Linux server Error code 17

UPDATE:
Solved. I used the command wrong:

ampinstmgr CreateInstance MyMinecraftServer Minecraft
Should have been:
ampinstmgr CreateInstance Minecraft

After this you get a prompt, and you can name the server.


OS Name/Version: AlmaLinux with Apache instead of Nginx

Product Name/Version: AMP version 2.5.0.0 (Callisto), built 22/03/2024

Problem Description:

[Info] Testing internet connection...
PING google.com(ams16s22-in-x0e.1e100.net (2a00:1450:400e:803::200e)) from 2a01:7c8:aac4:b7:5054:ff:fea7:8c11 : 56 data bytes
64 bytes from ams16s22-in-x0e.1e100.net (2a00:1450:400e:803::200e): icmp_seq=1 ttl=119 time=1.07 ms

--- google.com ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 1.073/1.073/1.073/0.000 ms
[Info] Checking licence key...
[Info] Licence Type:  AMP Professional
[Info] Current Usage: 0
[Info] Creating Instance: 'Minecraft'
[Info] Performing Step: Creating Environment
[Activity] Audit: Unknown/None [CreateInstance] Created instance Minecraft using MyMinecraftServer module on port 8080.
[Info] Performing Step: Downloading AMP
[Info] Performing Step: Unpacking
[Info] Performing Step: Configuring AMP
[14:42:00] [Core Info]            : Starting AMP version 2.5.0.0 (Callisto), built 22/03/2024 17:32
[14:42:00] [Core Info]            : Stream: Mainline / Release - built by CUBECODERS/buildbot on CCL-DEV
[14:42:00] [Core Info]            : Loading configuration from provisionargs.kvp...
[14:42:00] [Logger Info]          : Purging old audit logs (this may take a moment)...
[14:42:00] [Core Info]            : Running in a QEMU KVM environment.
[14:42:01] [Core Warning]         : Current time zone is set to CET - It is recommended you switch to UTC+0 to avoid time-zone related issues
[14:42:01] [Core Info]            : OS: Linux / x86_64
[14:42:01] [Core Info]            : CPU: Intel Xeon E312xx (Sandy Bridge, IBRS update) (2C/2T)
[14:42:01] [Core Info]            : RAM: 3653MB
[14:42:01] [Core Info]            : AMP Instance ID: 7dfc51fe-8def-46e8-9439-b50277a31a04
[14:42:01] [Core Warning]         : Running as non-standard user.
[14:42:01] [Core Info]            : Performing first-time setup for this instance, please wait...
[14:42:01] [Loader Error]         : Missing plugin/assembly: MyMinecraftServerModule (/home/sjoerdcraft/.ampdata/instances/Minecraft/Plugins/MyMinecraftServerModule/MyMinecraftServerModule.dll) for MyMinecraftServerModule
[14:42:01] [Loader Error]         : Unable to load module MyMinecraftServerModule: Could not load the file 'MyMinecraftServerModule'.
[14:42:01] [Core Error]           : FileNotFoundException
[14:42:01] [Core Error]           : [0] (FileNotFoundException) : Could not load the file 'MyMinecraftServerModule'.
[14:42:01] [Core Error]           :   at AppDomain.Load (String assemblyString)
  at (wrapper remoting-invoke-with-check) AppDomain.Load(string)
  at GSMyAdmin.ModuleLoader.GetPluginAssembly (String PluginName)
  at GSMyAdmin.ModuleLoader.LoadApplicationModule (String ModuleName)
[14:42:01] [Core Fatal]           : Failed to initialize application module.
[Error] Configuration Failed: AMP configuration failed with return code 17
[Info] Instance creation failed. AMP configuration failed with return code 17

Steps to reproduce:

  • Install Amp as root user
  • Manually use apache, instead of nginx
  • Ignore the error about not being able to set up Nginx
  • Use the command: ampinstmgr CreateInstance MyMinecraftServer Minecraft
  • IP binding: 0.0.0.0
  • Port 8080
  • Pro licence key

Actions taken to resolve so far: Well, the problem does not seem related to apache / nginx.

Why are you using CreateInstance manually instead of letting the install script setup AMP? You’ve got the arguments wrong anyway but there’s no reason to do that.

The install script sets up amp, but stops after it discovers I am not using nginx. I had to improvise.
I had the arguments wrong, hence my update at the top.
Modifying the install script so it can handle both nginx and apache would be convenient. Or at least have the option to skip the whole redirect and use domain:8080

Just say “No” when it asks for HTTPS then - it only sets up nginx if you say Yes. Apache isn’t suitable for AMP because it has relatively poor websockets support and isn’t as good at being a reverse proxy. But this doesn’t matter since if you select No it doesn’t use any external webserver at all and only uses the built in one.

Reading the output would have helped you here because it does say:

Apache2 is installed, which will conflict with nginx as required for AMPs reverse proxy - Aborting.
Either remove Apache2 and try again, or re-run this script and select ‘No’ when asked if you wish to use HTTPS

Thank you for your fast replies. That works aswell.

However, from a customer-service point of view, it would be convenient to cover apache somehow, without having to explain this to every customer who uses apache. I came across multiple discussions about apache, and the cubecoders reply can be summarized as (correct me if I’m wrong) : “you shouldn’t use apache, it sucks, now it is your problem. just remove apache from your server because I am sure AMP is the only thing useful on your server anyways”.

As a software engineer / PO / publisher I came across this and my advice is to include the option of apache somehow in the otherwise really good install script. Apache is used roughly as much as Nginx is, and I’m sure a lot of your customers install AMP on servers that are already set up for other purposes.

There’s nothing stopping you from running Apache on the same system as AMP. Literally the only thing you lose is the automated HTTPS setup. But that aside everything functions as normal.

Nginx is only being used as a reverse proxy, it’s not hosting AMP at all.

This is very clear to me now, thank you.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.