samba server mount points stopped working on CentOS 8 install, error: Failed to start SPNEGO handler for negprot OID list

Question

After about 6 months of smb working flawlessly on a home server, it is now failing to allow remote systems to mount with the following error message in /var/log/messages:

Jun 27 12:53:10 bike3 smbd[19385]: [2020/06/27 12:53:10.706872,  0] ../../source3/smbd/negprot.c:211(negprot_spnego)
Jun 27 12:53:10 bike3 smbd[19385]:  Failed to start SPNEGO handler for negprot OID list!

I am using a very basic smb.conf configuration, and have tried a variety of googled settings, with no luck:

[global]
        workgroup = WORKGROUP
        security = user
        log level = 3
        map to guest = bad user
        dns proxy = no
; tested various combinations:
        client use spnego = no
        client ntlmv2 auth = no
        client min protocol = SMB2
        client max protocol = SMB3


[pictures]
       comment = pictures
       path = /mnt/pictures
       public = yes
       browsable = yes
       writable = yes
       guest ok = yes
       read only = no

I have reinstalled all samba packages:

Reinstalled:
  samba-4.11.2-13.el8.x86_64                  samba-client-4.11.2-13.el8.x86_64            samba-client-libs-4.11.2-13.el8.x86_64      samba-common-4.11.2-13.el8.noarch
  samba-common-libs-4.11.2-13.el8.x86_64      samba-common-tools-4.11.2-13.el8.x86_64

I have tested from Windows 10 and OS X Mojave, both fail with the same error, here is the log level 3 in the log.smb:

[2020/06/27 13:06:11.367462,  3] ../../lib/util/access.c:371(allow_access)
  Allowed connection from 192.168.xxx.xxx (192.168.xxx.xxx)
[2020/06/27 13:06:11.368276,  3] ../../source3/smbd/oplock.c:1414(init_oplocks)
  init_oplocks: initializing messages.
[2020/06/27 13:06:11.368563,  3] ../../source3/smbd/server_exit.c:244(exit_server_common)
  Server exit (failed to receive smb request)
[2020/06/27 13:06:11.372050,  3] ../../lib/util/access.c:371(allow_access)
  Allowed connection from 192.168.1.197 (192.168.1.197)
[2020/06/27 13:06:11.372676,  3] ../../source3/smbd/oplock.c:1414(init_oplocks)
  init_oplocks: initializing messages.
[2020/06/27 13:06:11.372763,  3] ../../source3/smbd/process.c:1956(process_smb)
  Transaction 0 of length 73 (0 toread)
[2020/06/27 13:06:11.372787,  3] ../../source3/smbd/process.c:1549(switch_message)
  switch message SMBnegprot (pid 21109) conn 0x0
[2020/06/27 13:06:11.373194,  3] ../../source3/smbd/negprot.c:636(reply_negprot)
  Requested protocol [NT LM 0.12]
[2020/06/27 13:06:11.373220,  3] ../../source3/smbd/negprot.c:636(reply_negprot)
  Requested protocol [SMB 2.002]
[2020/06/27 13:06:11.373237,  3] ../../source3/smbd/negprot.c:636(reply_negprot)
  Requested protocol [SMB 2.???]
[2020/06/27 13:06:11.373469,  3] ../../source3/smbd/smb2_negprot.c:294(smbd_smb2_request_process_negprot)
  Selected protocol SMB2_FF
[2020/06/27 13:06:11.373856,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'gssapi_spnego' registered
[2020/06/27 13:06:11.373880,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'gssapi_krb5' registered
[2020/06/27 13:06:11.373895,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'gssapi_krb5_sasl' registered
[2020/06/27 13:06:11.373911,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'spnego' registered
[2020/06/27 13:06:11.373929,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'schannel' registered
[2020/06/27 13:06:11.373954,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'naclrpc_as_system' registered
[2020/06/27 13:06:11.373970,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'sasl-EXTERNAL' registered
[2020/06/27 13:06:11.373984,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'ntlmssp' registered
[2020/06/27 13:06:11.374000,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'ntlmssp_resume_ccache' registered
[2020/06/27 13:06:11.374016,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'http_basic' registered
[2020/06/27 13:06:11.374031,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'http_ntlm' registered
[2020/06/27 13:06:11.374048,  3] ../../auth/gensec/gensec_start.c:988(gensec_register)
  GENSEC backend 'http_negotiate' registered
[2020/06/27 13:06:11.374124,  1] ../../auth/gensec/spnego.c:418(gensec_spnego_create_negTokenInit_step)
  gensec_spnego_create_negTokenInit_step: Failed to setup SPNEGO negTokenInit request
[2020/06/27 13:06:11.374149,  0] ../../source3/smbd/negprot.c:211(negprot_spnego)
  Failed to start SPNEGO handler for negprot OID list!
[2020/06/27 13:06:11.374316,  3] ../../source3/smbd/smb2_server.c:3266(smbd_smb2_request_error_ex)
  smbd_smb2_request_error_ex: smbd_smb2_request_error_ex: idx[1] status[NT_STATUS_NO_MEMORY] || at ../../source3/smbd/smb2_negprot.c:307
[2020/06/27 13:06:11.374367,  3] ../../source3/smbd/negprot.c:771(reply_negprot)
  Selected protocol SMB 2.???
[2020/06/27 13:06:11.377729,  3] ../../source3/smbd/server_exit.c:244(exit_server_common)
  Server exit (NT_STATUS_END_OF_FILE)

Thanks in advance for any help.

Forgot to add: CentOS Linux release 8.2.2004 (Core) Auto-updates; selinux disabled during testing, firewalld disabled during testing — tsnorthj, Jun 27 '20 at 17:21

T M · Answer 1 · 2021-08-25T19:38:57.250

This is the top result on Google for this error, so even though it's a year+ old I'm going to toss on what solved this for me.

Disabling FIPS compliance.

There's probably a way to fix this by configuring SAMBA, to be FIPS-compliant, but for my use case, it was faster to just disable FIPS.

Edited to include the directions I followed

Credit for the how-to: https://www.thegeekdiary.com/how-to-disable-fips-mode-on-centos-rhel-7/

Remove dracut-fips packages.
yum remove dracut-fips*
Take a backup of the FIPS initramfs.
cp -p /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r).backup
Note: Check if the initramfs file has been created or not. Also, you can use another location instead of /boot/ to avoid space issues.
Recreate the initramfs file:
dracut -f
OR
dracut -f -v /boot/initramfs-$(uname -r).img $(uname -r)
Disable fips=1 value from the kernel command-line. Modify the kernel command line of the current kernel in the grub.cfg adding the following option “fips=0” to the GRUB_CMDLINE_LINUX key in the /etc/default/grub file and then rebuild the grub.cfg file:
Example of how GRUB_CMDLINE_LINUX line looks like:

GRUB_CMDLINE_LINUX="crashkernel=auto rd.lvm.lv=vg_os/root rd.lvm.lv=vg_os/swap rhgb quiet fips=0"

Changes to /etc/default/grub require rebuilding the grub.cfg file as follow:
grub2-mkconfig -o /boot/grub2/grub.cfg
Or If you have a UEFI-based run:
grub2-mkconfig -o /boot/efi/EFI/redhat/grub.cfg
Reboot the server for the changes to take effect:
shutdown -r now
Check that FIPS is not in enforcing mode after a reboot /proc/sys/crypto/fips_enabled should be 0.
Example:
cat /proc/sys/crypto/fips_enabled
0

samba server mount points stopped working on CentOS 8 install, error: Failed to start SPNEGO handler for negprot OID list

1 Answers1