[RESOLVED] "FAILED to start GSSAPI Proxy Daemon"

[simon_lefisch]

Hey everyone,

During boot I see the error message stated in the title. The boot log shows

Code: Select all

root@server:~$ cat /var/log/boot.log-20200408|grep gss*
         Starting Resets System Activity Logs...
See 'systemctl status gssproxy.service' for details.
[  OK  ] Started Resets System Activity Logs.

In doing so, I get the following:

Code: Select all

root@server:~$ systemctl status gssproxy.service
● gssproxy.service - GSSAPI Proxy Daemon
   Loaded: loaded (/usr/lib/systemd/system/gssproxy.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Sat 2020-04-11 16:20:33 PDT; 3 days ago
  Process: 15593 ExecStart=/usr/sbin/gssproxy -D (code=exited, status=1/FAILURE)

Apr 11 16:20:33 dmserver systemd[1]: Starting GSSAPI Proxy Daemon...
Apr 11 16:20:33 dmserver gssproxy[15593]: gssproxy[15594]: Failed to create Unix Socket! (98:Address already in use)
Apr 11 16:20:33 dmserver systemd[1]: gssproxy.service: control process exited, code=exited status=1
Apr 11 16:20:33 dmserver systemd[1]: Failed to start GSSAPI Proxy Daemon.
Apr 11 16:20:33 dmserver systemd[1]: Unit gssproxy.service entered failed state.
Apr 11 16:20:33 dmserver systemd[1]: gssproxy.service failed.

Now I have perused the web for this issue and found a few different things but nothing tells me what I can do to correct it. Can anyone shine a little light on this?

[simon_lefisch]

bueller.jpg (139.19 KiB) Viewed 5211 times

[pjsr2]

First, check that gssproxy really is not running:

Code: Select all

ps -e -f | grep gssproxy

Failed to create Unix Socket! (98:Address already in use)

The error message indicates problems with creation of the socket /var/lib/gssproxy/default.sock
This may be because the file already exists or has a permission problems or incorrect SELinux attributes, you get an error like the one you see in your log file.

On my system I have:

Code: Select all

# ls -ld /var/lib/gssproxy
drwxr-xr-x. 4 root root 55 Mar 29 16:46 /var/lib/gssproxy
# ls -lZd /var/lib/gssproxy
drwxr-xr-x. root root system_u:object_r:gssproxy_var_lib_t:s0 /var/lib/gssproxy
# ls -l /var/lib/gssproxy/default.sock
srw-rw-rw-. 1 root root 0 Mar 29 16:46 /var/lib/gssproxy/default.sock
# ls -lZ /var/lib/gssproxy/default.sock
srw-rw-rw-. root root system_u:object_r:gssproxy_var_lib_t:s0 /var/lib/gssproxy/                                              default.sock

Normally the date of these files is the boot time of your system as the files are created on boot when gssproxy is started.

Also check the pid and sock files in /var/run. These are created in the next step by gssproxy and if you have issues with
/var/lib/gssproxy/default.sock, you may also have issues with those.

Code: Select all

# ls -l /var/run/gssproxy.*
-rw-------. 1 root root 5 Mar 29 16:46 /var/run/gssproxy.pid
srw-rw-rw-. 1 root root 0 Mar 29 16:46 /var/run/gssproxy.sock
# ls -lZ /var/run/gssproxy.*
-rw-------. root root system_u:object_r:gssproxy_var_run_t:s0 /var/run/gssproxy.pid
srw-rw-rw-. root root system_u:object_r:gssproxy_var_run_t:s0 /var/run/gssproxy.sock

When your gssproxy has failed, try:

Code: Select all

sudo rm /var/lib/gssproxy/default.sock /var/run/gssproxy.pid /var/run/gssproxy.sock
sudo systemctl restart gssproxy

If your SELinux attributes are wrong, restore them:

Code: Select all

sudo restorecon -R /var/lib/gssproxy
sudo restorecon -R /var/run/gss*

Besides "systemctl status gssproxy", look at the output of

Code: Select all

journalctl -xe

That often shows more details of the error than systemctl status.

[simon_lefisch]

Thanks for the reply @pjsr2, and sorry for the late response.....life lol. I ran the commands and here is the output:

First, check that gssproxy really is not running:

Code: Select all

ps -e -f | grep gssproxy

Code: Select all

root@server:~$ ps -e -f | grep gssproxy
root     27806 27465  0 08:58 pts/3    00:00:00 grep --color=auto gssproxy

Failed to create Unix Socket! (98:Address already in use)

The error message indicates problems with creation of the socket /var/lib/gssproxy/default.sock
This may be because the file already exists or has a permission problems or incorrect SELinux attributes, you get an error like the one you see in your log file.

On my system I have:

Code: Select all

# ls -ld /var/lib/gssproxy
drwxr-xr-x. 4 root root 55 Mar 29 16:46 /var/lib/gssproxy
# ls -lZd /var/lib/gssproxy
drwxr-xr-x. root root system_u:object_r:gssproxy_var_lib_t:s0 /var/lib/gssproxy
# ls -l /var/lib/gssproxy/default.sock
srw-rw-rw-. 1 root root 0 Mar 29 16:46 /var/lib/gssproxy/default.sock
# ls -lZ /var/lib/gssproxy/default.sock
srw-rw-rw-. root root system_u:object_r:gssproxy_var_lib_t:s0 /var/lib/gssproxy/                                              default.sock

Normally the date of these files is the boot time of your system as the files are created on boot when gssproxy is started.

Code: Select all

root@server:~$ ls -ld /var/lib/gssproxy
drwxr-xr-x. 4 root root 4096 Mar 31 19:45 /var/lib/gssproxy
root@server:~$ ls -lZd /var/lib/gssproxy
drwxr-xr-x. root root system_u:object_r:gssproxy_var_lib_t:s0 /var/lib/gssproxy
root@server:~$ ls -l /var/lib/gssproxy/default.sock
srw-rw-rw-. 1 root root 0 Jan  8 18:20 /var/lib/gssproxy/default.sock
root@server:~$ ls -lZ /var/lib/gssproxy/default.sock
srw-rw-rw-. root root system_u:object_r:unlabeled_t:s0 /var/lib/gssproxy/default.sock

Also check the pid and sock files in /var/run. These are created in the next step by gssproxy and if you have issues with
/var/lib/gssproxy/default.sock, you may also have issues with those.

Code: Select all

# ls -l /var/run/gssproxy.*
-rw-------. 1 root root 5 Mar 29 16:46 /var/run/gssproxy.pid
srw-rw-rw-. 1 root root 0 Mar 29 16:46 /var/run/gssproxy.sock
# ls -lZ /var/run/gssproxy.*
-rw-------. root root system_u:object_r:gssproxy_var_run_t:s0 /var/run/gssproxy.pid
srw-rw-rw-. root root system_u:object_r:gssproxy_var_run_t:s0 /var/run/gssproxy.sock

Code: Select all

root@server:~$ ls -l /var/run/gssproxy.*
ls: cannot access /var/run/gssproxy.*: No such file or directory
root@server:~$ ls -lZ /var/run/gssproxy.*
ls: cannot access /var/run/gssproxy.*: No such file or directory

It looks like there is an issue when running ls -lZ /var/lib/gssproxy/default.sock as the output is
srw-rw-rw-. root root system_u:object_r:unlabeled_t:s0 /var/lib/gssproxy/default.sock

So I guess the question now is, how do I fix that?

[pjsr2]

Did you try to restore the selinux context of the directory and files in /var/lib/gssproxy?

Code: Select all

sudo restorecon -R /var/lib/gssproxy

Does that correct the selinux context of /var/lib/gssproxy/default.sock ?

If not, what is the output of

Code: Select all

semanage -l | grep gssproxy_var_lib_t

On my system the output is:

Code: Select all

[root@mycomputer]# semanage fcontext -l | grep gssproxy_var_lib_t
/var/lib/gssproxy(/.*)?                            all files          system_u:object_r:gssproxy_var_lib_t:s0
[root@mycomputer]#

[simon_lefisch]

Did you try to restore the selinux context of the directory and files in /var/lib/gssproxy?

Code: Select all

sudo restorecon -R /var/lib/gssproxy

Does that correct the selinux context of /var/lib/gssproxy/default.sock ?

@pjsr2 sorry for the late response again. I check the status of GSSPROXY and it was in a failed state. Ran the code above, checked status and it was active. Restarted the server to verify it loaded during boot and it did. Thank you for your help with this issue, it is much appreciated. Marking as resolved