you are FUCKING KIDDING ME.
Guy on IRC straight-up told me "just buy additional storage".
As if we aren't poor
As if we aren't about to be _homeless_ in a couple months
IN THIS ECONOMY
I...I can't even.
FUCKING BTRFS
Now it's got 60GB "unreachable" space. I tried defragging to fix that and just made it WORSE.
Hey #btrfs people: How safe is CRC32C as a checksum?
Should I be concerned that all the filesystems we've btrfs-convert'd so far defaulted to crc32c?
Should I tell it to use SHA256 for other converts in the future?
It looks like we can't change it on an already converted filesystem...
Hi friends who use #BTRFS on #Linux, with an operating #trash solution!
What is your setup? I am trying to use a btrfs subvolume as my .local/share/Trash folder so it doesn't get the snapshots and takes up space after being emptied. However, this confuses both #trash-cli (which now creates the .Trash-1000 folder in my home), and Nemo (which fails to trash the file).
Not to mention, that Nemo also fails to use the trash on a BTRFS drive with properly set up trash directory.
If you have a potentially (probably) dying SSD with BTRFS on it, would you rather:
1. Try to copy out data first
2. Try to use BTRFS repair utilities first, and then try to copy data
For reference this is from a used Thinkpad X390 which lately has been running extremely slowly and now not being able to boot properly (not reaching login screen). btrfs check found some errors (but FS was in use), UEFI diagnostic found no errors with the disk nor other hardware.
I’m using this sd-card image generator for a RiscV board, similar to others like RaspberryPi etc. NixOS on RISCV/VisionFive 2 - Official NixOS Wiki I changed the root filesystem from ext4 to Btrfs using this: sdImage = { compressImage = false; rootFilesystemCreator = "${pkgs.path}/nixos/lib/make-btrfs-fs.nix"; # do this manually as replacing the resize2fs string in postBootCommands # might be too complex. btrfs uses its own resize command. expandOnBoot = false; }; nixpkgs.overlays =...
Pupdate: I hopped on IRC and they're being extremely helpful.
Apparently it's nigh impossible to recover from out-of-space during a metadata rebalance, what the fuck... they're having us make a concatenated device out of two underlying devices using device-mapper (that's a thing???), offline grow the filesystem to cover both, cancel the balance and free up some space and shrink, then do away with the concatenated device.
And according to the manpages, the offline grow wouldn't work if we had RAID set up.
Just, wow. And they say "btrfs is stable now"...
Ylfingr, Devourer of Suns
NiceMicro
dei Muddr
met
Project Insanity
AskUbuntu
David -> 🇵🇹