Message ID | a70c9f6b4e45d9bcdc5c2f19182f89ef8e22c074.1625237782.git.anand.jain@oracle.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | btrfs/242: test case to fstrim on a degraded filesystem | expand |
On 2021/7/4 下午7:20, Anand Jain wrote: > Create a degraded btrfs filesystem and run fstrim on it. > > Signed-off-by: Anand Jain <anand.jain@oracle.com> > --- > tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ > tests/btrfs/242.out | 2 ++ > 2 files changed, 51 insertions(+) > create mode 100755 tests/btrfs/242 > create mode 100644 tests/btrfs/242.out > > diff --git a/tests/btrfs/242 b/tests/btrfs/242 > new file mode 100755 > index 000000000000..e946ee6ac7c2 > --- /dev/null > +++ b/tests/btrfs/242 > @@ -0,0 +1,49 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (C) 2021 Oracle. All Rights Reserved. > +# > +# FS QA Test 242 > +# > +# Test that fstrim can run on the degraded filesystem > +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() > +# [patch] btrfs: check for missing device in btrfs_trim_fs > + > + > +. ./common/preamble > +_begin_fstest auto quick replace trim > + > +# Import common functions. > +. ./common/filter > + > +# real QA test starts here > +_supported_fs btrfs > +_require_btrfs_forget_or_module_loadable Since you have wipefs requirement commented out, I guess you're also unsure which is the better way. Personally speaking, I think wipefs is better. As it's more generic and needs no kernel support. Despite that it looks good to me. Thanks, Qu > +_require_scratch_dev_pool 2 > +#_require_command "$WIPEFS_PROG" wipefs > + > +_scratch_dev_pool_get 2 > +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') > + > +_scratch_pool_mkfs "-m raid1 -d raid1" > +_scratch_mount > +_require_batched_discard $SCRATCH_MNT > + > +# Add a test file with some data. > +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null > + > +# Unmount the filesystem. > +_scratch_unmount > + > +# Mount the filesystem in degraded mode > +_btrfs_forget_or_module_reload > +_mount -o degraded $dev1 $SCRATCH_MNT > + > +# Run fstrim > +$FSTRIM_PROG $SCRATCH_MNT > + > +_scratch_dev_pool_put > + > +echo Silence is golden > + > +status=0 > +exit > diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out > new file mode 100644 > index 000000000000..a46d77702f23 > --- /dev/null > +++ b/tests/btrfs/242.out > @@ -0,0 +1,2 @@ > +QA output created by 242 > +Silence is golden >
On Sun, Jul 4, 2021 at 12:24 PM Anand Jain <anand.jain@oracle.com> wrote: > > Create a degraded btrfs filesystem and run fstrim on it. > > Signed-off-by: Anand Jain <anand.jain@oracle.com> > --- > tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ > tests/btrfs/242.out | 2 ++ > 2 files changed, 51 insertions(+) > create mode 100755 tests/btrfs/242 > create mode 100644 tests/btrfs/242.out > > diff --git a/tests/btrfs/242 b/tests/btrfs/242 > new file mode 100755 > index 000000000000..e946ee6ac7c2 > --- /dev/null > +++ b/tests/btrfs/242 > @@ -0,0 +1,49 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (C) 2021 Oracle. All Rights Reserved. > +# > +# FS QA Test 242 > +# > +# Test that fstrim can run on the degraded filesystem > +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() > +# [patch] btrfs: check for missing device in btrfs_trim_fs > + > + > +. ./common/preamble > +_begin_fstest auto quick replace trim > + > +# Import common functions. > +. ./common/filter > + > +# real QA test starts here > +_supported_fs btrfs > +_require_btrfs_forget_or_module_loadable > +_require_scratch_dev_pool 2 > +#_require_command "$WIPEFS_PROG" wipefs > + > +_scratch_dev_pool_get 2 > +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') > + > +_scratch_pool_mkfs "-m raid1 -d raid1" > +_scratch_mount > +_require_batched_discard $SCRATCH_MNT > + > +# Add a test file with some data. > +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null What's the point of adding this file (which you copied exactly from btrfs/223) if it's not needed to trigger the bug and it's not used anywhere in the test? Or at the very least, check that reading it after the fstrim returns the correct data, as it might help prevent some regression in the future. Thanks. > + > +# Unmount the filesystem. > +_scratch_unmount > + > +# Mount the filesystem in degraded mode > +_btrfs_forget_or_module_reload > +_mount -o degraded $dev1 $SCRATCH_MNT > + > +# Run fstrim > +$FSTRIM_PROG $SCRATCH_MNT > + > +_scratch_dev_pool_put > + > +echo Silence is golden > + > +status=0 > +exit > diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out > new file mode 100644 > index 000000000000..a46d77702f23 > --- /dev/null > +++ b/tests/btrfs/242.out > @@ -0,0 +1,2 @@ > +QA output created by 242 > +Silence is golden > -- > 2.27.0 >
