| Message ID | 20171124050144.23549-1-eguan@redhat.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
[cc: Karel Zak] On Fri, Nov 24, 2017 at 7:01 AM, Eryu Guan <eguan@redhat.com> wrote: > util-linux commit 6dede2f2f7c5 ("libmount: support MS_RDONLY on > write-protected devices") changed the error message on read-only > block device, and in the failure case printed one line message > instead of two (for details please see comments in common/filter), > and this change broke generic/050 and overlay/035. > > Fix it by adding more filter rules to _filter_ro_mount and updating > associated .out files to unify the output from both old and new > util-linux versions. > > Signed-off-by: Eryu Guan <eguan@redhat.com> > --- > v4 (also passed with both old & new util-linux): > - add more comments to explain the output differences between util-linux > versions > - print out message directly instead of using perl variables > - add word "device" to failed ro mount message Eryu, This looks good and you can take it as an ACK on the series. I've CC'ed Karel, so maybe he thinks about us fstests guys before making these sort of changes again... Thinking out loud, does xfstest even need to use mount program from util-linux? Do we ever need anything other than the bare libc mount(2)? We need it for -o loop, but that is the exception to the rule. For fs that have mount helpers (cifs,nfs), we could use mount.$FSTYP directly (what error formats are reported from the helpers??). For all other fs we can write a simple t_mount program to wrap libc mount(2) and not be dependent on util-linux error message formats. Maybe we can consider it next time util-linux changes.. Another idea to through in the air in the direction of Karel - Maybe it makes sense for util-linux to check some env variable and then print all error messages in a unified machine format, e.g.: fprintf("%s: errno=%d\n", progname, errno); Cheers, Amir. > > v3: > - document the filtered format in comments > - remove legacy sed filter, the perl filter covers the legacy case well > - filter out $SCRATCH_DEV/MNT too and use a consistent output > - remove the new filter_mount helper in overlay/035 > > common/filter | 59 +++++++++++++++++++++++++++++++++++++++++++++++---- > tests/generic/050 | 8 +++---- > tests/generic/050.out | 8 +++---- > tests/overlay/035 | 4 ++-- > tests/overlay/035.out | 4 ++-- > 5 files changed, 67 insertions(+), 16 deletions(-) > > diff --git a/common/filter b/common/filter > index a212c09aa138..9c33efacea6c 100644 > --- a/common/filter > +++ b/common/filter > @@ -397,11 +397,62 @@ _filter_ending_dot() > sed -e "s/\.$//" > } > > -# Older mount output referred to "block device" when mounting RO devices > -# It's gone in newer versions > +# Older mount output referred to "block device" when mounting RO devices. It's > +# gone in newer versions. v2.30 changed the output again. This filter is to > +# unify all read-only mount messages across all util-linux versions. > +# > +# for a successful ro mount: > +# ancient: mount: block device <device> is write-protected, mounting read-only > +# prior to v2.30: mount: <device> is write-protected, mounting read-only > +# v2.30 and later: mount: <mountpoint>: WARNING: device write-protected, mounted read-only. > +# > +# a failed ro mount: > +# ancient (two-line message): > +# mount: block device <device> is write-protected, mounting read-only > +# mount: cannot mount block device <device> read-only > +# prior to v2.30 (two-line message): > +# mount: <device> is write-protected, mounting read-only > +# mount: cannot mount <device> read-only > +# v2.30 and later (single-line message): > +# mount: <mountpoint>: cannot mount <device> read-only. > +# > +# a failed rw remount: > +# ancient: mount: cannot remount block device <device> read-write, is write-protected > +# prior to v2.30: mount: cannot remount <device> read-write, is write-protected > +# v2.30 and later: mount: <mountpoint>: cannot remount <device> read-write, is write-protected. > +# > +# Now use _filter_ro_mount to unify all these differences across old & new > +# util-linux versions. So the filtered format would be: > +# > +# successful ro mount: > +# mount: device write-protected, mounting read-only > +# > +# failed ro mount: > +# mount: device write-protected, mounting read-only > +# mount: cannot mount device read-only > +# > +# failed rw remount: > +# mount: cannot remount device read-write, is write-protected > _filter_ro_mount() { > - sed -e "s/mount: block device/mount:/g" \ > - -e "s/mount: cannot mount block device/mount: cannot mount/g" > + perl -ne ' > + if (/write-protected, mount.*read-only/) { > + # filter successful ro mount, and first line of prior to v2.30 > + # format failed ro mount > + print "mount: device write-protected, mounting read-only\n"; > + } elsif (/mount: .*: cannot mount.*read-only/) { > + # filter v2.30 format failed ro mount, convert single-line > + # message to two-line message > + print "mount: device write-protected, mounting read-only\n"; > + print "mount: cannot mount device read-only\n"; > + } elsif (/^mount: cannot mount .* read-only$/) { > + # filter prior to v2.30 format failed ro mount > + print "mount: cannot mount device read-only\n"; > + } elsif (/mount:.* cannot remount .* read-write.*/) { > + # filter failed rw remount > + print "mount: cannot remount device read-write, is write-protected\n"; > + } else { > + print "$_"; > + }' | _filter_ending_dot > } > > # Filter a failed mount output due to EUCLEAN and USTALE, util-linux changed > diff --git a/tests/generic/050 b/tests/generic/050 > index 5fa28a7648e5..efa45f04825b 100755 > --- a/tests/generic/050 > +++ b/tests/generic/050 > @@ -60,7 +60,7 @@ blockdev --setro $SCRATCH_DEV > # Mount it, and make sure we can't write to it, and we can unmount it again > # > echo "mounting read-only block device:" > -_scratch_mount 2>&1 | _filter_scratch | _filter_ro_mount > +_scratch_mount 2>&1 | _filter_ro_mount > > echo "touching file on read-only filesystem (should fail)" > touch $SCRATCH_MNT/foo 2>&1 | _filter_scratch > @@ -95,10 +95,10 @@ blockdev --setro $SCRATCH_DEV > # -o norecovery is used. > # > echo "mounting filesystem that needs recovery on a read-only device:" > -_scratch_mount 2>&1 | _filter_scratch | _filter_ro_mount > +_scratch_mount 2>&1 | _filter_ro_mount > > echo "unmounting read-only filesystem" > -_scratch_unmount 2>&1 | _filter_scratch > +_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot > > # > # This is the way out if the underlying device really is read-only. > @@ -106,7 +106,7 @@ _scratch_unmount 2>&1 | _filter_scratch > # data recovery hack. > # > echo "mounting filesystem with -o norecovery on a read-only device:" > -_scratch_mount -o norecovery 2>&1 | _filter_scratch | _filter_ro_mount > +_scratch_mount -o norecovery 2>&1 | _filter_ro_mount > > echo "unmounting read-only filesystem" > _scratch_unmount 2>&1 | _filter_scratch > diff --git a/tests/generic/050.out b/tests/generic/050.out > index fb90f6ea5819..7d70ddee83cb 100644 > --- a/tests/generic/050.out > +++ b/tests/generic/050.out > @@ -1,7 +1,7 @@ > QA output created by 050 > setting device read-only > mounting read-only block device: > -mount: SCRATCH_DEV is write-protected, mounting read-only > +mount: device write-protected, mounting read-only > touching file on read-only filesystem (should fail) > touch: cannot touch 'SCRATCH_MNT/foo': Read-only file system > unmounting read-only filesystem > @@ -12,12 +12,12 @@ going down: > unmounting shutdown filesystem: > setting device read-only > mounting filesystem that needs recovery on a read-only device: > -mount: SCRATCH_DEV is write-protected, mounting read-only > -mount: cannot mount SCRATCH_DEV read-only > +mount: device write-protected, mounting read-only > +mount: cannot mount device read-only > unmounting read-only filesystem > umount: SCRATCH_DEV: not mounted > mounting filesystem with -o norecovery on a read-only device: > -mount: SCRATCH_DEV is write-protected, mounting read-only > +mount: device write-protected, mounting read-only > unmounting read-only filesystem > setting device read-write > mounting filesystem that needs recovery with -o ro: > diff --git a/tests/overlay/035 b/tests/overlay/035 > index 64fcd708105e..05447741a1ba 100755 > --- a/tests/overlay/035 > +++ b/tests/overlay/035 > @@ -69,7 +69,7 @@ mkdir -p $lowerdir1 $lowerdir2 $upperdir $workdir > $MOUNT_PROG -t overlay -o"lowerdir=$lowerdir2:$lowerdir1" \ > $OVL_BASE_SCRATCH_MNT $SCRATCH_MNT > touch $SCRATCH_MNT/foo 2>&1 | _filter_scratch > -_scratch_remount rw 2>&1 | _filter_scratch > +_scratch_remount rw 2>&1 | _filter_ro_mount > $UMOUNT_PROG $SCRATCH_MNT > > # Make workdir immutable to prevent workdir re-create on mount > @@ -79,7 +79,7 @@ $CHATTR_PROG +i $workdir > # Verify that overlay is mounted read-only and that it cannot be remounted rw. > _overlay_scratch_mount_dirs $lowerdir2 $upperdir $workdir > touch $SCRATCH_MNT/bar 2>&1 | _filter_scratch > -_scratch_remount rw 2>&1 | _filter_scratch > +_scratch_remount rw 2>&1 | _filter_ro_mount > > # success, all done > status=0 > diff --git a/tests/overlay/035.out b/tests/overlay/035.out > index 5a5f67771402..e08ba2ebc691 100644 > --- a/tests/overlay/035.out > +++ b/tests/overlay/035.out > @@ -1,5 +1,5 @@ > QA output created by 035 > touch: cannot touch 'SCRATCH_MNT/foo': Read-only file system > -mount: cannot remount SCRATCH_DEV read-write, is write-protected > +mount: cannot remount device read-write, is write-protected > touch: cannot touch 'SCRATCH_MNT/bar': Read-only file system > -mount: cannot remount SCRATCH_DEV read-write, is write-protected > +mount: cannot remount device read-write, is write-protected > -- > 2.14.3 > -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Nov 24, 2017 at 10:04:33AM +0200, Amir Goldstein wrote: > [cc: Karel Zak] > > On Fri, Nov 24, 2017 at 7:01 AM, Eryu Guan <eguan@redhat.com> wrote: > > util-linux commit 6dede2f2f7c5 ("libmount: support MS_RDONLY on > > write-protected devices") changed the error message on read-only > > block device, and in the failure case printed one line message > > instead of two (for details please see comments in common/filter), > > and this change broke generic/050 and overlay/035. > > > > Fix it by adding more filter rules to _filter_ro_mount and updating > > associated .out files to unify the output from both old and new > > util-linux versions. > > > > Signed-off-by: Eryu Guan <eguan@redhat.com> > > --- > > v4 (also passed with both old & new util-linux): > > - add more comments to explain the output differences between util-linux > > versions > > - print out message directly instead of using perl variables > > - add word "device" to failed ro mount message > > > Eryu, > > This looks good and you can take it as an ACK on the series. > > I've CC'ed Karel, so maybe he thinks about us fstests guys before making > these sort of changes again... Frankly, we have never promised that things like warning messages (or another messages) are stable interface. It's fragile to depend on this stuff... > Thinking out loud, does xfstest even need to use mount program from > util-linux? Do we ever need anything other than the bare libc mount(2)? > We need it for -o loop, but that is the exception to the rule. Well, I don't think that create a parallel universe is the best solution. > For fs that have mount helpers (cifs,nfs), we could use mount.$FSTYP > directly (what error formats are reported from the helpers??). For all other > fs we can write a simple t_mount program to wrap libc mount(2) and not be > dependent on util-linux error message formats. > Maybe we can consider it next time util-linux changes.. > > Another idea to through in the air in the direction of Karel - > Maybe it makes sense for util-linux to check some env variable > and then print all error messages in a unified machine format, e.g.: > fprintf("%s: errno=%d\n", progname, errno); This is good idea, I can try to implement it into libmount. Karel
