There's Talk Again About Btrfs For Fedora

[RahulSundaram]

https://www.suse.com/products/server/technical-information/[/URL]

Hint: Look at the SP2 press release and what they commericlally support and recommend. If they dont consider it trustworthy, do you think Oracle and SUSE would support it? Your refusal to answer that basic question tells a lot.

[xeekei]

Chris Mason is no Jeff Bonwick. I doubt many people would want to use a Mason distro. Sun, on the other hand, was trusted by a lot of people for putting out a dependable OS.

That wasn't my point...

[jwilliams]

Hint: Look at the SP2 press release and what they commericlally support and recommend. If they dont consider it trustworthy, do you think Oracle and SUSE would support it? Your refusal to answer that basic question tells a lot.

And you continue to grasp at straws. In fact, they do not consider btrfs trustworthy for widespread usage, which is why it is not the default filesystem. Your inability to grasp this simple concept tells a lot.

The fact is that no major linux distro trusts btrfs enough to make it the default filesystem.

[jwilliams]

That wasn't my point...

Right, because what I wrote actually made sense, so it couldn't be your point.

[Ericg]

You really need to stop grasping at straws. The default filesystem for SUSE is ext3.

RHEL, CentOS, Fedora, SUSE, Ubuntu and Debian. None of the major linux distros trust btrfs enough to use it as the default filesystem. There is a reason -- because btrfs is not trustworthy. It is not ready for widespread, serious usage.

https://www.suse.com/products/server...l-information/

I really wouldnt be trusting SUSE's choice of recommendation purely based off of the "Default" choice...the default is still ext3, not even ext4. By your logic, ext4 is therefore not ready for widespread serious use just because SUSE says "No" to it being the default.

btrfs has 3 major blockers right now for "mainstream" use.

1) mkfs.btrfs doesn't always work (correctly) and I know that from personal experience-- I installed Arch Linux on THIS laptop about a month ago and mkfs.btrfs would run correctly but mount would refuse to mount it, saying the system had a bad superblock along with a few other errors. At first I thought it was the drive, but mkfs.ext4 worked just fine. (Not just one attempt, multiple attempts always the same thing: ext4 is created and mounted successfully, btrfs is created successfully (no errors) but refuses to mount)

2) btrfsck needs to be written around the fsck API's so that a proper fsck.btrfs can be available, this is mainly an issue for initramfs creation (atleast mkinitcpio looks for fsck.$filesystem directly) and scripts that merely call "fsck"

3) the VM issue needs to be sorted out. For those that don't know, btrfs suuuuuuuucks at small random writes, which is exactly what databases and VM images handle and those need to be properly sorted out otherwise you have crap performance from day 1 that gradually gets worse and worse ....fun

--Eric--


EDIT: "btrfs scrub" is only an "fsck" utility if there are copies available to be used. As far as I can find online all "btrfs scrub" does is go through the file system and checks the checksum of every file, if one is damaged it deletes the file and does in-place replace of a saved backup of that file...if theres no backup available, then you just have a corrupted file.

^If im wrong on that, please include a link to the source because thats all my own research turned up.

[jwilliams]

By your logic, ext4 is therefore not ready for widespread serious use just because SUSE says "No" to it being the default.

No, that may be your "logic", but I did not say it. Your "logic" is flawed.

If ext4 offered a large number of additional features over ext3 (which it does not), and none of the major distros used ext4 as the default filesystem, then you would have a reasonable argument. But you do not.

[xeekei]

Right, because what I wrote actually made sense, so it couldn't be your point.

No, I'm telling you that you missed the point. Mason needs to convince the distros, while Sun only needed to convince themselves. It's a lot easier to convince yourself than other people.

IF (you do know about hypothetical scenarios, I hope?) Chris Mason had his own distro that was big and respected, he probably would've used Btrfs on it already. Putting the development speed effectively on par with ZFS, or at least close to it. Now stop playing internet bully, no one likes those.

[jwilliams]

No, I'm telling you that you missed the point.

No, I saw what you wrote, it just does not make sense. Who cares what a hypothetical Chris Mason distro would use for a default filesystem. It is irrelevant. What is relevant is what widely trusted, large distros choose for their default filesystem.

[smitty3268]

Or rather, a couple trolls that are so loud.

/grabs popcorn and leans back

[xeekei]

No, I saw what you wrote, it just does not make sense. Who cares what a hypothetical Chris Mason distro would use for a default filesystem. It is irrelevant. What is relevant is what widely trusted, large distros choose for their default filesystem.

Okay, you are still not getting it. The difference is that in Btrfs case, these "widely trusted, large distros" didn't invent it. While in ZFS case, it did. I'm talking about uneven playing fields. I'm trying to get you to stop bashing the development time.