Hi Steve, Paulo, and maintainers of CIFS, We have come across a problem where kernels 5.0-rc1 and onwards cannot mount a multi tier DFS setup, while kernels 4.20 and below can mount the share fine. The DFS tiering structure looks like this: Domain virtual DFS (i.e. \\company.com\folders\share) |-- Domain controller DFS (i.e. \\regional-dc.company.com\folders\share) |-- Regional DFS Server (i.e. \\regional-dfs.company.com\folders\share) |-- Actual file server (i.e. \\regional-svr.company.com\share) On the 5.x series kernels, after getting the DFS referrals list through to the Regional DFS Server, which responds with the correct server/share, instead of going to the Actual file server, the kernel backtracks from the Regional DFS Server back to the Domain controller and requests the share there. Of course, this share does not exist on the Domain controller, as it only exists on the Actual file server, and the connection dies. We have collected a packet capture, and the flow looks like this: 4.18.0-21-generic Ubuntu kernel - Good Host request/response -------------------------------------------- ---------------------------------------------------- Domain controller / Domain DFS Root company.com\folders Domain controller / Domain DFS Root Referral List Regional Domain Controller / Domain DFS Root start convo Regional Domain Controller / Domain DFS Root <Regional Domain Controller>\Folders\Country\<Share> referral Regional Domain Controller / Domain DFS Root <Regional Domain Controller>\Folders\Country\<Share> referral Regional DFS server start convo Regional DFS server <Regional DFS Server>\Root\Country\<Share> Regional DFS server STATUS_PATH_NOT_COVERED Regional DFS server request referrals Regional DFS server Referral List Actual File Server convo started Actual File Server <Actual File Server>\<Share> Actual File Server Good response 5.0.0-26-generic Ubuntu kernel - Bad Host request/response -------------------------------------------- ------------------------------------------- Domain controller / Domain DFS Root company.com\folders Regional Domain Controller / Domain DFS Root start convo Regional Domain Controller / Domain DFS Root <Regional Domain Controller>\Folders\Country\<Share> Regional Domain Controller / Domain DFS Root STATUS_PATH_NOT_COVERED Regional DFS server start convo Regional DFS server <Regional DFS Server>\Root\Country\<Share> Regional DFS server STATUS_PATH_NOT_COVERED Regional Domain Controller / Domain DFS Root <Regional DFS Server>\Root\Country\<Share> Regional Domain Controller / Domain DFS Root STATUS_PATH_NOT_COVERED If you are interested in any parts of the packet capture, let us know and I will provide you with portions that you need. We also enabled CIFS dynamic debugging with: echo 'module cifs +p' > /sys/kernel/debug/dynamic_debug/control echo 'file fs/cifs/* +p' > /sys/kernel/debug/dynamic_debug/control echo 7 > /proc/fs/cifs/cifsFYI >From there the debugging output was more or less the same between the two kernel versions, until the problematic area: Linux 4.18: Full log: https://paste.ubuntu.com/p/D9XwBbvTXc/ Status code returned 0xc0000257 STATUS_PATH_NOT_COVERED fs/cifs/smb2maperror.c: Mapping SMB2 status code 0xc0000257 to POSIX err -66 fs/cifs/connect.c: build_unc_path_to_root: full_path=\\<Regional DFS Server>\Root\Country\<Share> fs/cifs/smb2ops.c: smb2_get_dfs_refer path <\<Regional DFS Server>\Root\Country\<Share>> fs/cifs/misc.c: num_referrals: 1 dfs flags: 0x2 ... fs/cifs/dns_resolve.c: dns_resolve_server_name_to_ip: resolved: <Actual File Server> to <IPV4 Address> fs/cifs/connect.c: Username: XXX // mounts the share successfully Linux 5.0: Full log: https://paste.ubuntu.com/p/9sXPj7WMQv/ Status code returned 0xc0000257 STATUS_PATH_NOT_COVERED fs/cifs/smb2maperror.c: Mapping SMB2 status code 0xc0000257 to POSIX err -66 fs/cifs/connect.c: build_unc_path_to_root: full_path=\\<Regional DFS Server>\Root\Country\<Share> fs/cifs/connect.c: build_unc_path_to_root: full_path=\\<Regional DFS Server>\Root\Country\<Share> fs/cifs/dfs_cache.c: do_dfs_cache_find: search path: \<Regional DFS Server>\Root\Country\<Share> fs/cifs/dfs_cache.c: do_dfs_cache_find: cache miss fs/cifs/dfs_cache.c: do_dfs_cache_find: DFS referral request for \<Regional DFS Server>\Root\Country\<Share> fs/cifs/smb2ops.c: smb2_get_dfs_refer path <\<Regional DFS Server>\Root\Country\<Share>> fs/cifs/smb2pdu.c: SMB2 IOCTL Status code returned 0xc0000225 STATUS_NOT_FOUND fs/cifs/smb2maperror.c: Mapping SMB2 status code 0xc0000225 to POSIX err -2 // mounting the share fails shortly after Since there seems to be problems with how referrals are handled, I examined the commit history, and came across the new DFS caching feature introduced in 5.0-rc1. I built a test kernel at: commit e7b602f43719fc6173ae86d2de8f6f07c6858ddd Author: Paulo Alcantara <palcantara@xxxxxxx> Date: Wed Nov 14 15:38:51 2018 -0200 Subject: cifs: Save TTL value when parsing DFS referrals Which is the commit directly before the first commit of the DFS caching feature, commit 54be1f6c1c37498bba557049df646cc239fa37e3 Author: Paulo Alcantara <palcantara@xxxxxxx> Date: Wed Nov 14 16:01:21 2018 -0200 Subject: cifs: Add DFS cache routines I also built a test kernel at the end of the patch series which implements the DFS caching feature, at: commit 14e92c5dc7a1a1d4a82fb7142b5642837fef962a Author: Steve French <stfrench@xxxxxxxxxxxxx> Date: Mon Dec 24 01:05:22 2018 -0600 Subject: cifs: Minor Kconfig clarification We then tried to mount the CIFS share, and found that: e7b602f43719fc6173ae86d2de8f6f07c6858ddd -> mounts successfully 14e92c5dc7a1a1d4a82fb7142b5642837fef962a -> fails to mount. So there is a problem somewhere in the DFS caching feature, covered in the following commits: https://paste.ubuntu.com/p/XcNwp3dVBV/ We have also tried 5.3.5 mainline kernel, and the issue is still present. We are available to help out with gathering more debugging data and trying test patches. Please let us know what data to collect, or patches to test, and we will collect whatever you need to get this fixed. Thanks, Matthew Ruffell Sustaining Engineer @ Canonical
