Message ID | 20210420075351.16059-1-liwang@redhat.com |
---|---|
State | Accepted |
Headers | show |
Series | mkfs: print more info for debugging | expand |
> We can NOT reproduce the problem by manual with both ppc64le and s390x, > so let's print more useful info from test when getting fail. Reviewed-by: Petr Vorel <pvorel@suse.cz> LGTM, just some remarks below > mkfs01 1 TPASS: 'mkfs -t ext4 /dev/loop0 ' passed. > mkfs01 2 TFAIL: 'mkfs -t ext4 /dev/loop0 16000' failed, not expected. > mkfs01 3 TPASS: 'mkfs -t ext4 -c /dev/loop0 ' passed. > mkfs01 4 TPASS: 'mkfs -V ' passed. > mkfs01 5 TPASS: 'mkfs -h ' passed. > Signed-off-by: Li Wang <liwang@redhat.com> > --- > testcases/commands/mkfs/mkfs01.sh | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > diff --git a/testcases/commands/mkfs/mkfs01.sh b/testcases/commands/mkfs/mkfs01.sh > index 90368190d..55296c8df 100755 > --- a/testcases/commands/mkfs/mkfs01.sh > +++ b/testcases/commands/mkfs/mkfs01.sh > @@ -128,7 +128,8 @@ mkfs_test() > if [ -n "$device" ]; then > mkfs_verify_type "$fs_type" "$device" > if [ $? -ne 0 ]; then > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected type." Maybe "unexpected type"? I'd also change ${mkfs_cmd} to $mkfs_cmd (more readable). > + cat temp > return > fi > fi > @@ -136,7 +137,8 @@ mkfs_test() > if [ -n "$size" ]; then > mkfs_verify_size "$fs_type" "$size" > if [ $? -ne 0 ]; then > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected size." The same here. Kind regards, Petr > + cat temp > return > fi > fi
> > --- a/testcases/commands/mkfs/mkfs01.sh > > +++ b/testcases/commands/mkfs/mkfs01.sh > > @@ -128,7 +128,8 @@ mkfs_test() > > if [ -n "$device" ]; then > > mkfs_verify_type "$fs_type" "$device" > > if [ $? -ne 0 ]; then > > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected type." > Maybe "unexpected type"? > > I'd also change ${mkfs_cmd} to $mkfs_cmd (more readable). Sounds good, I modified it with your suggestions and pushed it.
----- Original Message ----- > We can NOT reproduce the problem by manual with both ppc64le and s390x, > so let's print more useful info from test when getting fail. It's mkfs_verify_size, test expects to find 90% blocks (of device size), but lately we miss that by ~1-2%: # mkfs.ext3 /dev/loop0 16000 mke2fs 1.46.2 (28-Feb-2021) /dev/loop0 contains a ext3 file system created on Fri Apr 16 05:09:23 2021 Proceed anyway? (y,N) y Discarding device blocks: done Creating filesystem with 16000 1k blocks and 4000 inodes Filesystem UUID: d66c4c37-13ce-41df-b3c4-345ec1641e1f Superblock backups stored on blocks: 8193 Allocating group tables: done Writing inode tables: done Creating journal (1024 blocks): done Writing superblocks and filesystem accounting information: done mount /dev/loop0 /mnt/test # df -T Filesystem Type 1K-blocks Used Available Use% Mounted on /dev/loop0 ext3 14343 19 13524 1% /mnt/test > > mkfs01 1 TPASS: 'mkfs -t ext4 /dev/loop0 ' passed. > mkfs01 2 TFAIL: 'mkfs -t ext4 /dev/loop0 16000' failed, not expected. > mkfs01 3 TPASS: 'mkfs -t ext4 -c /dev/loop0 ' passed. > mkfs01 4 TPASS: 'mkfs -V ' passed. > mkfs01 5 TPASS: 'mkfs -h ' passed. > > Signed-off-by: Li Wang <liwang@redhat.com> > --- > testcases/commands/mkfs/mkfs01.sh | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/testcases/commands/mkfs/mkfs01.sh > b/testcases/commands/mkfs/mkfs01.sh > index 90368190d..55296c8df 100755 > --- a/testcases/commands/mkfs/mkfs01.sh > +++ b/testcases/commands/mkfs/mkfs01.sh > @@ -128,7 +128,8 @@ mkfs_test() > if [ -n "$device" ]; then > mkfs_verify_type "$fs_type" "$device" > if [ $? -ne 0 ]; then > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected type." > + cat temp > return > fi > fi > @@ -136,7 +137,8 @@ mkfs_test() > if [ -n "$size" ]; then > mkfs_verify_size "$fs_type" "$size" > if [ $? -ne 0 ]; then > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected size." > + cat temp > return > fi > fi > -- > 2.30.2 > > > -- > Mailing list info: https://lists.linux.it/listinfo/ltp > >
On Tue, Apr 20, 2021 at 8:17 PM Jan Stancek <jstancek@redhat.com> wrote: > > > > ----- Original Message ----- > > We can NOT reproduce the problem by manual with both ppc64le and s390x, > > so let's print more useful info from test when getting fail. > > It's mkfs_verify_size, test expects to find 90% blocks (of device size), > but lately we miss that by ~1-2%: > > # mkfs.ext3 /dev/loop0 16000 > mke2fs 1.46.2 (28-Feb-2021) Thanks for the info. How reproducible of this issue, and which arch/platform you used? Or, can you reproduce it with a downgrade e2fsprogs version? > /dev/loop0 contains a ext3 file system > created on Fri Apr 16 05:09:23 2021 > Proceed anyway? (y,N) y > Discarding device blocks: done > Creating filesystem with 16000 1k blocks and 4000 inodes > Filesystem UUID: d66c4c37-13ce-41df-b3c4-345ec1641e1f > Superblock backups stored on blocks: > 8193 > > Allocating group tables: done > Writing inode tables: done > Creating journal (1024 blocks): done > Writing superblocks and filesystem accounting information: done > > mount /dev/loop0 /mnt/test > # df -T > Filesystem Type 1K-blocks Used Available Use% Mounted on > /dev/loop0 ext3 14343 19 13524 1% /mnt/test > > > > > > mkfs01 1 TPASS: 'mkfs -t ext4 /dev/loop0 ' passed. > > mkfs01 2 TFAIL: 'mkfs -t ext4 /dev/loop0 16000' failed, not expected. > > mkfs01 3 TPASS: 'mkfs -t ext4 -c /dev/loop0 ' passed. > > mkfs01 4 TPASS: 'mkfs -V ' passed. > > mkfs01 5 TPASS: 'mkfs -h ' passed. > > > > Signed-off-by: Li Wang <liwang@redhat.com> > > --- > > testcases/commands/mkfs/mkfs01.sh | 6 ++++-- > > 1 file changed, 4 insertions(+), 2 deletions(-) > > > > diff --git a/testcases/commands/mkfs/mkfs01.sh > > b/testcases/commands/mkfs/mkfs01.sh > > index 90368190d..55296c8df 100755 > > --- a/testcases/commands/mkfs/mkfs01.sh > > +++ b/testcases/commands/mkfs/mkfs01.sh > > @@ -128,7 +128,8 @@ mkfs_test() > > if [ -n "$device" ]; then > > mkfs_verify_type "$fs_type" "$device" > > if [ $? -ne 0 ]; then > > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected type." > > + cat temp > > return > > fi > > fi > > @@ -136,7 +137,8 @@ mkfs_test() > > if [ -n "$size" ]; then > > mkfs_verify_size "$fs_type" "$size" > > if [ $? -ne 0 ]; then > > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected size." > > + cat temp > > return > > fi > > fi > > -- > > 2.30.2 > > > > > > -- > > Mailing list info: https://lists.linux.it/listinfo/ltp > > > > >
----- Original Message ----- > On Tue, Apr 20, 2021 at 8:17 PM Jan Stancek <jstancek@redhat.com> wrote: > > > > > > > > ----- Original Message ----- > > > We can NOT reproduce the problem by manual with both ppc64le and s390x, > > > so let's print more useful info from test when getting fail. > > > > It's mkfs_verify_size, test expects to find 90% blocks (of device size), > > but lately we miss that by ~1-2%: > > > > # mkfs.ext3 /dev/loop0 16000 > > mke2fs 1.46.2 (28-Feb-2021) > > Thanks for the info. > > How reproducible of this issue, and which arch/platform you used? It seemed pretty consistent with Fedora-Rawhide-20210415.n.0 on power9 ppc64le (9006-22P, J:5279637) > Or, can you reproduce it with a downgrade e2fsprogs version? Not immediately, I lost reservation of the machine. > > > > /dev/loop0 contains a ext3 file system > > created on Fri Apr 16 05:09:23 2021 > > Proceed anyway? (y,N) y > > Discarding device blocks: done > > Creating filesystem with 16000 1k blocks and 4000 inodes > > Filesystem UUID: d66c4c37-13ce-41df-b3c4-345ec1641e1f > > Superblock backups stored on blocks: > > 8193 > > > > Allocating group