Tarkov Fika Mod instance failing to start

Technical Support
,
1

OS Name/Version: Ubuntu 24.04.3

Product Name/Version: AMP 2.6.2.8 - 20250909.1

Problem Description:
The Tarkov Fika Mod instance fails to start, throwing an error.

This task could not be completed: Running Update Tasks - - State: 3.

/bin/bash failed with exit code 2

Looking through the logs, I can see the following:

Cloning into 'srv'...
Note: switching to 'c6a4e5ef8d77e5195dc2bc6449b9e69b19605474'.
You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.
If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:
  git switch -c <new-branch-name>
Or undo this operation with:
  git switch -
Turn off this advice by setting config variable advice.detachedHead to false
batch response: Expected JSON type, got: "text/plain"
batch response: Expected JSON type, got: "text/plain"
batch response: Expected JSON type, got: "text/plain"
batch response: Expected JSON type, got: "text/plain"
batch response: Expected JSON type, got: "text/plain"
batch response: Expected JSON type, got: "text/plain"
Failed to fetch some objects from 'https://lfs.sp-tarkov.com/sp-tarkov/server'

I can confirm that Git LFS is installed on the machine and is initialised using git lfs install.

Steps to reproduce:

  • Create Tarkov Fika Mod instance
  • Launch the Tarkov Fika Mod instance with default settings
  • Failed to start after reaching step 3

Actions taken to resolve so far:

  • Installed Git LFS and made sure it’s operational

  • Attempted to clone the repo from https://lfs.sp-tarkov.com/sp-tarkov/server which gave me an error saying it’s empty:

    git clone https://lfs.sp-tarkov.com/sp-tarkov/server
Cloning into ‘server’…
warning: You appear to have cloned an empty repository.
2

That’s on the SPT devs. Their server is flaky at best

3

Is there a way to bypass this, such as downloading the server files manually and placing them in the instances directory?
Or is this a deeper rooted issue?

4

It’s a deeper issue. The only way to get the server on Linux is to build it from source, and you need all the files to do that.

5

:rofl:

No. Don’t put this on us.

Trinagan, that’s the incorrect LFS server that you’re using, so you obviously aren’t on the latest commit, or your workflow is modifying it manually, either way that needs to updated to match the latest on develop.

Also, you’ve tried to directly clone the LFS server… What are you doing?

This is user error.

6

You know, I had a suspicion this could be the case.

I don’t have control of which server the instance tries to pull the files from; the setup is triggered automatically upon starting the instance, and it all seems to be done inside Docker.

I have just gone back to try again now, and it has now managed to successfully download the files and launch…

So I guess that got resolved either way?

As for trying to clone the LFS server.. I was trying to work with what the logs were giving me :man_shrugging:

7

AMP clones the repo and runs git-lfs pull, as per the GH instructions. So it was a repo issue.

As to what the user tried to do manually to get around it, nothing to do with AMP.

Anyway, seems to be solved.

8

Hello! Anyway you could update the git-lfs download links to the new domain for older versions? Or point me in the direction of being able to update my own instance. Much Appreciated.

(Figured I’d slap this on an already similar support request)

9

You will have to make that request on the SPT GitHub. Given they expressly don’t support anything other than the latest version, I suspect they won’t backport older LFS files