39

The command mount.cifs is found not being able to run in a gentoo system with systemd

ae429-1105 etc # mount -t cifs //file.abc.edu.au/user /home/directory/path -o credentials=/etc/user,rw,iocharset=utf8,file_mode=0777,dir_mode=0777
mount error(2): No such file or directory
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

It has been confirmed that the existence and accessibility of the mountpoint /home/directory/path and credential file /etc/user. Also the relevant modules and services has been enabled, i.e.,

 ae429-1105 etc # lsmod |egrep 'fuse|cifs'
 fuse                   72589  5 
 cifs                  312131  0

and

ae429-1105 etc # systemctl -t service -a |grep Samba
nmbd.service                         loaded active   running Samba NetBIOS                     name server
smbd.service                         loaded active   running Samba SMB/CIFS     server
winbindd.service                     loaded inactive dead    Samba Winbind daemon

This problem has been identified by many users, e.g. one example . ALSO NOTE that the same command executed in my Ubuntu/debian system is able to mount successfully.

Other information in the problematic machine:

ae429-1105 etc # mount.cifs --version
mount.cifs version: 6.1

the version of mount.cifs installed in debian/ubuntu is 6.0

Chenming Zhang
  • 1,037
  • 1
  • 12
  • 14

14 Answers14

22

You might need to provide the vers= option to the mount command to force version 3.0 if you're trying to mount a share from a newer version of Windows. One of our fileservers was recently upgraded to 2012R2 and that's when my mount stopped working. Setting it to vers=3.0 fixed the issue. Like most Samba/CIFS errors the "No such file or directory" message isn't much help.

As an example:

# mount -t cifs //win2012r2/someshare -o cred=/home/foo/.cifs_user,vers=3.0 /mnt/tmp

..where I have my domain, username and password contained in the .cifs_user file:

user=MyUser
password=MyPassword
domain=MyDomain

Apparently, smbmount uses a newer version of the SMB protocol by default since it worked without issue or any special options.

Notice below that the default protocol version is 1.0.

From the mount.cifs man page:

vers=arg
       SMB protocol version. Allowed values are:

   · 1.0 - The classic CIFS/SMBv1 protocol.

   · 2.0 - The SMBv2.002 protocol. This was initially introduced in Windows Vista Service Pack 1, and Windows Server 2008. Note  that  the  initial  release
     version of Windows Vista spoke a slightly different dialect (2.000) that is not supported.

   · 2.1 - The SMBv2.1 protocol that was introduced in Microsoft Windows 7 and Windows Server 2008R2.

   · 3.0 - The SMBv3.0 protocol that was introduced in Microsoft Windows 8 and Windows Server 2012.

   · 3.02 or 3.0.2 - The SMBv3.0.2 protocol that was introduced in Microsoft Windows 8.1 and Windows Server 2012R2.

   · 3.1.1 or 3.11 - The SMBv3.1.1 protocol that was introduced in Microsoft Windows 10 and Windows Server 2016.

   · 3 - The SMBv3.0 protocol version and above.

   · default - Tries to negotiate the highest SMB2+ version supported by both the client and server.

   If no dialect is specified on mount vers=default is used.  To check Dialect refer to /proc/fs/cifs/DebugData

   Note too that while this option governs the protocol version used, not all features of each version are available.

