So, maybe I missed it, maybe I’m stupid, but I’m looking for something.
We are using webmin on our proxy, very nice. BUT we have 12 ubuntu server and I was wondering, logging everyone to check the status, what if we could have a webpage with everything on the same location.
What I’m thinking is : the dashboard > system information x12.
Each one connecting to every server.
Instantly able to check uptime, update, disk space for everyone easily.
Again, maybe I’m just dumb, but it could be nice
You are using Webmin right? not Virtualmin - yet you imply 12 “servers”
if, only if, then I would concur that being able to monitor multiple servers would b nice bit I do not think it is possible .
I am using webmin. When opening https://webmin.com/ I select forum and was send here.
I said 12 servers, it’s 12 Ubuntu servers, but only 1 has webmin right now.
I CAN install webmin on each servers, but it’s useless if need to open each page to check them. It’s faster to just ssh.
sorry for my confusion here but 12 Ubuntu servers implies to me 12 different host connections ie. one IP per server. I might be ignorant here but I am pretty sure Webmin does not manage groups of servers
I’m ignorant too, but I was wondering if anyone has done like a page with 12 different connection. If it’s not possible too bad, but I like the interface of this dashboard.
I do this using system monitors adding each required service to one webmin server however you need to ensure each physical server is running webmin in order that the server that needs to display the data can read the data from the other servers. Note well there was a smalll bug using debain/ubuntu which stopped this working but I’m sure our bug fixer shoulders will give you the load down on the bug and on how to fix it
This gets pretty close to the line in terms of Webmin is a server administration tool that offers up system stats on the node it is installed on. At 12 servers you probably really need monitoring software. It isn’t just if the node is up, it becomes about if the services on those nodes are running.
I think it was mentioned above that you can use the Webmin Server’s index to add servers and then use System Monitors to see what’s up on each server. There was some discussion in the past about making the Systems Monitor more ‘configurable/useful’. Might have been @jimr1 that made a thread or two about this?
I generally use Prometheus (and Loki for logs, and Alloy as the agent) for this kind of thing. Maybe a big hammer, depending on how much data you really want, but if you’ve got 12 servers, you probably need more better history and query abilities than the basic graphs that Webmin provides.
As server admin you can check all you 12 server from one interface. sudo apt install cockpit
now login https://SERVER-IP:9090 Login with your root user, find ‘add host’ from menu. Add all you 11 server’s ip & login info & add them (you have to install cockpit in all your servers).
Cockpit communicates between your servers. Now you can followup your 12 server from one instance.
Here’s a subset of tasks you can perform on each host running Cockpit:
Inspect and change network settings
Configure a firewall
Manage storage (including RAID and LUKS partitions)
Create and manage virtual machines
Download and run containers
Browse and search system logs
Inspect a system’s hardware
Upgrade software
Keep tabs on performance
Manage user accounts
Inspect and interact with systemd-based services
Use a terminal on a remote server in your local web browser
I tried the webmin server index…but it look so weird. I add a server, then I could see the interface of second server…but couldn’t go back. Add 3 more, can select them but still see server ‘2’.
I tried to install cockpit…but it was looking “too” simple.
IT
IS
PERFECT.
Simple, easy.
I had one server who refuse to connect, had to remove the old key in the server “1” to connect with SSH and it’s now ok.
Only trouble I have now : with 2 of them I have an error :
sender does not match (:1.3038 vs :1.3034)Cannot refresh cache whilst offline
It happened when I tried to see updates.
Internet suggest : systemctl disable network-manager.service
Don’t want to fucked up a prod server…Can I really disable this service?
Edit : I tried to on the less usefull server : unit file network-manager-service does not exist.
But thank you again for the answer tawfiq.
NetworkManager is a system network service that manages network devices and connections. It manages Ethernet, WiFi, mobile broadband (WWAN), and PPPoE devices. systemctl disable network-manager.service will not work.
Possibly your system uses networkd-dispatcher.service instead.
Playing in production server is not a good idea. Check twice what you are doing.
I test with a server used only for GLPI, in the same building I am.
Solved with :
nmcli d
systemctl stop NetworkManager.service
systemctl disable NetworkManager.service
Great
Now only missing option for give a surname to server with ugly hostname and change order in the list…but it’s decoration.
Weird point : some server can connect as admin only with “Turn on administrative access”, other need the password.
Very happy with this new tool, I think I will use it for a long time, thank you very much tawfiq.