trapexit

Please follow the instructions in the support docs regarding issues like this. At the very least you need to say what version of the software you're using.

Please provide the details asked for in the docs and in the ticket creation page. The docs present fully functional examples. Your comments are too vague to help with.

I need a trace from an app experiencing the issue. Your trace is of df and it looks to exit fine and the query of mergerfs took 0.0096 seconds to...

Using basic tooling is what I suggest using in the docs but a filesystem is a complex beast with many types of functions so getting more than one function testing...

There mergerfs trace shows nothing but a gap in time around then which `df` had issued a `stat` request. Without logs upto the event or external information there isn't much...

Having traces will only show who is being slow. It won't necessarily diagnose anything. There is very little going on in mergerfs that can cause slowdown. It's primarily just a...

I would prefer moving such discussion to the [discussions](https://github.com/trapexit/mergerfs/discussions) page rather than overloading an existing ticket. But to respond: there is no such thing as a timeout in filesystem APIs....

I don't believe writeback caching is what the OP is asking about. I think the OP needs to be a lot more specific about what they are asking though. Do...

The [description](https://github.com/trapexit/mergerfs-tools?tab=readme-ov-file#mergerfsbalance) of the tool is: > Will move files from the most filled drive (percentage wise) to the least filled drive. Will do so till the most and least...

I'm not following. That's what it effectively does. Move file from filesystem with least free space % wise to the most. Repeat till difference between all filesystems is < 2%...