Xibo server running but no Webpage access

Get Help
1

I have a real mystery going here. I have been running Xibo on a Ubuntu OS (LAMP but with a Desktop) for a number of years. I have been running the “Compose” install with no major problems. I upgraded from 1.8.1 over to the 2.x version earlier this year and did not encounter any problems until now.

The symptoms are as follows. The server is running and the CMS is apparently running “docker-compose up -d” with green on the xibo_cms_xmr, xibo_cms_db and xibo_cms_web.

Normally this would mean that if the CMS is running, using “localhost:” on the server would give you the webpage login, and that has always been the test of success. Then one could then login from another PC on the network, using a browser to the CMS address.

However in the last 3 weeks, even though the CMS is running (apparently) the webpage, fails to appear reporting that the server is not available. This manifests itself in Firefox, Chrome, Brave, Internet Explorer.

I thought it was an issue with the Virtual Machine (we host it on VMWARE) and attempted to restore the server to an earlier version prior to an OS update, but this had no effect. I then tried downing the server using “docker-compose down” and bringing it up again. but this didn’t seem to work as well.

I then tried about 30 minutes later and was able to login.

Wierd. There are no firewalls, so that would be a factor. I then created a new Ubuntu host completely with the latest CMS and tested it as well. It come on immediately.

Then again today we spotted that the old server was not accessible over the weekend but was working fine this morning, and we were then very surprised by the new server after restarting was ALSO not available for 15 minutes.

Very mysterious. Is there any way to check on the status of the CMS, or something we can do to identify the issue.

2

When I was struggling with the Xibo setup, after running the Docker -compose down message, I would then delete everything from the file system so it would be created anew when I ran the UP command.

I do not know if deleting everything from the ‘C:\XIBO\xibo-docker\xibo-docker-2.0.3\shared’ folder mattered, but starting fresh seemed like the right thing to do.
Good luck!

3

Mine is not working at all after I upgraded. I cannot figure out why it is not working. But like you everything seems to be working properly, yet it will not display the web page.

4

What is really the mystery is that it ONLY happens when you upgrade or work on the hosting OS (Linux/Ubuntu/Lubuntu in my case and you have to restart the OS. I am not in the habit of restarting the OS, this isn’t Windows, but sometimes if there is a big upgrade or new kernel, then an update is required. When this is done the CMS seems to be running OK after the reboot (it is supposed to start automatically, right?) but the CMS webpage where your login doesn’t appear and the message says that the “connection is reset” and that the website might not be running.

I found the last time that it was not working at 10:00pm at night, but when I got into work the following morning it was working, 10 hours later. I had to do a reboot the following day and it misbehaved again, I went out for a quick break to clear my head and grab a cup of tea, and when I returned 15 minutes later the site was working perfectly. This occurs on an upgraded CMS and a clean install of a CMS.

I wonder if the Xibo gurus have some sort of method that we can use to poke the CMS on the hosting OS and see what is working and what is not.

5

Like a validation script? That would be awesome!

6

I have exactly the same problem

Debian 9, with docker
sometimes it works, and other time not.

when I use the command “top”, there is no problem with CPU or memory

PS : sorry for my english, I am a french guy…

7

Is it possible that either DHCP or a stupid user is bringing up the same IP address as your server on some other machine on the network? When you have this problem again, open up a command prompt window on your workstation, and try to Ping the IP address of your server. If you get responses from more than one address, that’s your problem. Alternatively, you could use software like nmap on a Linux workstation, ‘Angry ip scanner’ (yes, that’s really what it’s called, and I recommend the 2.11 version which doesn’t require Java) on a Windows machine, or Fing on an Android cell phone.

8

The IP checks out. No problem there. I think it is the database! If I open up a browser on the host machine and type in localhost, I get the same error as if I try to login from another workstation browser. This “bug” only happens if you have to reboot the host machine after a software update. If you do that, then you lose the web interface. If I run a root terminal and cd to /opt/xibo and down the xibo instance using “compose down” then the CMS, the WEB and the DB components all come up green and report that they are down. However, when I use “compose up -d” then the CMS and the WEB components show as green on the results but NOT the DB (Database)

I have just had the same issue this morning. The login webpage refused to display after a host reboot. I downed compose and brought it up again and got the same error, BUT after 30 minutes the login page appeared finally!