| Message ID | 53cdc0c4696655a0e7a5eb62612a7b87ff4d48cd.1720547422.git.boris@bur.io (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | btrfs: add test for subvolid reuse with squota | expand |
On Tue, Jul 09, 2024 at 10:51:04AM -0700, Boris Burkov wrote: > Squotas are likely to leave qgroups that outlive deleted subvolids. > Before the kernel patch > 2b8aa78cf127 ("btrfs: qgroup: fix qgroup id collision across mounts") > this would lead to a repeated subvolid which would collide on an > existing qgroup id and error out with EEXIST. In snapshot creation, this > would lead to a read only fs. > > Add a test which exercises the path that could create duplicate > subvolids but with squotas enabled, which should avoid the trap. > > Signed-off-by: Boris Burkov <boris@bur.io> Reviewed-by: Josef Bacik <josef@toxicpanda.com> Thanks, Josef
On 10/07/2024 01:51, Boris Burkov wrote: > Squotas are likely to leave qgroups that outlive deleted subvolids. > Before the kernel patch > 2b8aa78cf127 ("btrfs: qgroup: fix qgroup id collision across mounts") > this would lead to a repeated subvolid which would collide on an > existing qgroup id and error out with EEXIST. In snapshot creation, this > would lead to a read only fs. > > Add a test which exercises the path that could create duplicate > subvolids but with squotas enabled, which should avoid the trap. > > Signed-off-by: Boris Burkov <boris@bur.io> Reviewed-by: Anand Jain <anand.jain@oracle.com> Applied. Thanks, Anand > --- > tests/btrfs/331 | 45 +++++++++++++++++++++++++++++++++++++++++++++ > tests/btrfs/331.out | 2 ++ > 2 files changed, 47 insertions(+) > create mode 100755 tests/btrfs/331 > create mode 100644 tests/btrfs/331.out > > diff --git a/tests/btrfs/331 b/tests/btrfs/331 > new file mode 100755 > index 000000000..8a99d5527 > --- /dev/null > +++ b/tests/btrfs/331 > @@ -0,0 +1,45 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (c) 2024 Meta Platforms, Inc. All Rights Reserved. > +# > +# FS QA Test 331 > +# > +# Test that btrfs does not recycle subvolume ids across remounts in a way that > +# breaks squotas. > +# > +. ./common/preamble > +_begin_fstest auto quick qgroup snapshot > + > +_fixed_by_kernel_commit 2b8aa78cf127 \ > + "btrfs: qgroup: fix qgroup id collision across mounts" > + > +. ./common/btrfs > +_supported_fs btrfs > +_require_scratch_enable_simple_quota > + > +_scratch_mkfs >> $seqres.full > +_scratch_mount > +$BTRFS_UTIL_PROG quota enable --simple $SCRATCH_MNT > + > +# Create a gap in the subvolume IDs > +sv=$SCRATCH_MNT/sv > +for i in $(seq 100); do > + $BTRFS_UTIL_PROG subvolume create $sv.$i >> $seqres.full > +done > +for i in $(seq 50 100); do > + $BTRFS_UTIL_PROG subvolume delete $sv.$i >> $seqres.full > +done > + > +# Cycle mount triggers reading the tree_root's free objectid. > +_scratch_cycle_mount > + > +# Create snapshots that could go into the used subvolid space. > +$BTRFS_UTIL_PROG subvolume create $sv.BOOM >> $seqres.full > +for i in $(seq 10); do > + $BTRFS_UTIL_PROG subvolume snapshot $sv.BOOM $sv.BOOM.$i >> $seqres.full > +done > + > +# success, all done > +echo "Silence is golden" > +status=0 > +exit > diff --git a/tests/btrfs/331.out b/tests/btrfs/331.out > new file mode 100644 > index 000000000..0097e61d8 > --- /dev/null > +++ b/tests/btrfs/331.out > @@ -0,0 +1,2 @@ > +QA output created by 331 > +Silence is golden
diff --git a/tests/btrfs/331 b/tests/btrfs/331 new file mode 100755 index 000000000..8a99d5527 --- /dev/null +++ b/tests/btrfs/331 @@ -0,0 +1,45 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (c) 2024 Meta Platforms, Inc. All Rights Reserved. +# +# FS QA Test 331 +# +# Test that btrfs does not recycle subvolume ids across remounts in a way that +# breaks squotas. +# +. ./common/preamble +_begin_fstest auto quick qgroup snapshot + +_fixed_by_kernel_commit 2b8aa78cf127 \ + "btrfs: qgroup: fix qgroup id collision across mounts" + +. ./common/btrfs +_supported_fs btrfs +_require_scratch_enable_simple_quota + +_scratch_mkfs >> $seqres.full +_scratch_mount +$BTRFS_UTIL_PROG quota enable --simple $SCRATCH_MNT + +# Create a gap in the subvolume IDs +sv=$SCRATCH_MNT/sv +for i in $(seq 100); do + $BTRFS_UTIL_PROG subvolume create $sv.$i >> $seqres.full +done +for i in $(seq 50 100); do + $BTRFS_UTIL_PROG subvolume delete $sv.$i >> $seqres.full +done + +# Cycle mount triggers reading the tree_root's free objectid. +_scratch_cycle_mount + +# Create snapshots that could go into the used subvolid space. +$BTRFS_UTIL_PROG subvolume create $sv.BOOM >> $seqres.full +for i in $(seq 10); do + $BTRFS_UTIL_PROG subvolume snapshot $sv.BOOM $sv.BOOM.$i >> $seqres.full +done + +# success, all done +echo "Silence is golden" +status=0 +exit diff --git a/tests/btrfs/331.out b/tests/btrfs/331.out new file mode 100644 index 000000000..0097e61d8 --- /dev/null +++ b/tests/btrfs/331.out @@ -0,0 +1,2 @@ +QA output created by 331 +Silence is golden
Squotas are likely to leave qgroups that outlive deleted subvolids. Before the kernel patch 2b8aa78cf127 ("btrfs: qgroup: fix qgroup id collision across mounts") this would lead to a repeated subvolid which would collide on an existing qgroup id and error out with EEXIST. In snapshot creation, this would lead to a read only fs. Add a test which exercises the path that could create duplicate subvolids but with squotas enabled, which should avoid the trap. Signed-off-by: Boris Burkov <boris@bur.io> --- tests/btrfs/331 | 45 +++++++++++++++++++++++++++++++++++++++++++++ tests/btrfs/331.out | 2 ++ 2 files changed, 47 insertions(+) create mode 100755 tests/btrfs/331 create mode 100644 tests/btrfs/331.out