This fix is only needed in nautilus. The issue
was observed during upstream teuthology testing in CentOS 7.5
machines.
File "/usr/share/ceph/mgr/volumes/fs/async_cloner.py", line 153, in do_clone
bulk_copy(fs_handle, src_path, dst_path, should_cancel)
File "/usr/share/ceph/mgr/volumes/fs/async_cloner.py", line 144, in bulk_copy
cptree(source_path, dst_path)
File "/usr/share/ceph/mgr/volumes/fs/async_cloner.py", line 134, in cptree
sync_attrs(fs_handle, d_full_dst, stx)
File "/usr/share/ceph/mgr/volumes/fs/async_cloner.py", line 77, in sync_attrs
fs_handle.lchown(target_path, source_statx["uid"], source_statx["gid"])
The issue wasn't observed in master/octopus teuthology
testing.
Signed-off-by: Ramana Raja <rraja@redhat.com>
def sync_attrs(fs_handle, target_path, source_statx):
try:
- fs_handle.lchown(target_path, source_statx["uid"], source_statx["gid"])
+ fs_handle.lchown(target_path, int(source_statx["uid"]), int(source_statx["gid"]))
fs_handle.lutimes(target_path, (time.mktime(source_statx["atime"].timetuple()),
time.mktime(source_statx["mtime"].timetuple())))
except cephfs.Error as e: