* xfstest 318 @ 2013-10-24 15:44 Christoph Hellwig 2013-10-24 17:05 ` Dwight Engen 2013-10-24 20:11 ` [PATCH] xfstests: generic/318 use symbolic namespaced ids Dwight Engen 0 siblings, 2 replies; 7+ messages in thread From: Christoph Hellwig @ 2013-10-24 15:44 UTC (permalink / raw) To: Dwight Engen; +Cc: xfs This test has always been failing for me on my Debian stable system with the output below, any idea why this could be? --- tests/generic/318.out 2013-09-09 14:41:08.000000000 +0000 +++ /root/xfstests/results//generic/318.out.bad 2013-10-24 15:44:01.000000000 +0000 @@ -17,7 +17,7 @@ # group: 65534 user::rw- user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:3:rw- #effective:r-- group::r-- group:0:--x #effective:--- mask::r-- @@ -43,7 +43,7 @@ # group: 65534 user::rw- user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:3:rw- #effective:r-- group::r-- group:0:--x #effective:--- mask::r-- _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: xfstest 318 2013-10-24 15:44 xfstest 318 Christoph Hellwig @ 2013-10-24 17:05 ` Dwight Engen 2013-10-24 20:11 ` [PATCH] xfstests: generic/318 use symbolic namespaced ids Dwight Engen 1 sibling, 0 replies; 7+ messages in thread From: Dwight Engen @ 2013-10-24 17:05 UTC (permalink / raw) To: Christoph Hellwig; +Cc: xfs On Thu, 24 Oct 2013 08:44:14 -0700 Christoph Hellwig <hch@infradead.org> wrote: > This test has always been failing for me on my Debian stable system > with the output below, any idea why this could be? Looks like the | getfacl_filter_id isn't doing any good on the getfacl that is being run in the userns because the uids don't match $acl[12] (which is the point), and the uids that _acl_setup_ids picked are not the same distance apart as they were on my system. So on your system $acl2 is not $acl1 + 1. Let me see about running the second getfacl through a filter that knows how to replace it based on the delta between $acl1 and $acl2, I think that should always work. > --- tests/generic/318.out 2013-09-09 14:41:08.000000000 +0000 > +++ /root/xfstests/results//generic/318.out.bad 2013-10-24 > 15:44:01.000000000 +0000 @@ -17,7 +17,7 @@ > # group: 65534 > user::rw- > user:0:r-x #effective:r-- > -user:1:rw- #effective:r-- > +user:3:rw- #effective:r-- > group::r-- > group:0:--x #effective:--- > mask::r-- > @@ -43,7 +43,7 @@ > # group: 65534 > user::rw- > user:0:r-x #effective:r-- > -user:1:rw- #effective:r-- > +user:3:rw- #effective:r-- > group::r-- > group:0:--x #effective:--- > mask::r-- _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH] xfstests: generic/318 use symbolic namespaced ids 2013-10-24 15:44 xfstest 318 Christoph Hellwig 2013-10-24 17:05 ` Dwight Engen @ 2013-10-24 20:11 ` Dwight Engen 2013-10-25 9:31 ` Christoph Hellwig 2013-11-04 20:25 ` Rich Johnston 1 sibling, 2 replies; 7+ messages in thread From: Dwight Engen @ 2013-10-24 20:11 UTC (permalink / raw) To: Christoph Hellwig; +Cc: xfs Christoph, I think the following should fix 318 for you. -- - use namespace ids that correspond to the initial acl ids and match them symbolically in the output - also ensure that all uids have a mapping in the namespace so we don't get the unmapped uid (65534) in the output Signed-off-by: Dwight Engen <dwight.engen@oracle.com> --- tests/generic/318 | 21 +++++++++++++++++++-- tests/generic/318.out | 20 ++++++++++---------- 2 files changed, 29 insertions(+), 12 deletions(-) diff --git a/tests/generic/318 b/tests/generic/318 index d3bce51..810cff4 100755 --- a/tests/generic/318 +++ b/tests/generic/318 @@ -63,6 +63,23 @@ _require_scratch _need_to_be_root _acl_setup_ids _require_acls +ns_acl1=0 +ns_acl2=`expr $acl2 - $acl1` +ns_acl3=`expr $acl3 - $acl1` + +_getfacl_filter_nsid() +{ + sed \ + -e "s/user:$ns_acl1/user:nsid1/" \ + -e "s/user:$ns_acl2/user:nsid2/" \ + -e "s/user:$ns_acl3/user:nsid3/" \ + -e "s/group:$ns_acl1/group:nsid1/" \ + -e "s/group:$ns_acl2/group:nsid2/" \ + -e "s/group:$ns_acl3/group:nsid3/" \ + -e "s/: $ns_acl1/: nsid1/" \ + -e "s/: $ns_acl2/: nsid2/" \ + -e "s/: $ns_acl3/: nsid3/" +} _print_getfacls() { @@ -70,7 +87,7 @@ _print_getfacls() getfacl --absolute-names -n $file 2>/dev/null | _filter_scratch | _getfacl_filter_id echo "From user_ns" - $nsexec -U -M "0 $acl1 1000" -G "0 $acl2 1000" getfacl --absolute-names -n $file 2>/dev/null | _filter_scratch | _getfacl_filter_id + $nsexec -U -M "0 $acl1 1000" -G "0 $acl1 1000" getfacl --absolute-names -n $file 2>/dev/null | _filter_scratch | _getfacl_filter_nsid } umount $SCRATCH_DEV >/dev/null 2>&1 @@ -85,7 +102,7 @@ chown $acl1.