   The default since v4.13.5 is for the client and server to negotiate the highest possible version greater than or equal to 2.1. In kernels prior to v4.13,
   the default was 1.0. For kernels between v4.13 and v4.13.5 the default is 3.0.
foobrew
  • 344
  • I had a similar problem with the "nounix" flag which must not be supported in v1.0. Changing to v2.0 (the most recent available for me) fixed the problem. Also file permissions are more sensible with vers=2.0 (755 instead of 777) – cxrodgers Aug 12 '17 at 21:10
  • 4
    Thanks a lot for the solution related to vers= option! It worked for me, only backwards...After upgrading opensuse leap from 42.3 version to 15.1, an fstab entry for mounting a network drive, which worked, stopped working in 15.1. I used the option vers=1.0 and guess what...Probably leap 15.1 uses a newer version of SMB protocol that was not able to find the remote directory. – John Jul 13 '19 at 14:54
  • 1
    Connecting to a share hosted on Windows Server 2003 from Ubuntu 19.04 continually failed for me until I added vers=1.0 to my options list. Thanks! – user8675309 Aug 12 '19 at 21:03
  • It worked for me, EXCEPT: I had to state Version TWO vers=2.0 to mount the samba shares of my 5-year-old NAS system... with 3.0 I got above error. – Frank N Sep 08 '19 at 13:15
  • etc/fstab users: Just put that vers=3.0 (or 2.0...) right and spaceless before your other options, i.e. vers=2.0,guest,uid=1000,iocharset… – Frank N Sep 08 '19 at 13:16
  • In my case it was needed to explicitly state vers=1.0. – Alex Yu Aug 27 '20 at 14:33
13

Try to install the package keyutils:

sudo apt-get install keyutils

Not sure exactly why this helps, maybe someone else has an answer here. But at least it did the trick for me: with keyutils the cifs mount worked just fine.

Klaus
  • 239
  • Please add some information on how this would resolve the problem stated in the question. What does this package do, and how does it figure into the problem raised by the OP? – Haxiel Mar 20 '19 at 15:03
  • 1
    Good question. I am not sure how the package keyutils helps. In my case at least this is what did the trick. After install of keyutils, my cifs mount worked just fine, whereas before I did get the error message "mount error(2): No such file or directory", just as in the OP. – Klaus Mar 24 '19 at 11:41
  • Duplicate of this other answer – Chris Davies Jul 13 '19 at 15:36
  • It's works for ubuntu 20.04 ... I am not sure how the package keyutils helps too.. – W. Dan May 12 '22 at 08:51
11

Can you use nodfs option? ie for your -o options input pass the input as below.

-o credentials=/etc/user,rw,iocharset=utf8,file_mode=0777,dir_mode=0777,nodfs

i.e appended ,nodfs

It worked for me.

αғsнιη
  • 41,407
Sanath
  • 111
6

I ran into this on Ubuntu 18.04. The problem was I needed the keyutils package to do Kerberos authentication (sec=krb5 mount option), which was not installed along with cifs-utils (which provided mount.cifs). I'm not sure if the package name is the same on Gentoo or not. (Thanks to https://forum.zentyal.org/index.php?topic=18601.0 for the solution.)

Chris
  • 331
4

You might need to change the sec parameter: this setting made it work on my setup:

mount.cifs ... -o sec=ntlm

Relevant extract of man mount.cifs:

sec= Security mode. Allowed values are:

  • none - attempt to connection as a null user (no name)
  • krb5 - Use Kerberos version 5 authentication
  • krb5i - Use Kerberos authentication and forcibly enable packet signing
  • ntlm - Use NTLM password hashing
  • ntlmi - Use NTLM password hashing and force packet signing
  • ntlmv2 - Use NTLMv2 password hashing
  • ntlmv2i - Use NTLMv2 password hashing and force packet signing
  • ntlmssp - Use NTLMv2 password hashing encapsulated in Raw NTLMSSP message

  • ntlmsspi - Use NTLMv2 password hashing encapsulated in Raw NTLMSSP message, and force packet signing

    The default in mainline kernel versions prior to v3.8 was sec=ntlm. In v3.8, the default was changed to sec=ntlmssp.

    If the server requires signing during protocol negotiation, then it may be enabled automatically. Packet signing may also be enabled automatically if it's enabled in /proc/fs/cifs/SecurityFlags.

Benoit Duffez
  • 505
2

I wanted to add another source of this problem I encountered today. Once you change the user id of a unix user, the smb user created via smbpasswd may not able to authenticate for the samba share anymore resulting in the same error.