On Fri, Nov 24, 2017 at 12:38 PM, Karel Zak <kzak@redhat.com> wrote: > On Fri, Nov 24, 2017 at 10:04:33AM +0200, Amir Goldstein wrote: >> [cc: Karel Zak] >> >> On Fri, Nov 24, 2017 at 7:01 AM, Eryu Guan <eguan@redhat.com> wrote: >> > util-linux commit 6dede2f2f7c5 ("libmount: support MS_RDONLY on >> > write-protected devices") changed the error message on read-only >> > block device, and in the failure case printed one line message >> > instead of two (for details please see comments in common/filter), >> > and this change broke generic/050 and overlay/035. >> > >> > Fix it by adding more filter rules to _filter_ro_mount and updating >> > associated .out files to unify the output from both old and new >> > util-linux versions. >> > >> > Signed-off-by: Eryu Guan <eguan@redhat.com> >> > --- >> > v4 (also passed with both old & new util-linux): >> > - add more comments to explain the output differences between util-linux >> > versions >> > - print out message directly instead of using perl variables >> > - add word "device" to failed ro mount message >> >> >> Eryu, >> >> This looks good and you can take it as an ACK on the series. >> >> I've CC'ed Karel, so maybe he thinks about us fstests guys before making >> these sort of changes again... > > Frankly, we have never promised that things like warning messages (or > another messages) are stable interface. It's fragile to depend on this > stuff... > I know. No complains. Just wanted to inform you of the downstream ripples. >> Thinking out loud, does xfstest even need to use mount program from >> util-linux? Do we ever need anything other than the bare libc mount(2)? >> We need it for -o loop, but that is the exception to the rule. > > Well, I don't think that create a parallel universe is the best > solution. > I agree. However, xfstest uses 'mount' is a very particular way: - User has to provide both block device and mount point - xfstests already checks that those block device and mount point are not in use in /proc/mounts *before* calling 'mount' So I prefer the 'machine output format' solution going forward, but since xfstests will need to continue supporting old util-linux installations it might be less clumsy to use a simple C wrapper to mount(2), then to carry these error format filters. xfstests can also check for availability of the new machine format in 'mount' and if it doesn't exist fall back to using its private mount helper. >> For fs that have mount helpers (cifs,nfs), we could use mount.$FSTYP >> directly (what error formats are reported from the helpers??). For all other >> fs we can write a simple t_mount program to wrap libc mount(2) and not be >> dependent on util-linux error message formats. >> Maybe we can consider it next time util-linux changes.. >> >> Another idea to through in the air in the direction of Karel - >> Maybe it makes sense for util-linux to check some env variable >> and then print all error messages in a unified machine format, e.g.: >> fprintf("%s: errno=%d\n", progname, errno); > > This is good idea, I can try to implement it into libmount. > That would be great. Thanks, Amir. -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Sun, Nov 26, 2017 at 10:56:40AM +0200, Amir Goldstein wrote: > >> Thinking out loud, does xfstest even need to use mount program from > >> util-linux? Do we ever need anything other than the bare libc mount(2)? > >> We need it for -o loop, but that is the exception to the rule. > > > > Well, I don't think that create a parallel universe is the best > > solution. > > > > I agree. However, xfstest uses 'mount' is a very particular way: > - User has to provide both block device and mount point > - xfstests already checks that those block device and mount point > are not in use in /proc/mounts *before* calling 'mount' OK, I see, than the wrapper is not so