$acl1 $file # set acls from init_user_ns, to be checked from inside the userns setfacl -n -m u:$acl2:rw,g:$acl2:r $file # set acls from inside userns, to be checked from init_user_ns -$nsexec -s -U -M "0 $acl1 1000" -G "0 $acl2 1000" setfacl -n -m u:root:rx,g:root:x $file +$nsexec -s -U -M "0 $acl1 1000" -G "0 $acl1 1000" setfacl -n -m u:root:rx,g:$ns_acl2:x $file _print_getfacls diff --git a/tests/generic/318.out b/tests/generic/318.out index e2b42a4..6ff2bf8 100644 --- a/tests/generic/318.out +++ b/tests/generic/318.out @@ -13,13 +13,13 @@ other::r-- From user_ns # file: SCRATCH_MNT/file1 -# owner: 0 -# group: 65534 +# owner: nsid1 +# group: nsid1 user::rw- -user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:nsid1:r-x #effective:r-- +user:nsid2:rw- #effective:r-- group::r-- -group:0:--x #effective:--- +group:nsid2:--x #effective:--- mask::r-- other::r-- @@ -39,13 +39,13 @@ other::r-- From user_ns # file: SCRATCH_MNT/file1 -# owner: 0 -# group: 65534 +# owner: nsid1 +# group: nsid1 user::rw- -user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:nsid1:r-x #effective:r-- +user:nsid2:rw- #effective:r-- group::r-- -group:0:--x #effective:--- +group:nsid2:--x #effective:--- mask::r-- other::r-- -- 1.7.1 _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] xfstests: generic/318 use symbolic namespaced ids 2013-10-24 20:11 ` [PATCH] xfstests: generic/318 use symbolic namespaced ids Dwight Engen @ 2013-10-25 9:31 ` Christoph Hellwig 2013-11-04 20:11 ` Rich Johnston 2013-11-04 20:25 ` Rich Johnston 1 sibling, 1 reply; 7+ messages in thread From: Christoph Hellwig @ 2013-10-25 9:31 UTC (permalink / raw) To: Dwight Engen; +Cc: xfs On Thu, Oct 24, 2013 at 04:11:55PM -0400, Dwight Engen wrote: > Christoph, I think the following should fix 318 for you. This fixes the test for me, thanks. Reviewed-by: Christoph Hellwig <hch@lst.de> _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] xfstests: generic/318 use symbolic namespaced ids 2013-10-25 9:31 ` Christoph Hellwig @ 2013-11-04 20:11 ` Rich Johnston 2013-11-04 20:21 ` Rich Johnston 0 siblings, 1 reply; 7+ messages in thread From: Rich Johnston @ 2013-11-04 20:11 UTC (permalink / raw) To: Christoph Hellwig, Dwight Engen; +Cc: xfs On 10/25/2013 04:31 AM, Christoph Hellwig wrote: > On Thu, Oct 24, 2013 at 04:11:55PM -0400, Dwight Engen wrote: >> Christoph, I think the following should fix 318 for you. > > This fixes the test for me, thanks. > > Reviewed-by: Christoph Hellwig <hch@lst.de> > > _______________________________________________ > xfs mailing list > xfs@oss.sgi.com > http://oss.sgi.com/mailman/listinfo/xfs > Hey Dwight, This still fails for me. --Rich diff -u tests/generic/318.out /usr/src/tot/xfstests/results//generic/318.out.bad --- tests/generic/318.out 2013-11-04 13:03:03.000000000 -0600 +++ /usr/src/tot/xfstests/results//generic/318.out.bad 2013-11-04 14:08:32.000000000 -0600 @@ -17,7 +17,7 @@ # group: 65534 user::rw- user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:12:rw- #effective:r-- group::r-- group:0:--x #effective:--- mask::r-- @@ -43,7 +43,7 @@ # group: 65534 user::rw- user:0:r-x #effective:r-- -user:1:rw- #effective:r-- +user:12:rw- #effective:r-- group::r-- group:0:--x #effective:--- mask::r-- _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] xfstests: generic/318 use symbolic namespaced ids 2013-11-04 20:11 ` Rich Johnston @ 2013-11-04 20:21 ` Rich Johnston 0 siblings, 0 replies; 7+ messages in thread From: Rich Johnston @ 2013-11-04 20:21 UTC (permalink / raw) To: Christoph Hellwig, Dwight Engen; +Cc: xfs On 11/04/2013 02:11 PM, Rich Johnston wrote: >> > Hey Dwight, > > This still fails for me. > > --Rich > Unsend sorry yes it does fix it, sorry for the noise. _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] xfstests: generic/318 use symbolic namespaced ids 2013-10-24 20:11 ` [PATCH] xfstests: generic/318 use symbolic namespaced ids Dwight Engen 2013-10-25 9:31 ` Christoph Hellwig @ 2013-11-04 20:25 ` Rich Johnston 1 sibling, 0 replies; 7+ messages in thread From: Rich Johnston @ 2013-11-04 20:25 UTC (permalink / raw) To: Dwight Engen; +Cc: xfs This has been committed. Thanks --Rich commit c4d4fb21ef3b08e530512071e798bffeea6f2c74 Author: Dwight Engen <dwight.engen@oracle.com> Date: Thu Oct 24 20:11:55 2013 +0000 xfstests: generic/318 use symbolic namespaced ids _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2013-11-04 20:25 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2013-10-24 15:44 xfstest 318 Christoph Hellwig 2013-10-24 17:05 ` Dwight Engen 2013-10-24 20:11 ` [PATCH] xfstests: generic/318 use symbolic namespaced ids Dwight Engen 2013-10-25 9:31 ` Christoph Hellwig 2013-11-04 20:11 ` Rich Johnston 2013-11-04 20:21 ` Rich Johnston 2013-11-04 20:25 ` Rich Johnston
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox