes.c | 57 ++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 55 insertions(+), 2 deletions(-)