Skip to contentSkip to conversation

Linux swap file not activating after reboot

DISCUSSION Michael Ruiz 4 participants 6 messages Jan 16, 2020 – Jan 17, 2020 Btrfs file system

lore
Share & tools
lore
#0 lore raw

Hi,
I have a //@swap subvolume and i have a swapfile within it. I mount the
subvolume like such in fstab:

rw,ssd,nofail,noautodefrag,nodatacow,nodatasum,subvolid=1234,subvol=/@swap

It mounts correctly, but 1/15/20 4:20 PM kernel I get:

BTRFS warning (device dm-0): swapfile must not be copy-on-write

I did chattr +C on the empty swapfile as per arch wiki instructions. The only problem is that it does not work after reboot... Does btrfs allow subvolumes to have different mount options?

#1 lore raw
Quoted 6 lines · Michael Ruiz

Hi,
I have a //@swap subvolume and i have a swapfile within it. I mount the
subvolume like such in fstab:

rw,ssd,nofail,noautodefrag,nodatacow,nodatasum,subvolid=1234,subvol=/@swap

Quoted 7 lines

It mounts correctly, but 1/15/20 4:20 PM kernel I get:

BTRFS warning (device dm-0): swapfile must not be copy-on-write

I did chattr +C on the empty swapfile as per arch wiki instructions. The only problem is that it does not work after reboot... Does btrfs allow subvolumes to have different mount options?

If that file has nodatacow, then it may be possible that you had a snapshot of that subvolume, and caused problem.

Thanks,
Qu

#2 lore raw
In reply to Michael Ruiz
Quoted 9 lines · Michael Ruiz

Hi,
I have a //@swap subvolume and i have a swapfile within it. I mount the
subvolume like such in fstab:

rw,ssd,nofail,noautodefrag,nodatacow,nodatasum,subvolid=1234,subvol=/@swap

It mounts correctly, but 1/15/20 4:20 PM kernel I get:

BTRFS warning (device dm-0): swapfile must not be copy-on-write

There are two reasons why the message is printed, one is when the file does not have the C attribute and another one when the the existing file extents need to be COWed (same case as if the file is NOCOW and has been snapshotted).

Plain reboot will not change the C attribute, so either there's a snapshot of /@snap or the check of a used swapfile is wrong.

I tested it here, a swapfile that got almost full after a stress test, followed by reboot and swapon (without any change to the file) was ok.

Doing a snapshot and swapon resulted in the message you saw.

After deleting the snapshot and waiting until it gets cleaned, swapon did not activate the file anymore. Filefrag or fiemap don't report any shared extents so here I' expect that the file should be in a valid state for swapon.

Omar, any ideas?

#3 lore raw
Quoted 22 lines · David Sterba
Quoted 9 lines · Michael Ruiz

Hi,
I have a //@swap subvolume and i have a swapfile within it. I mount the
subvolume like such in fstab:

rw,ssd,nofail,noautodefrag,nodatacow,nodatasum,subvolid=1234,subvol=/@swap

It mounts correctly, but 1/15/20 4:20 PM kernel I get:

BTRFS warning (device dm-0): swapfile must not be copy-on-write

There are two reasons why the message is printed, one is when the file does not have the C attribute and another one when the the existing file extents need to be COWed (same case as if the file is NOCOW and has been snapshotted).

Plain reboot will not change the C attribute, so either there's a snapshot of /@snap or the check of a used swapfile is wrong.

I tested it here, a swapfile that got almost full after a stress test, followed by reboot and swapon (without any change to the file) was ok.

Quoted 8 lines

Doing a snapshot and swapon resulted in the message you saw.

After deleting the snapshot and waiting until it gets cleaned, swapon did not activate the file anymore. Filefrag or fiemap don't report any shared extents so here I' expect that the file should be in a valid state for swapon.

Omar, any ideas?

Hm, we're hitting this check in can_nocow_extent():

if (btrfs_file_extent_generation(leaf, fi) <=
btrfs_root_last_snapshot(&root->root_item))

That check was added in 78d4295b1eee ("btrfs: lift some btrfs_cross_ref_exist checks in nocow path") as an optimization. Even if we comment that out, we'll hit the similar check in btrfs_cross_ref_exist():

        /* If extent created before last snapshot => it's definitely shared */
        if (btrfs_extent_generation(leaf, ei) <=
            btrfs_root_last_snapshot(&root->root_item))

That's not quite right in exactly this case that the snapshot has been deleted. Apparently we've been doing unnecessary COW for this case. I'll need to think about how to safely avoid these checks without too much of a performance hit.

