Just throwing this out so that people can search on it: I now have an example file with a filename length of 100 characters, however most will be wide characters since its mainly Japanese - in python:
=================================
len(file_name)
100
len(bytes(file_name.encode()))
246
=================================
So in UTF-8, its more than double the length! Presumably a character
won't be more than 2 bytes long, so hidden characters??
Interacting with this in aufs is causing 'File name too long' errors
(rsync reports error code 36).
Reading the manpage, I think this is covered with:
=====================================================================
The whiteout prefix (.wh.) is reserved on all branches. Users
should not handle the filename
begins with this prefix. In order to future whiteout, the
maximum filename length is limited
by the longest value - 4 * 2 - 1 - 4 = 242. It means you
cannot handle such long name in
aufs, even if it surely exists on the underlying branch fs. The
readdir(3)/getdents(2) call
show you such name, but the d_type is set to DT_UNKNOWN. It may
be a violation of POSIX.
=====================================================================
246 is pretty close to the extfs/btrfs max of 255B, so I doubt its going
to get 'solved' any time soon.
The name is stupidly long anyway, wide characters or not.
signature.asc
Description: OpenPGP digital signature
------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports.http://sdm.link/zohodev2dev