tables: done > > Writing inode tables: done > > Creating journal (1024 blocks): done > > Writing superblocks and filesystem accounting information: done > > > > mount /dev/loop0 /mnt/test > > # df -T > > Filesystem Type 1K-blocks Used > > Available Use% Mounted on > > /dev/loop0 ext3 14343 19 > > 13524 1% /mnt/test > > > > > > > > > > mkfs01 1 TPASS: 'mkfs -t ext4 /dev/loop0 ' passed. > > > mkfs01 2 TFAIL: 'mkfs -t ext4 /dev/loop0 16000' failed, not expected. > > > mkfs01 3 TPASS: 'mkfs -t ext4 -c /dev/loop0 ' passed. > > > mkfs01 4 TPASS: 'mkfs -V ' passed. > > > mkfs01 5 TPASS: 'mkfs -h ' passed. > > > > > > Signed-off-by: Li Wang <liwang@redhat.com> > > > --- > > > testcases/commands/mkfs/mkfs01.sh | 6 ++++-- > > > 1 file changed, 4 insertions(+), 2 deletions(-) > > > > > > diff --git a/testcases/commands/mkfs/mkfs01.sh > > > b/testcases/commands/mkfs/mkfs01.sh > > > index 90368190d..55296c8df 100755 > > > --- a/testcases/commands/mkfs/mkfs01.sh > > > +++ b/testcases/commands/mkfs/mkfs01.sh > > > @@ -128,7 +128,8 @@ mkfs_test() > > > if [ -n "$device" ]; then > > > mkfs_verify_type "$fs_type" "$device" > > > if [ $? -ne 0 ]; then > > > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > > > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected > > > type." > > > + cat temp > > > return > > > fi > > > fi > > > @@ -136,7 +137,8 @@ mkfs_test() > > > if [ -n "$size" ]; then > > > mkfs_verify_size "$fs_type" "$size" > > > if [ $? -ne 0 ]; then > > > - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." > > > + tst_res TFAIL "'${mkfs_cmd}' failed, not expected > > > size." > > > + cat temp > > > return > > > fi > > > fi > > > -- > > > 2.30.2 > > > > > > > > > -- > > > Mailing list info: https://lists.linux.it/listinfo/ltp > > > > > > > > > > > -- > Regards, > Li Wang > >
----- Original Message ----- > On Tue, Apr 20, 2021 at 8:17 PM Jan Stancek <jstancek@redhat.com> wrote: > > > > > > > > ----- Original Message ----- > > > We can NOT reproduce the problem by manual with both ppc64le and s390x, > > > so let's print more useful info from test when getting fail. > > > > It's mkfs_verify_size, test expects to find 90% blocks (of device size), > > but lately we miss that by ~1-2%: > > > > # mkfs.ext3 /dev/loop0 16000 > > mke2fs 1.46.2 (28-Feb-2021) > > Thanks for the info. > > How reproducible of this issue, and which arch/platform you used? > Or, can you reproduce it with a downgrade e2fsprogs version? It starts with this commit: https://github.com/tytso/e2fsprogs/commit/59037c5357d39c6d0f14a0aff70e67dc13eafc84
diff --git a/testcases/commands/mkfs/mkfs01.sh b/testcases/commands/mkfs/mkfs01.sh index 90368190d..55296c8df 100755 --- a/testcases/commands/mkfs/mkfs01.sh +++ b/testcases/commands/mkfs/mkfs01.sh @@ -128,7 +128,8 @@ mkfs_test() if [ -n "$device" ]; then mkfs_verify_type "$fs_type" "$device" if [ $? -ne 0 ]; then - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." + tst_res TFAIL "'${mkfs_cmd}' failed, not expected type." + cat temp return fi fi @@ -136,7 +137,8 @@ mkfs_test() if [ -n "$size" ]; then mkfs_verify_size "$fs_type" "$size" if [ $? -ne 0 ]; then - tst_res TFAIL "'${mkfs_cmd}' failed, not expected." + tst_res TFAIL "'${mkfs_cmd}' failed, not expected size." + cat temp return fi fi
We can NOT reproduce the problem by manual with both ppc64le and s390x, so let's print more useful info from test when getting fail. mkfs01 1 TPASS: 'mkfs -t ext4 /dev/loop0 ' passed. mkfs01 2 TFAIL: 'mkfs -t ext4 /dev/loop0 16000' failed, not expected. mkfs01 3 TPASS: 'mkfs -t ext4 -c /dev/loop0 ' passed. mkfs01 4 TPASS: 'mkfs -V ' passed. mkfs01 5 TPASS: 'mkfs -h ' passed. Signed-off-by: Li Wang <liwang@redhat.com> --- testcases/commands/mkfs/mkfs01.sh | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)