Btrfs devel - Dec 2011 - [PATCH] xfstests: check if qa user can execute commands

_require_user only checks if the qa user exists in /etc/passwd. It may exist
though still not be able to execute anything (e.g. with a /bin/false shell).

_user_do cannot determine failure to execute the given command, as it uses
_user_do_filter for post-processing which succeeds even if the command
fails. Thus, the check should be performed by _require_user.

Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
---
 common.rc |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/common.rc b/common.rc
index cab0b64..f540a79 100644
--- a/common.rc
+++ b/common.rc
@@ -897,6 +897,8 @@ _require_user()
     qa_user=fsgqa
     _cat_passwd | grep -q $qa_user
     [ "$?" == "0" ] || _notrun "$qa_user user not
defined."
+    echo /bin/true | su $qa_user
+    [ "$?" == "0" ] || _notrun "$qa_user cannot
execute commands."
 }
 
 _filter_user_do()
-- 
1.7.2.2

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Tue, Dec 06, 2011 at 10:56:29AM +0100, Jan Schmidt
wrote:
> _require_user only checks if the qa user exists in /etc/passwd. It may
exist
> though still not be able to execute anything (e.g. with a /bin/false
shell).
> 
> _user_do cannot determine failure to execute the given command, as it uses
> _user_do_filter for post-processing which succeeds even if the command
> fails. Thus, the check should be performed by _require_user.
Looks good.  Wouldn''t

	su $qa_user -c /bin/true

be the more normal way to write that test?

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On 06.12.2011 16:13, Christoph Hellwig wrote:
> On Tue, Dec 06, 2011 at 10:56:29AM +0100, Jan Schmidt wrote:
>> _require_user only checks if the qa user exists in /etc/passwd. It may
exist
>> though still not be able to execute anything (e.g. with a /bin/false
shell).
>>
>> _user_do cannot determine failure to execute the given command, as it
uses
>> _user_do_filter for post-processing which succeeds even if the command
>> fails. Thus, the check should be performed by _require_user.
> 
> Looks good.  Wouldn''t
> 
> 	su $qa_user -c /bin/true
> 
> be the more normal way to write that test?
I agree. I saw this and thought I''d adapt the (sometimes strange)
predominant bash programming style:

 921 _user_do()
 922 {
 923     if [ "$HOSTOS" == "IRIX" ]
 924         then
 925         echo $1 | /bin/bash "su $qa_user 2>&1" |
_filter_user_do
 926     else
 927         echo $1 | su $qa_user 2>&1 | _filter_user_do
 928     fi
 929 }

-Jan
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Tue, Dec 06, 2011 at 04:16:45PM +0100, Jan Schmidt
wrote:
> I agree. I saw this and thought I''d adapt the (sometimes strange)
> predominant bash programming style:
Maybe the IRIX su command didn''t support -c.  I''ll put your
patch in
as-is.

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html