So if you changed your unix user id via usermod -u 1000 my_user then you may run into problems. The fix for me was to delete and re-add the smb user afterwards:

smbpasswd -x my_user
smbpasswd -a my_user
Ryad
  • 21
  • 1
    While true, how is this related to the original question? – RalfFriedl Jun 16 '19 at 14:12
  • 2
    As I said, if you change the user id of a user, the same error of the original question appears. So if anyone did the same and finds this thread, he or she might find my hint useful. – Ryad Jun 17 '19 at 19:24
1

I was running into this same "mount error(2): No such file or directory" error using mount.cifs on a CentOS 7 VM. I never determined exactly why the error was being generated when using the default ntlm security (and the variants), but I did discover that using Kerberos authentication worked around the problem. So my final working command line looked like this:

mount.cifs -v -o domain=MYCODOMAIN,sec=krb5 //winserver/sharename /mnt/mymountpoint

whereas this command that gave the "no such file or directory" error was:

mount.cifs -v -o username=myusername,domain=MYCODOMAIN //winserver/sharename /mnt/mymountpoint

To use Kerberos, I installed "krb5-workstation" package and configured it.

Mark Edington
  • 523
1

Add a $ to the end, like this //winserver/sharename$

mount.cifs -v -o username=myusername,domain=MYCODOMAIN //winserver/sharename$ /mnt/mymountpoint
Jeff Schaller
  • 67,283
  • 35
  • 116
  • 255
Fahri çetin
  • 11
  • Wow! Any idea what the '$' does? It fixed it for me but no clue to why – Gabriel Fair Jun 28 '19 at 20:22
  • 1
    The $ sign is an administrative share in the Windows share context, if activated by the system, a user with administrative rights can access all paths. Example \MY-SERVER\c$ – Phil795 Oct 25 '19 at 09:31
1

With me it worked by putting "vers = 1.0" as before -> credentials = / root / .dbx.credentials, vers = 1.0, uid = 1001, gid = 100, rw

Carlos Roberto da Silva Junior
  • 11
1

Could you have just upgraded the linux kernel?

Are you on a corporate network?

If so - try disabling IPv6 in /etc/sysctl.conf

to do this open it with sudo nano /etc/sysctl.conf and add the following:

net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.lo.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1

And then run sudo sysctl -p

Remount and try again.

This worked for me when I upgraded LTS all the way from 16.04 through 18.04 to 20.04, and I had this problem.

RGD2
  • 199
0

One solution might be to manually install keyutils since it is not a (hard) dependency of cifs-utils anymore.

Information why keyutils are not installed anymore can be found here: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=822841

And a launchpad bug report can be found here: https://bugs.launchpad.net/ubuntu/+source/cifs-utils/+bug/1772148

tuk8
  • 11
0

Apparently samba throws this entirely un-helpful error in an astoundingly wide range of cases. In my case, the issue was an option configured for the share in the server's smb.conf (vfs objects = vfs_fruit) which the server didn't like, even though testparm didn't complain about it.

ghostly_s
  • 164
0

This error could mean a variety of things. Scenario I saw was that Samba on the server had been updated, but the service was not restarted. This caused it to have errors loading shared modules. Restarting Samba on the server resolved. So: check the server logs and consider restarting Samba.

Andrew Marshall
  • 2,990
0

I got this error because I was using a hardened Ubuntu VM and the firewall was blocking the outgoing connection. I checked sudo ufw status verbose, found outgoing connections were not always allowed, and then added the relevant ports:

ufw allow out from any to a.b.c.d port 137 proto udp
ufw allow out from any to a.b.c.d port 138 proto udp
ufw allow out from any to a.b.c.d port 139 proto tcp
ufw allow out from any to a.b.c.d port 445 proto tcp

where a.b.c.d should be replaced with the remote server's IP address.

EDIT: The error output included the line:

Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) and kernel log messages (dmesg)

so I ran dmesg and spotted these two lines, which set me on the right track:

[ 6331.419294] CIFS VFS: Error connecting to socket. Aborting operation.
[ 6331.419581] CIFS VFS: cifs_mount failed w/return code = -2
IpsRich
  • 767