Backups fail - fatal: confused by unstable object source data for example

[McMelonTV]

Hey, my backups keep failing with the error below. It worked before players started actually playing survival. After that this keeps popping up everytime a backup runs.

[21:35:19] [pool-5-thread-5/ERROR]: [STDERR] fatal: confused by unstable object source data for 3302de592c5eaec86662966e7564b207d2461348
[21:35:19] [pool-5-thread-5/ERROR]: [STDOUT] add 'entities/r.2.0.mca'
[21:35:19] [pool-5-thread-5/ERROR]: Exit 128 when executing: git -C /home/container/world add -v .
net.pcal.fastback.utils.ProcessException: Exit 128 when executing: git -C /home/container/world add -v .
        at net.pcal.fastback.utils.ProcessUtils.doExec(ProcessUtils.java:75) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.utils.ProcessUtils.doExec(ProcessUtils.java:44) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.repo.CommitUtils.native_commit(CommitUtils.java:133) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.repo.CommitUtils.doCommitSnapshot(CommitUtils.java:82) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.repo.RepoImpl.doCommitSnapshot(RepoImpl.java:121) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.commands.SchedulableAction$2.lambda$getTask$0(SchedulableAction.java:51) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at net.pcal.fastback.mod.AutosaveListener.lambda$run$0(AutosaveListener.java:72) ~[fastback-0.17.0+1.20.4-fabric.jar:?]
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539) ~[?:?]
        at java.util.concurrent.FutureTask.run(FutureTask.java:264) ~[?:?]
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136) ~[?:?]
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635) ~[?:?]
        at java.lang.Thread.run(Thread.java:840) ~[?:?]

FastBack Info

[21:40:30] [Server thread/INFO]: FastBack version: 0.17.0+1.20.4-fabric
[21:40:30] [Server thread/INFO]: native git installed: true
[21:40:31] [Server thread/INFO]: git: git version 2.34.1
[21:40:31] [Server thread/INFO]: git-lfs: git-lfs/3.4.1 (GitHub; linux amd64; go 1.21.1)
[21:40:31] [Server thread/INFO]: World size: 24 GB
[21:40:31] [Server thread/INFO]: Local backup size: 24 GB

[pcal43]

Tough to say. That error is coming deep from within the bowels of git. I've never seen it before.

A quick google search suggests one of the following:

• You're getting unlucky with timing and world files are getting written at just the wrong time. I try to lock minecraft from flushing updates while backup is happening but maybe there's gap.
• A really big file is getting added (1gb+ say)
• Some sort of disk corruption

I'll leave this open but I'm unlikely to have time to investigate it myself.

[RaeTheGit]

I seem to be running into the same issue. The server console reports:

Creating backup snapshot 2024-03-21_10-06-05

pool-20-thread-4/ERROR

Exit 128 when executing: git -C /home/amp/.ampdata/instances/WeirdMods201/Minecraft commit -m u2YH/2024-03-21_10-06-05

Console

net.pcal.fastback.utils.ProcessException: Exit 128 when executing: git -C /home/amp/.ampdata/instances/WeirdMods201/Minecraft commit -m u2YH/2024-03-21_10-06-05
at net.pcal.fastback.utils.ProcessUtils.doExec(ProcessUtils.java:75) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.utils.ProcessUtils.doExec(ProcessUtils.java:44) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.repo.CommitUtils.native_commit(CommitUtils.java:140) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.repo.CommitUtils.doCommitSnapshot(CommitUtils.java:82) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.repo.RepoImpl.doCommitSnapshot(RepoImpl.java:121) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.commands.LocalCommand.lambda$run$1(LocalCommand.java:67) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at net.pcal.fastback.commands.Commands.lambda$gitOp$0(Commands.java:125) ~[fastback-0.15.6+1.20.1-forge.jar%23407!/:?] {re:classloading}
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539) ~[?:?] {}
at java.util.concurrent.FutureTask.run(FutureTask.java:264) ~[?:?] {}
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136) ~[?:?] {}
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635) ~[?:?] {}
at java.lang.Thread.run(Thread.java:833) ~[?:?] {re:mixin}

pool-20-thread-4/INFO

fastback.chat.commit-failed
Backup complete. Time elapsed: 1s

Is there any detail I can provide to help get to the bottom of this?
Just in case, this is MC 1.20.1 with NeoForge as modloader.

[McMelonTV]

It seemed to have been caused by a logging mod for me. Are you running anything like that by a chance?