New install, no GUI

Fresh install of current stable iso [Rockstor-Leap15.6-generic.x86_64-5.0.15-0.install.iso] on somewhat old VMware (6.5)

I ran the install but I don’t have a dhcp server. During the install there were several bits of text indicating a problem but they flashed by too fast to read. This was right when I was prompted to set the password, this operation was not interrupted.

After a reboot there is a bit of text indicating something didn’t happen due to a missing dependency but it scrolled off the screen so I can’t read it.

I set a static ip address using nmcli but the nginx service is not running so no gui.

[systemctl status nginx] gives:

× nginx.service - The nginx HTTP and reverse proxy server - 30-rockstor-nginx-override.conf
Loaded: loaded (/usr/lib/systemd/system/nginx.service; disabled; vendor preset: disabled)
Drop-In: /etc/systemd/system/nginx.service.d
└─30-rockstor-nginx-override.conf
Active: failed (Result: exit-code) since Tue 2025-02-04 14:18:09 GMT; 51s ago
Process: 14483 ExecStartPre=/usr/sbin/nginx -t (code=exited, status=0/SUCCESS)
Process: 14484 ExecStartPre=/usr/sbin/nginx -t -c /opt/rockstor/etc/nginx/nginx.conf (code=exited, status=1/FAILURE)

Feb 04 14:18:08 “server name” systemd[1]: Starting The nginx HTTP and reverse proxy server - 30-rockstor-nginx-override.conf…
Feb 04 14:18:09 “server name” nginx[14483]: nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
Feb 04 14:18:09 “server name” nginx[14483]: nginx: configuration file /etc/nginx/nginx.conf test is successful
Feb 04 14:18:09 “server name” nginx[14484]: nginx: [emerg] cannot load certificate “/opt/rockstor/certs/rockstor.cert”: BIO_new_file() failed (SSL: error:02001002:system library:fopen:No such file or directory:fopen(‘/opt/rockstor/certs/rockstor.cert’,‘r’) error:2006D080:BIO routines:BIO_new_file:no such file)
Feb 04 14:18:09 [14484]: nginx: configuration file /opt/rockstor/etc/nginx/nginx.conf test failed
Feb 04 14:18:09 “server name” systemd[1]: nginx.service: Control process exited, code=exited, status=1/FAILURE
Feb 04 14:18:09 “server name” systemd[1]: nginx.service: Failed with result ‘exit-code’.
Feb 04 14:18:09 “server name” systemd[1]: Failed to start The nginx HTTP and reverse proxy server - 30-rockstor-nginx-override.conf.

Detailed step by step instructions to reproduce the problem

per above

Web-UI screenshot

none possible

Error Traceback provided on the Web-UI

none possible

What to do?

I suspect that the Rockstor services failed due to a dependency on network manager. I am saying that since I’ve seen this in the past (and in some other threads, too) on my own VM installation, albeit not VMWare, but Virtualbox.

Network Manager in a VM tends be more likely to fail due to timeout for interfaces to acquire IP addresses etc. I found this to be a problem with some bridged adapter setups (fixed IP or dhcp, too).

To see whether that is the issue you can check first:

systemctl status network ← will check whether Network Manager failed

If that is “dead” you can try to restart it with:
systemctl start network (or systemctl restart network)

Then check the status again.

If it indeed started up, you can manually execute the startup procedure by executing the three commands (each after the previous successfully completed):

systemctl start rockstor-build
systemctl start rockstor-pre
systemctl start rockstor

Once (if) the third command returns without error message, check the status or rockstor again:
systemtcl status rockstor

If that shows as running, give it a few more minutes until it finishes all its startup tasks (including starting nginx, etc.) and then see whether you can get to the WebUI.

If you are observing it from your VMWare terminal you will likely see some messages scroll across the screen mentioning nginx (hopefully not error but success messsages).

If that does work out, you will need to take a look at your network adapter configuration at the VMWare level (simplest for me on VirtualBox was to use a host-only adapter, as it would connect really quickly and was sufficient for my testing purposes).

Somewhere aroun 5.0.9-0 (I think) the requirement for a rockstor startup was changed to be dependent on Network Manager having successfully started, so if, e.g. there is a timeout on obtaining an IP address (since you mentioned you don’t have a dhcp server connected that gives out IP addresses), or Network Manager recognizing a static IP that you might have set/not set. Might also require to use nmcli or ip command line stuff to set a static IP address right up front (so, it’s persistent across reboots) for the successful network manager and Rockstor bootup, which also looks like you did at some point.

Out of curiousity, what does the below command show you?

ip addr show

2 Likes

Hi, many thanks for your reply.

The VM is offline so I can’t check at the moment but, iirc, before I set the ip address, [ip addr show] just showed the interfaces lo and eth0.

Tomorrow I’ll spin up the VM and get the rest of the info. What you say about network manager makes sense.

Thanks,

Eric

2 Likes

There is also a section on having a static IP set in the documentation:

https://rockstor.com/docs/installation/special-install-scenarios-howto.html#assign-static-ip-address-after-installation-but-before-webui-setup

3 Likes

Thanks for this and your other comments!

1 Like

@MRC-MBU Hello again.

@Hooverdan & @MRC-MBU Re:

We also have our top-level systemd servcie:

rockstor-bootstrap

and each depends on it’s predecessor. So one can just start “rockstor-bootstrap” and it should start the cascade appropriately.

We have them documented in the later-half of the following doc section: Contributing to Rockstor - Overview — Rockstor documentation copied in here for convenience:

  • rockstor-build # runs build.sh if .venv dir is not found
  • rockstor-pre # starts .venv/bin/initrock after rockstor-build
  • rockstor # starts .venv/bin/supervisord -c etc/supervisord.conf after rockstor-pre
  • rockstor-bootstrap # starts .venv/bin/bootstrap after rockstor

Hope that helps.

2 Likes