On Sun, Jul 4, 2021 at 12:24 PM Anand Jain <anand.jain@oracle.com> wrote: > > Create a degraded btrfs filesystem and run fstrim on it. > > Signed-off-by: Anand Jain <anand.jain@oracle.com> > --- > tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ > tests/btrfs/242.out | 2 ++ > 2 files changed, 51 insertions(+) > create mode 100755 tests/btrfs/242 > create mode 100644 tests/btrfs/242.out > > diff --git a/tests/btrfs/242 b/tests/btrfs/242 > new file mode 100755 > index 000000000000..e946ee6ac7c2 > --- /dev/null > +++ b/tests/btrfs/242 > @@ -0,0 +1,49 @@ > +#! /bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# Copyright (C) 2021 Oracle. All Rights Reserved. > +# > +# FS QA Test 242 > +# > +# Test that fstrim can run on the degraded filesystem > +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() > +# [patch] btrfs: check for missing device in btrfs_trim_fs > + > + > +. ./common/preamble > +_begin_fstest auto quick replace trim Also, this does not belong to the 'replace' group (again copied from btrfs/223 it seems). Thanks. > + > +# Import common functions. > +. ./common/filter > + > +# real QA test starts here > +_supported_fs btrfs > +_require_btrfs_forget_or_module_loadable > +_require_scratch_dev_pool 2 > +#_require_command "$WIPEFS_PROG" wipefs > + > +_scratch_dev_pool_get 2 > +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') > + > +_scratch_pool_mkfs "-m raid1 -d raid1" > +_scratch_mount > +_require_batched_discard $SCRATCH_MNT > + > +# Add a test file with some data. > +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null > + > +# Unmount the filesystem. > +_scratch_unmount > + > +# Mount the filesystem in degraded mode > +_btrfs_forget_or_module_reload > +_mount -o degraded $dev1 $SCRATCH_MNT > + > +# Run fstrim > +$FSTRIM_PROG $SCRATCH_MNT > + > +_scratch_dev_pool_put > + > +echo Silence is golden > + > +status=0 > +exit > diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out > new file mode 100644 > index 000000000000..a46d77702f23 > --- /dev/null > +++ b/tests/btrfs/242.out > @@ -0,0 +1,2 @@ > +QA output created by 242 > +Silence is golden > -- > 2.27.0 >
On 05/07/2021 16:50, Filipe Manana wrote: > On Sun, Jul 4, 2021 at 12:24 PM Anand Jain <anand.jain@oracle.com> wrote: >> >> Create a degraded btrfs filesystem and run fstrim on it. >> >> Signed-off-by: Anand Jain <anand.jain@oracle.com> >> --- >> tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ >> tests/btrfs/242.out | 2 ++ >> 2 files changed, 51 insertions(+) >> create mode 100755 tests/btrfs/242 >> create mode 100644 tests/btrfs/242.out >> >> diff --git a/tests/btrfs/242 b/tests/btrfs/242 >> new file mode 100755 >> index 000000000000..e946ee6ac7c2 >> --- /dev/null >> +++ b/tests/btrfs/242 >> @@ -0,0 +1,49 @@ >> +#! /bin/bash >> +# SPDX-License-Identifier: GPL-2.0 >> +# Copyright (C) 2021 Oracle. All Rights Reserved. >> +# >> +# FS QA Test 242 >> +# >> +# Test that fstrim can run on the degraded filesystem >> +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() >> +# [patch] btrfs: check for missing device in btrfs_trim_fs >> + >> + >> +. ./common/preamble >> +_begin_fstest auto quick replace trim > > Also, this does not belong to the 'replace' group (again copied from > btrfs/223 it seems). Oh no. Will fix it. Thanks. - Anand > > Thanks. > >> + >> +# Import common functions. >> +. ./common/filter >> + >> +# real QA test starts here >> +_supported_fs btrfs >> +_require_btrfs_forget_or_module_loadable >> +_require_scratch_dev_pool 2 >> +#_require_command "$WIPEFS_PROG" wipefs >> + >> +_scratch_dev_pool_get 2 >> +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') >> + >> +_scratch_pool_mkfs "-m raid1 -d raid1" >> +_scratch_mount >> +_require_batched_discard $SCRATCH_MNT >> + >> +# Add a test file with some data. >> +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null >> + >> +# Unmount the filesystem. >> +_scratch_unmount >> + >> +# Mount the filesystem in degraded mode >> +_btrfs_forget_or_module_reload >> +_mount -o degraded $dev1 $SCRATCH_MNT >> + >> +# Run fstrim >> +$FSTRIM_PROG $SCRATCH_MNT >> + >> +_scratch_dev_pool_put >> + >> +echo Silence is golden >> + >> +status=0 >> +exit >> diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out >> new file mode 100644 >> index 000000000000..a46d77702f23 >> --- /dev/null >> +++ b/tests/btrfs/242.out >> @@ -0,0 +1,2 @@ >> +QA output created by 242 >> +Silence is golden >> -- >> 2.27.0 >> > >
