Why wouldn’t I use ext4, if I don’t need copy-on-write? There is nothing wrong with ext4, it works fine and doesn’t corrupt easily. It requires no special setup and no maintenance. It’s an excellent filesystem…
Sure, it doesn’t have the fancy features of Btrfs or ZFS, but if I have no need for such features, what benefit would I have from using these filesystems over ext4?
Yeah i wonder why…coz according to my own experience, BTRFS stands for:
unrecoverably-corrupt-all-the-files-on-simple-electricity-failure-spike-fs
Seems very progressive to me, especially considering all of my ext4 disks were completely fine. 
I approve of your method of composing acronyms. 
Reminds me of INTERCAL.
Isn’t F2FS better for SSDs? ( I feel like I asked this some time ago)
I would be curious when did it happen in your real experience? What version of Btrfs?
Let me think…It was around last months of 2021 i believe, maybe little earlier.
But since i constantly reading about bugs and horrible failures of BTRFS it shouldn’t matter that much, at least YET 
It would have been more a reliable poll if the poll allowed multiple choice, to reflect your situation of multiple FS.
BTRFS will become the dominant desktop/workstation file system over they coming years.
Itsfoss points out the advantages well.
I got that from https://itsfoss.com/btrfs-default-fedora/
BTRFS has been the default file system since Fedora 33 in 2020, so I think the trend will continue as new users coming from Windows don’t want to confront these geeky Linux file management issues. BTRFS will manage all that file stuff for them. Just my opinion.
AFAIK Linux kernel 5.15 or newer enables Btrfs profile of metadata “DUP” by default, that means you get two same metadata. If one of both metadata is broken or corrupted but last one survives so you might be able to mount the partition and recover your data back.
Who installed Linux Kernel older than 5.15 initially, then he should manually convert “Single” metadata to “DUP” Metadata by btrfs balance
What?
Which Linux file management issues?
One of the core reasons to leave Windoze is to loose it’s absolutely horrible, unstable, constantly fragmented ntfs.
ext4 is something that just works without any problems and works good, unless you really need some funky features of btrfs / zfs and other ones of course…And any of those are still better than Windoze crap
Both of those points are absolute 
, in my view…if they’re not compared it with NTFS of course
Especially hardware lifespan, because the only things that affect lifespan are:
SSD: amount of data written (BTRFS should technically use more write operations by it’s COW nature as far as i understand, especially considering some of it’s bugs during last couple of years which could kill some SSDs if they wouldn’t be fixed fast enough)
HDD: number of start / stop of spin motor (btw that’s why most forms of HDD powersaving involving stopping motors SUCKS, and should be absolutely turned off) and general lifespan expectations for given model…so nothing FS is actually involved in.
What are these geeky Linux file management issues exactly that will be managed by BTRFS for me?
Take it up with the source. I simply quoted him
A huge welcome to @dreadnought aboard the good ship 
@
Windoze users coming across may find our way of doing things a bit geeky 
First I heard of this. Do you know where you saw this?
See the changelog of Btrfs, there are some bufixes.
https://btrfs.readthedocs.io/en/latest/CHANGES.html
How to convert single to DUP
That is nice, I thought DUP was only for rotational device, I’m running a non-rotation device (SSD) and DUP is set:
Data,single: Size:15.01GiB, Used:13.32GiB (88.74%)
/dev/sda2 15.01GiB
Metadata,DUP: Size:1.00GiB, Used:464.58MiB (45.37%)
/dev/sda2 2.00GiB
System,DUP: Size:8.00MiB, Used:16.00KiB (0.20%)
/dev/sda2 16.00MiB
The big thing that caught me is snapshots…and yes I know other file systems can sort of do snapshots too, but BTRFS was designed for it. Being able to roll back to a snapshot is a huge timesaver I’ve used many times in my vSphere environments, and being able to do it without some wacky multi-partition Acronis setup on my laptop is fantastic.
I would always re-install my Linux setups if I had major problems until I put it on my primary laptop. Balancing needing to use my laptop vs an annoying problem usually pushed me in the direction of troubleshooting instead. Most of the times I did a re-install was back when I was still using Ubuntu and it’s derivatives. Mainly caused by trying to get newer software that wasn’t in the official repos yet. I haven’t really had those types of issues since moving to Manjaro and now EOS. Of course my experience with Linux also leveled up in between as well.
I am not sure, but someone cannot mount / import ZFS mirror (RAID 1 in TrueNAS) with two SSDs after a some corrupted upgrade. but he cannot recover it.
He wrote:
No, unfortunately not. On https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-5E/ it states pretty clearly that this cant be fixed. So I gave up trying and sucked up my dataloss.
https://www.reddit.com/r/zfs/comments/wd3j5t/zfs_import_invalid_label/
Come on. A person having a problem or even a problem existing has no bearing on one thing being more stable than another thing. There isn’t even any explanation in that topic about what the actual issue was or how the pool was configured.
Do you how many posts on btrfs filesystems being unrecoverable there are?
I know you are a fan of btrfs but to argue it is more reliable than zfs is just plain silly. I have used both filesystems extensively over a period of many years.
I personally have lost multiple btrfs volumes over the years. The problem with btrfs is that it is too sensitive to data corruption when that corruption occurs in the metadata. There are too many situations that result in total data loss. They tried to solve this by putting an extra copy of the metadata in place by default but failures still happen.
With zfs, it is much more likely that your data will be recoverable even if your pool becomes corrupt.
That being said, any filesystem is capable of becoming totally corrupt.
I am not trying to say that btrfs is bad. It should be pretty clear if you read my posts that I am an advocate of btrfs.