Re: [opensuse-factory] btrfs woes

Hi, You could try to mount it once with the clear_cache mount option. I know it helped me in the past (few years ago) when such a thing happened. btrfs-freespace should recalculate the actual free space. Aferwards, mount normally again. Kind regards, Erwin On Tue, May 28, 2019 at 1:53 PM Erwin Van de Velde wrote:

Hi,

You could try to mount it once with the clear_cache mount option. I know it helped me in the past (few years ago) when such a thing happened. Aferwards, mount normally again.

Kind regards, Erwin

On Mon, May 27, 2019 at 6:59 PM Oliver Kurz wrote:

...

On Monday, 27 May 2019 02.12.51 CEST Peter Suetterlin wrote:

...

Hi list,

my laptop just died away while running. First thought it was an application error (kstars/EKOS), as that one stoped updating, while e.g. my emacs still worked. But after some 10 more seconds everything stoped responding, and I finally had to press the power button.

After that, the system boot would hang reproducably telling me it's waiting for ridiculous stuff like saving the kill switch state, flushing the journal to persistent storage and some 2 others I couldn't remember.

After three unsuccessful trials, I booted from a rescue disk. Expecting bad things, I tried mounting my root partition - but it did so without complaints.

To be sure, I ran an 'btrfs scrub start /mnt' that finished after some 1-2 minutes. Not really having more ideas I rebooted, and now the system came up fine.

Looking at the last logfile entries I see it was running btrfsmainainance, doing trim and balance. URL: http://paste.opensuse.org/24906643

Interesting cutouts are May 27 00:02:10 woodstock.pitnet btrfs-balance.sh[17806]: /dev/sda2 73G 67G 5.6G 93% / .... May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): balance: start -musage=1 -susage=1 May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): relocating block group 398848950272 flags metadata May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): found 2 extents May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): 1 enospc errors during balance May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): balance: ended with status: -28 May 27 00:02:10 woodstock.pitnet btrfs-balance.sh[17806]: ERROR: error during balancing '/': No space left on device ..... May 27 00:02:10 woodstock.pitnet btrfs-balance.sh[17806]: Done, had to relocate 1 out of 84 chunks May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): balance: start -musage=10 -susage=10 May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): relocating block group 398848950272 flags system May 27 00:02:10 woodstock.pitnet kernel: BTRFS info (device sda2): relocating block group 398815395840 flags metadata May 27 00:02:10 woodstock.pitnet kernel: assertion failed: root, file: fs/btrfs/relocation.c, line: 804 May 27 00:02:10 woodstock.pitnet kernel: ------------[ cut here ]------------ May 27 00:02:10 woodstock.pitnet kernel: kernel BUG at fs/btrfs/ctree.h:3518!

That's where the log ends. Not sure what to do with this. Why is it running out of space with 5.6G left? what are those two last messages?

I can not really answer both those questions however running into "out of space" myself even though there should be enough free space (and metadata space) I deleted more data until all subsequent balance runs with the increasing "-musage" and "-susage" numbers successfully finished.

...

Should this go to bugzilla? Even without proper/complete error messages?

I looked on bugzilla and did not find a report with the specific line "kernel BUG at fs/btrfs/ctree.h:3518!" so I recommend to report a bug.

-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org