bad idea (if you're able to keep it simple and stupid:-) > So I prefer the 'machine output format' solution going forward, but > since xfstests will need to continue supporting old util-linux installations > it might be less clumsy to use a simple C wrapper to mount(2), then > to carry these error format filters. > xfstests can also check for availability of the new machine format in > 'mount' and if it doesn't exist fall back to using its private mount helper. If you want to implement the wrapper than it's probably better use it in all cases and then you can remove all the backward compatibility filters. In thins case it seems better to depend on the wrapper only. So, the unified messages maintained by unit-linux mount(8) will be unnecessary. Karel
On Sun, Nov 26, 2017 at 10:56:40AM +0200, Amir Goldstein wrote: > On Fri, Nov 24, 2017 at 12:38 PM, Karel Zak <kzak@redhat.com> wrote: > > On Fri, Nov 24, 2017 at 10:04:33AM +0200, Amir Goldstein wrote: > >> [cc: Karel Zak] > >> > >> On Fri, Nov 24, 2017 at 7:01 AM, Eryu Guan <eguan@redhat.com> wrote: > >> > util-linux commit 6dede2f2f7c5 ("libmount: support MS_RDONLY on > >> > write-protected devices") changed the error message on read-only > >> > block device, and in the failure case printed one line message > >> > instead of two (for details please see comments in common/filter), > >> > and this change broke generic/050 and overlay/035. > >> > > >> > Fix it by adding more filter rules to _filter_ro_mount and updating > >> > associated .out files to unify the output from both old and new > >> > util-linux versions. > >> > > >> > Signed-off-by: Eryu Guan <eguan@redhat.com> > >> > --- > >> > v4 (also passed with both old & new util-linux): > >> > - add more comments to explain the output differences between util-linux > >> > versions > >> > - print out message directly instead of using perl variables > >> > - add word "device" to failed ro mount message > >> > >> > >> Eryu, > >> > >> This looks good and you can take it as an ACK on the series. > >> > >> I've CC'ed Karel, so maybe he thinks about us fstests guys before making > >> these sort of changes again... > > > > Frankly, we have never promised that things like warning messages (or > > another messages) are stable interface. It's fragile to depend on this > > stuff... > > > > I know. No complains. Just wanted to inform you of the downstream > ripples. Agreed. > > >> Thinking out loud, does xfstest even need to use mount program from > >> util-linux? Do we ever need anything other than the bare libc mount(2)? > >> We need it for -o loop, but that is the exception to the rule. > > > > Well, I don't think that create a parallel universe is the best > > solution. > > > > I agree. However, xfstest uses 'mount' is a very particular way: > - User has to provide both block device and mount point > - xfstests already checks that those block device and mount point > are not in use in /proc/mounts *before* calling 'mount' The problem is it's not a mount-only issue, outputs from other util-linux commands and commands from coreutils have been changing over time too, we really can't "fork" & maintain a local wrapper all the changing pieces. As long as fstests still go with the 'match golden image' path, it's not likely to avoid such problems completely. OTOH, the output changes are not so frequent (such a change per 3-4 years maybe?), I think the maintain burden of filters is much less than maintaining local wrappers of such commands. And commands like mount are fs-related, it's better to get them some (indirect) test coverage in fstests too, as we test other fs-specific userspace tools like xfsprogs, e2fsprogs etc. Thanks, Eryu -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/common/filter b/common/filter index a212c09aa138..9c33efacea6c 100644 --- a/common/filter +++ b/common/filter @@ -397,11 +397,62 @@ _filter_ending_dot() sed -e "s/\.