On 05/07/2021 16:45, Filipe Manana wrote: > On Sun, Jul 4, 2021 at 12:24 PM Anand Jain <anand.jain@oracle.com> wrote: >> >> Create a degraded btrfs filesystem and run fstrim on it. >> >> Signed-off-by: Anand Jain <anand.jain@oracle.com> >> --- >> tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ >> tests/btrfs/242.out | 2 ++ >> 2 files changed, 51 insertions(+) >> create mode 100755 tests/btrfs/242 >> create mode 100644 tests/btrfs/242.out >> >> diff --git a/tests/btrfs/242 b/tests/btrfs/242 >> new file mode 100755 >> index 000000000000..e946ee6ac7c2 >> --- /dev/null >> +++ b/tests/btrfs/242 >> @@ -0,0 +1,49 @@ >> +#! /bin/bash >> +# SPDX-License-Identifier: GPL-2.0 >> +# Copyright (C) 2021 Oracle. All Rights Reserved. >> +# >> +# FS QA Test 242 >> +# >> +# Test that fstrim can run on the degraded filesystem >> +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() >> +# [patch] btrfs: check for missing device in btrfs_trim_fs >> + >> + >> +. ./common/preamble >> +_begin_fstest auto quick replace trim >> + >> +# Import common functions. >> +. ./common/filter >> + >> +# real QA test starts here >> +_supported_fs btrfs >> +_require_btrfs_forget_or_module_loadable >> +_require_scratch_dev_pool 2 >> +#_require_command "$WIPEFS_PROG" wipefs >> + >> +_scratch_dev_pool_get 2 >> +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') >> + >> +_scratch_pool_mkfs "-m raid1 -d raid1" >> +_scratch_mount >> +_require_batched_discard $SCRATCH_MNT >> + >> +# Add a test file with some data. >> +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null > > What's the point of adding this file (which you copied exactly from > btrfs/223) if it's not needed to trigger the bug and it's not used > anywhere in the test? > Or at the very least, check that reading it after the fstrim returns > the correct data, as it might help prevent some regression in the > future. There is no need for data to trigger the bug. I wanted some data as practically it is. I like the idea to prevent some regression in the future- I will add code to check if data is correct after fstrim. Thanks, Anand > > Thanks. > >> + >> +# Unmount the filesystem. >> +_scratch_unmount >> + >> +# Mount the filesystem in degraded mode >> +_btrfs_forget_or_module_reload >> +_mount -o degraded $dev1 $SCRATCH_MNT >> + >> +# Run fstrim >> +$FSTRIM_PROG $SCRATCH_MNT >> + >> +_scratch_dev_pool_put >> + >> +echo Silence is golden >> + >> +status=0 >> +exit >> diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out >> new file mode 100644 >> index 000000000000..a46d77702f23 >> --- /dev/null >> +++ b/tests/btrfs/242.out >> @@ -0,0 +1,2 @@ >> +QA output created by 242 >> +Silence is golden >> -- >> 2.27.0 >> > >
diff --git a/tests/btrfs/242 b/tests/btrfs/242 new file mode 100755 index 000000000000..e946ee6ac7c2 --- /dev/null +++ b/tests/btrfs/242 @@ -0,0 +1,49 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2021 Oracle. All Rights Reserved. +# +# FS QA Test 242 +# +# Test that fstrim can run on the degraded filesystem +# Kernel requires fix for the null pointer deref in btrfs_trim_fs() +# [patch] btrfs: check for missing device in btrfs_trim_fs + + +. ./common/preamble +_begin_fstest auto quick replace trim + +# Import common functions. +. ./common/filter + +# real QA test starts here +_supported_fs btrfs +_require_btrfs_forget_or_module_loadable +_require_scratch_dev_pool 2 +#_require_command "$WIPEFS_PROG" wipefs + +_scratch_dev_pool_get 2 +dev1=$(echo $SCRATCH_DEV_POOL | $AWK_PROG '{ print $1 }') + +_scratch_pool_mkfs "-m raid1 -d raid1" +_scratch_mount +_require_batched_discard $SCRATCH_MNT + +# Add a test file with some data. +$XFS_IO_PROG -f -c "pwrite -S 0xab 0 10M" $SCRATCH_MNT/foo > /dev/null + +# Unmount the filesystem. +_scratch_unmount + +# Mount the filesystem in degraded mode +_btrfs_forget_or_module_reload +_mount -o degraded $dev1 $SCRATCH_MNT + +# Run fstrim +$FSTRIM_PROG $SCRATCH_MNT + +_scratch_dev_pool_put + +echo Silence is golden + +status=0 +exit diff --git a/tests/btrfs/242.out b/tests/btrfs/242.out new file mode 100644 index 000000000000..a46d77702f23 --- /dev/null +++ b/tests/btrfs/242.out @@ -0,0 +1,2 @@ +QA output created by 242 +Silence is golden
Create a degraded btrfs filesystem and run fstrim on it. Signed-off-by: Anand Jain <anand.jain@oracle.com> --- tests/btrfs/242 | 49 +++++++++++++++++++++++++++++++++++++++++++++ tests/btrfs/242.out | 2 ++ 2 files changed, 51 insertions(+) create mode 100755 tests/btrfs/242 create mode 100644 tests/btrfs/242.out