Thanks for the report!

#4 lore raw
Quoted 49 lines · PST you
Quoted 25 lines · David Sterba
Quoted 4 lines · Michael Ruiz

Hi,

I have a //@swap subvolume and i have a swapfile within it. I mount the

Quoted 8 lines

subvolume like such in fstab:

rw,ssd,nofail,noautodefrag,nodatacow,nodatasum,subvolid=1234,subvol=/@s wap

It mounts correctly, but 1/15/20 4:20 PM kernel I get:

BTRFS warning (device dm-0): swapfile must not be copy-on-write

There are two reasons why the message is printed, one is when the file does not have the C attribute and another one when the the existing file extents need to be COWed (same case as if the file is NOCOW and has been snapshotted).

Plain reboot will not change the C attribute, so either there's a snapshot of /@snap or the check of a used swapfile is wrong.

I tested it here, a swapfile that got almost full after a stress test, followed by reboot and swapon (without any change to the file) was ok.

Quoted 8 lines

Doing a snapshot and swapon resulted in the message you saw.

After deleting the snapshot and waiting until it gets cleaned, swapon did not activate the file anymore. Filefrag or fiemap don't report any shared extents so here I' expect that the file should be in a valid state for swapon.

Omar, any ideas?

Hm, we're hitting this check in can_nocow_extent():

if (btrfs_file_extent_generation(leaf, fi) <=
btrfs_root_last_snapshot(&root->root_item))

That check was added in 78d4295b1eee ("btrfs: lift some btrfs_cross_ref_exist checks in nocow path") as an optimization. Even if we comment that out, we'll hit the similar check in btrfs_cross_ref_exist():

        /* If extent created before last snapshot => it's definitely shared
*/ if (btrfs_extent_generation(leaf, ei) <=
            btrfs_root_last_snapshot(&root->root_item))
Quoted 6 lines

That's not quite right in exactly this case that the snapshot has been deleted. Apparently we've been doing unnecessary COW for this case. I'll need to think about how to safely avoid these checks without too much of a performance hit.

Thanks for the report!

My solution was to boot into an arch live usb, unlock my dmcrypt partition and mount the btrfs partition to /mnt. After that I created a subvolume on the @ directory (top level 5) instead of a subvolume of my root (/) partition. So now my subvolume layout is like this:

ID 256 gen 45343 top level 5 path @
ID 257 gen 45346 top level 5 path @home
ID 258 gen 45346 top level 5 path @log
ID 259 gen 44303 top level 5 path @srv
ID 260 gen 44650 top level 5 path @pkg
ID 261 gen 45346 top level 5 path @tmp
ID 1607 gen 43963 top level 5 path @swap

The strange thing to me is that I didn't ask for snapshots of this subvolume, although I do keep snapshots of my / directory, I was under the impression that snapshots would not be recursive and go into the /swap subvolume. I can also confirm I had the +C attribute while getting this error. So now I am able to mount this subvolume with it's own options, whereas before I guess it inherited options from the root dir which has CoW enabled. The problem is now resolved by doing this. Thanks for the responses.

#5 lore raw
Quoted 13 lines · Michael Ruiz

My solution was to boot into an arch live usb, unlock my dmcrypt partition and mount the btrfs partition to /mnt. After that I created a subvolume on the @ directory (top level 5) instead of a subvolume of my root (/) partition. So now my subvolume layout is like this:

ID 256 gen 45343 top level 5 path @
ID 257 gen 45346 top level 5 path @home
ID 258 gen 45346 top level 5 path @log
ID 259 gen 44303 top level 5 path @srv
ID 260 gen 44650 top level 5 path @pkg
ID 261 gen 45346 top level 5 path @tmp
ID 1607 gen 43963 top level 5 path @swap

Quoted 3 lines

The strange thing to me is that I didn't ask for snapshots of this subvolume, although I do keep snapshots of my / directory, I was under the impression that snapshots would not be recursive and go into the /swap subvolume.

No, snapshots are not recursive.

Quoted 5 lines

I can
also confirm I had the +C attribute while getting this error. So now I am able to mount this subvolume with it's own options, whereas before I guess it inherited options from the root dir which has CoW enabled. The problem is now resolved by doing this. Thanks for the responses.

The file attribute has precedence over the mount options, so it should not matter if you had nodatacow or not when creating the file (assuming you set the +C flag manually).