$//" } -# Older mount output referred to "block device" when mounting RO devices -# It's gone in newer versions +# Older mount output referred to "block device" when mounting RO devices. It's +# gone in newer versions. v2.30 changed the output again. This filter is to +# unify all read-only mount messages across all util-linux versions. +# +# for a successful ro mount: +# ancient: mount: block device <device> is write-protected, mounting read-only +# prior to v2.30: mount: <device> is write-protected, mounting read-only +# v2.30 and later: mount: <mountpoint>: WARNING: device write-protected, mounted read-only. +# +# a failed ro mount: +# ancient (two-line message): +# mount: block device <device> is write-protected, mounting read-only +# mount: cannot mount block device <device> read-only +# prior to v2.30 (two-line message): +# mount: <device> is write-protected, mounting read-only +# mount: cannot mount <device> read-only +# v2.30 and later (single-line message): +# mount: <mountpoint>: cannot mount <device> read-only. +# +# a failed rw remount: +# ancient: mount: cannot remount block device <device> read-write, is write-protected +# prior to v2.30: mount: cannot remount <device> read-write, is write-protected +# v2.30 and later: mount: <mountpoint>: cannot remount <device> read-write, is write-protected. +# +# Now use _filter_ro_mount to unify all these differences across old & new +# util-linux versions. So the filtered format would be: +# +# successful ro mount: +# mount: device write-protected, mounting read-only +# +# failed ro mount: +# mount: device write-protected, mounting read-only +# mount: cannot mount device read-only +# +# failed rw remount: +# mount: cannot remount device read-write, is write-protected _filter_ro_mount() { - sed -e "s/mount: block device/mount:/g" \ - -e "s/mount: cannot mount block device/mount: cannot mount/g" + perl -ne ' + if (/write-protected, mount.*read-only/) { + # filter successful ro mount, and first line of prior to v2.30 + # format failed ro mount + print "mount: device write-protected, mounting read-only\n"; + } elsif (/mount: .*: cannot mount.*read-only/) { + # filter v2.30 format failed ro mount, convert single-line + # message to two-line message + print "mount: device write-protected, mounting read-only\n"; + print "mount: cannot mount device read-only\n"; + } elsif (/^mount: cannot mount .* read-only$/) { + # filter prior to v2.30 format failed ro mount + print "mount: cannot mount device read-only\n"; + } elsif (/mount:.* cannot remount .* read-write.*/) { + # filter failed rw remount + print "mount: cannot remount device read-write, is write-protected\n"; + } else { + print "$_"; + }' | _filter_ending_dot } # Filter a failed mount output due to EUCLEAN and USTALE, util-linux changed diff --git a/tests/generic/050 b/tests/generic/050 index 5fa28a7648e5..efa45f04825b 100755 --- a/tests/generic/050 +++ b/tests/generic/050 @@ -60,7 +60,7 @@ blockdev --setro $SCRATCH_DEV # Mount it, and make sure we can't write to it, and we can unmount it again # echo "mounting read-only block device:" -_scratch_mount 2>&1 | _filter_scratch | _filter_ro_mount +_scratch_mount 2>&1 | _filter_ro_mount echo "touching file on read-only filesystem (should fail)" touch $SCRATCH_MNT/foo 2>&1 | _filter_scratch @@ -95,10 +95,10 @@ blockdev --setro $SCRATCH_DEV # -o norecovery is used. # echo "mounting filesystem that needs recovery on a read-only device:" -_scratch_mount 2>&1 | _filter_scratch | _filter_ro_mount +_scratch_mount 2>&1 | _filter_ro_mount echo "unmounting read-only filesystem" -_scratch_unmount 2>&1 | _filter_scratch +_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot # # This is the way out if the underlying device really is read-only. @@ -106,7 +106,7 @@ _scratch_unmount 2>&1 | _filter_scratch # data recovery hack. # echo "mounting filesystem with -o norecovery on a read-only device:" -_scratch_mount -o norecovery 2>&1 | _filter_scratch | _filter_ro_mount +_scratch_mount -o norecovery 2>&1 | _filter_ro_mount echo "unmounting read-only filesystem" _scratch_unmount 2>&1 | _filter_scratch diff --git a/tests/generic/050.out b/tests/generic/050.out index fb90f6ea5819..7d70ddee83cb 100644 --- a/tests/generic/050.out +++ b/tests/generic/050.out @@ -1,7 +1,7 @@ QA output created by 050 setting device read-only mounting read-only block device: -mount: SCRATCH_DEV is write-protected, mounting read-only +mount: device write-protected, mounting read-only touching file on read-only filesystem (should fail) touch: cannot touch 'SCRATCH_MNT/foo': Read-only file system unmounting read-only filesystem @@ -12,12 +12,12 @@ going down: unmounting shutdown filesystem: setting device read-only mounting filesystem that needs recovery on a read-only device: -mount: SCRATCH_DEV is write-protected, mounting read-only -mount: cannot mount SCRATCH_DEV read-only +mount: device write-protected, mounting read-only +mount: cannot mount device read-only unmounting read-only filesystem umount: SCRATCH_DEV: not mounted mounting filesystem with -o norecovery on a read-only device: -mount: SCRATCH_DEV is write-protected, mounting read-only +mount: device write-protected, mounting read-only unmounting read-only filesystem setting device read-write mounting filesystem that needs recovery with -o ro: diff --git a/tests/overlay/035 b/tests/overlay/035 index 64fcd708105e..05447741a1ba 100755 --- a/tests/overlay/035 +++ b/tests/overlay/035 @@ -69,7 +69,7 @@ mkdir -p $lowerdir1 $lowerdir2 $upperdir $workdir $MOUNT_PROG -t overlay -o"lowerdir=$lowerdir2:$lowerdir1" \ $OVL_BASE_SCRATCH_MNT $SCRATCH_MNT touch $SCRATCH_MNT/foo 2>&1 | _filter_scratch -_scratch_remount rw 2>&1 | _filter_scratch +_scratch_remount rw 2>&1 | _filter_ro_mount $UMOUNT_PROG $SCRATCH_MNT # Make workdir immutable to prevent workdir re-create on mount @@ -79,7 +79,7 @@ $CHATTR_PROG +i $workdir # Verify that overlay is mounted read-only and that it cannot be remounted rw. _overlay_scratch_mount_dirs $lowerdir2 $upperdir $workdir touch $SCRATCH_MNT/bar 2>&1 | _filter_scratch -_scratch_remount rw 2>&1 | _filter_scratch +_scratch_remount rw 2>&1 | _filter_ro_mount # success, all done status=0 diff --git a/tests/overlay/035.out b/tests/overlay/035.out index 5a5f67771402..e08ba2ebc691 100644 --- a/tests/overlay/035.out +++ b/tests/overlay/035.out @@ -1,5 +1,5 @@ QA output created by 035 touch: cannot touch 'SCRATCH_MNT/foo': Read-only file system -mount: cannot remount SCRATCH_DEV read-write, is write-protected +mount: cannot remount device read-write, is write-protected touch: cannot touch 'SCRATCH_MNT/bar': Read-only file system -mount: cannot remount SCRATCH_DEV read-write, is write-protected +mount: cannot remount device read-write, is write-protected
util-linux commit 6dede2f2f7c5 ("libmount: support MS_RDONLY on write-protected devices") changed the error message on read-only block device, and in the failure case printed one line message instead of two (for details please see comments in common/filter), and this change broke generic/050 and overlay/035. Fix it by adding more filter rules to _filter_ro_mount and updating associated .out files to unify the output from both old and new util-linux versions. Signed-off-by: Eryu Guan <eguan@redhat.com> --- v4 (also passed with both old & new util-linux): - add more comments to explain the output differences between util-linux versions - print out message directly instead of using perl variables - add word "device" to failed ro mount message v3: - document the filtered format in comments - remove legacy sed filter, the perl filter covers the legacy case well - filter out $SCRATCH_DEV/MNT too and use a consistent output - remove the new filter_mount helper in overlay/035 common/filter | 59 +++++++++++++++++++++++++++++++++++++++++++++++---- tests/generic/050 | 8 +++---- tests/generic/050.out | 8 +++---- tests/overlay/035 | 4 ++-- tests/overlay/035.out | 4 ++-- 5 files changed, 67 insertions(+), 16 deletions(-)