Cannot access admin GUI or SSH

NethServer Version: 7.4

I recently installed NethServer and setup a few modules. The issue that cropped up suddenly with no recent changes is that I can’t access web gui or ssh. Get connection timed out on both. Disabling shorewall doesn’t fix the problem. netstat -l shows 980 is listening. All modules installed are functioning. Just can’t access the web gui. /var/log/httpd-admin shows non errors.

If you need more info I will provide it.

You can try to start the services from command line or view their status:

systemctl start httpd-admin sshd
systemctl -l status httpd-admin sshd

It’s possible that some errors were logged in /var/log/messages, they could provide some clues about the cause.

Edit: I’m assuming you have NethServer 7.4

Another option, if services are already started but not responding, is to restart them.

Yes I have already done this. httpd-admin is running. rebooted server and everything.

Do you use virtualization? I had these kind of problems with a Nethserver as KVM on an old machine.

Can you ping something from the Nethserver?

Which modules do you have installed?

rpm -qa nethserver-*

Like for example I have nextcloud installed. It is working just fine. The virtual hosts are working just fine. But httpd-admin and ssh are broken for some reason

Please post the config of the services, maybe there’s an error:

config show httpd-admin

config show sshd

Config is ok.

You may try removing packages not needed, check which packages are uninstalled:

yum autoremove

Restarting services:

systemctl restart sshd httpd-admin
signal-event firewall-adjust

this removed about 100 packages but even with server reboot still same results

OK. so some development. Seems i can access it but not with the FQDN. but can access nextcloud using same FQDN

For example

:// doesn’t work.
://XXX.XXX.XXX.XXX:980 does work
:// does work

This is really weird.

Did you try other browser or other client?

Do you use custom templates in /etc/e-smith/templates-custom/?

yes I tried multiple locations. i haven’t manually modified any of the files in templates-custom

Can you reach the ssh port now via IP address on port 1983?

You may check if DNS is giving correct results more times:


Check if the port is open from a client:

yum install nmap
nmap -p 980 X.X.X.X
nmap -p 980

Some more config to post, maybe there’s an error, mask public IPs, MACs, secrets…:

config show SystemName
config show DomainName
config show dns
ip addr
db networks show
db vhosts show

1 Like

nmap shows port open

ip addr=shows normal lo and eth0
db networks show=eth0(role=green) and ppp0(not sure where this came from)
vhosts=shows only default

Honestly, I am out of ideas:

  • port 980 is open
  • httpd-admin reachable via IP but not via hostname
  • httpd reachable via IP and hostname
  • no errors in /var/log/httpd…

Did you set a static IP for green network?
Do you have all updates installed?
Does SSH work again?

Last instance would be a backup and reinstall via disaster recovery but I’d wait for some more ideas:

I just realized why. Sorry for so much work over something so stupid. that FQDN is routed through cloudflare and I accidentally proxied that FQDN instead of bypassing CF

1 Like

No problem, glad you got it working!