You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lenya.apache.org by bu...@apache.org on 2006/09/26 17:36:39 UTC

DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG�
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=40609>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND�
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=40609


thorsten@apache.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID




------- Additional Comments From thorsten@apache.org  2006-09-26 15:36 -------
If you switch the rights then it will work again. 

I do not consider this a bug, but a feature. Proofs that the ac is working in
locking out the ones that you define. 

If you did not got locked out then we need to worry. 

-- 
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Thorsten Scherler <th...@wyona.com>.
On Tue, 2006-09-26 at 17:50 +0200, Jörn Nettingsmeier wrote:
> before we start a REOPEN war on bugzilla, let's discuss this here first.
> 
> imho an exception caused by a normal gui operation is always a bug.
> imnsho a "deny" credential for the role "edit" that leads to a loss of
> "visit" for the user admin is a blocker bug.
> 
> bugzilla@apache.org wrote:
> > ------- Additional Comments From thorsten@apache.org  2006-09-26
> > 15:36 ------- If you switch the rights then it will work again.
> 
> please elaborate.

http://lenya.apache.org/docs/1_4/reference/ac.html especially (it is
nearly the same use case as in the bug)
http://lenya.apache.org/docs/1_4/reference/ac.html#samples

> 
> > I do not consider this a bug, but a feature. Proofs that the ac is
> > working in locking out the ones that you define.
> > 
> > If you did not got locked out then we need to worry.
> 
> either you have totally misread my bug report or i'm totally at sea
> about how ac is supposed to work.
> 

Hmm, the only thing is that lenya may (or better should, looking at this
bug) warn you in the case that you are going to lock yourself out (like
described in your bug report and our documentation). However this would
be enhancement and not a bug since the ac is working like described in
the specification.

salu2
-- 
Thorsten Scherler
COO Spain
Wyona Inc.  -  Open Source Content Management  -  Apache Lenya
http://www.wyona.com                   http://lenya.apache.org
thorsten.scherler@wyona.com                thorsten@apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Jann Forrer <ja...@id.unizh.ch>.
Thorsten Scherler wrote:

[ ... ]

>>> the usecase permissions for tab.acLive are "admin/grant".
>>> lenya (as an admin) should be able to commit the second change, even
>>> after taking away the edit permission for one of its own group
>>> accreditables, as changing ac calls for the "admin" role (which was
>>> never touched). that seems like a big bug.
>>>
>> I reconstruct this ac behavior and i agree with joern. I really can't
>> find a reason why an admin user is not able to change ac entries (if for
>> the group edit the role edit is denied). The consequence of the first
>> transaction should only prevent the user lenya (if he is a member of the
>> group edit) from editing the content of a document.
> 
> Hmm, did you try to create another admin user (which is not in the
> editor group) and logged in? 
> 
> Try http://lenya.zones.apache.org:9999/default/authoring/concepts.html
> before the next build with user lenya. -> Access Denied
> 
> Now try with (user: admin, pass: 123456) -> Access Grant
>
> It looks out the user lenya which is as well member of the editor.
> http://lenya.apache.org/docs/1_4/reference/ac.html#Concept
> "The order of credentials at a node is important."
>

Ok i did not realize that the order of the credentials is important.
But does that make sense? If I deny editing for the group edit but grant
access for the admin group, I think it is more intuitive to assume that
an admin user could at least access and modify the node in the site area
 even as a member of the group edit.


> http://lenya.zones.apache.org:9999/default/authoring/links.html
> Here I changed the order allowing the admin group first and then deny
> the editors. Login with lenya now works like a charm.
> 
> ...and even if we use a slightly different use case it pins down to
> http://lenya.apache.org/docs/1_4/reference/ac.html#samples
> 
> If we want to change this see my bugzilla comment.
> 

I read the comment and basically I think we should do the change. But
frankly said i do not fully understand the consequences yet.


>>
>>> the user does not get a meaningful error message, and lenya throws an
>>> exception. that is perhaps a small bug, but it needs to be fixed.
>>>
>> As joern said I think no gui operation should throw an nasty java
>> exception. The user should be redirected to some page e.g. the login
>> page or getting a meaningful error message.
> 
> Like I stated in the bug, I agree to catch this exception before it
> happens asking the user if he wants continue regardless he will get
> looked out afterward. If he agrees then yes there should be a redirect
> instead of an exception.
> 
+1

jann

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Thorsten Scherler <th...@apache.org>.
On Thu, 2006-09-28 at 08:58 +0200, Jann Forrer wrote:
> Joern Nettingsmeier wrote:
> > Thorsten Scherler wrote:
> >> On Tue, 2006-09-26 at 17:50 +0200, Jörn Nettingsmeier wrote:
> >>> before we start a REOPEN war on bugzilla, let's discuss this here first.
> >>>
> >>> imho an exception caused by a normal gui operation is always a bug.
> >>> imnsho a "deny" credential for the role "edit" that leads to a loss of
> >>> "visit" for the user admin is a blocker bug.
> >> The thing is a user can have more then one role. Lenya e.g. has the role
> >> "edit, admin, visit", now we test whether we find credentials for this
> >> roles, if there is a credential (with high priority) which deny access
> >> for editor then lenya get locked out. To allow lenya and not any other
> >> editor have a look at our documentation. 
> > 
> > again:
> > 
> > log in as user lenya.
> > go to site overview, select page "document type examples".
> > open "ac authoring" tab.
> > as an inheritable credential, add { group:edit, role:edit, deny}.
> > press "add".
> > as an inheritable credential, add { group:admin, role:edit: grant}.
> > press "add". pooof.
> > 
> > 
> > the problems are these:
> > 
> > 
> > the usecase permissions for tab.acLive are "admin/grant".
> > lenya (as an admin) should be able to commit the second change, even
> > after taking away the edit permission for one of its own group
> > accreditables, as changing ac calls for the "admin" role (which was
> > never touched). that seems like a big bug.
> > 
> 
> I reconstruct this ac behavior and i agree with joern. I really can't
> find a reason why an admin user is not able to change ac entries (if for
> the group edit the role edit is denied). The consequence of the first
> transaction should only prevent the user lenya (if he is a member of the
> group edit) from editing the content of a document.

Hmm, did you try to create another admin user (which is not in the
editor group) and logged in? 

Try http://lenya.zones.apache.org:9999/default/authoring/concepts.html
before the next build with user lenya. -> Access Denied

Now try with (user: admin, pass: 123456) -> Access Grant

It looks out the user lenya which is as well member of the editor.
http://lenya.apache.org/docs/1_4/reference/ac.html#Concept
"The order of credentials at a node is important."

http://lenya.zones.apache.org:9999/default/authoring/links.html
Here I changed the order allowing the admin group first and then deny
the editors. Login with lenya now works like a charm.

...and even if we use a slightly different use case it pins down to
http://lenya.apache.org/docs/1_4/reference/ac.html#samples

If we want to change this see my bugzilla comment.

> 
> 
> > the user does not get a meaningful error message, and lenya throws an
> > exception. that is perhaps a small bug, but it needs to be fixed.
> > 
> 
> As joern said I think no gui operation should throw an nasty java
> exception. The user should be redirected to some page e.g. the login
> page or getting a meaningful error message.

Like I stated in the bug, I agree to catch this exception before it
happens asking the user if he wants continue regardless he will get
looked out afterward. If he agrees then yes there should be a redirect
instead of an exception.

salu2

> 
> 
> [...]
> 
> Jann
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
> For additional commands, e-mail: dev-help@lenya.apache.org
> 
-- 
thorsten

"Together we stand, divided we fall!" 
Hey you (Pink Floyd)


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Jann Forrer <ja...@id.unizh.ch>.
Joern Nettingsmeier wrote:
> Thorsten Scherler wrote:
>> On Tue, 2006-09-26 at 17:50 +0200, Jörn Nettingsmeier wrote:
>>> before we start a REOPEN war on bugzilla, let's discuss this here first.
>>>
>>> imho an exception caused by a normal gui operation is always a bug.
>>> imnsho a "deny" credential for the role "edit" that leads to a loss of
>>> "visit" for the user admin is a blocker bug.
>> The thing is a user can have more then one role. Lenya e.g. has the role
>> "edit, admin, visit", now we test whether we find credentials for this
>> roles, if there is a credential (with high priority) which deny access
>> for editor then lenya get locked out. To allow lenya and not any other
>> editor have a look at our documentation. 
> 
> again:
> 
> log in as user lenya.
> go to site overview, select page "document type examples".
> open "ac authoring" tab.
> as an inheritable credential, add { group:edit, role:edit, deny}.
> press "add".
> as an inheritable credential, add { group:admin, role:edit: grant}.
> press "add". pooof.
> 
> 
> the problems are these:
> 
> 
> the usecase permissions for tab.acLive are "admin/grant".
> lenya (as an admin) should be able to commit the second change, even
> after taking away the edit permission for one of its own group
> accreditables, as changing ac calls for the "admin" role (which was
> never touched). that seems like a big bug.
> 

I reconstruct this ac behavior and i agree with joern. I really can't
find a reason why an admin user is not able to change ac entries (if for
the group edit the role edit is denied). The consequence of the first
transaction should only prevent the user lenya (if he is a member of the
group edit) from editing the content of a document.


> the user does not get a meaningful error message, and lenya throws an
> exception. that is perhaps a small bug, but it needs to be fixed.
> 

As joern said I think no gui operation should throw an nasty java
exception. The user should be redirected to some page e.g. the login
page or getting a meaningful error message.


[...]

Jann

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Joern Nettingsmeier <po...@uni-due.de>.
Thorsten Scherler wrote:
> On Tue, 2006-09-26 at 17:50 +0200, Jörn Nettingsmeier wrote:
>> before we start a REOPEN war on bugzilla, let's discuss this here first.
>>
>> imho an exception caused by a normal gui operation is always a bug.
>> imnsho a "deny" credential for the role "edit" that leads to a loss of
>> "visit" for the user admin is a blocker bug.
> 
> The thing is a user can have more then one role. Lenya e.g. has the role
> "edit, admin, visit", now we test whether we find credentials for this
> roles, if there is a credential (with high priority) which deny access
> for editor then lenya get locked out. To allow lenya and not any other
> editor have a look at our documentation. 

again:

log in as user lenya.
go to site overview, select page "document type examples".
open "ac authoring" tab.
as an inheritable credential, add { group:edit, role:edit, deny}.
press "add".
as an inheritable credential, add { group:admin, role:edit: grant}.
press "add". pooof.


the problems are these:


the usecase permissions for tab.acLive are "admin/grant".
lenya (as an admin) should be able to commit the second change, even
after taking away the edit permission for one of its own group
accreditables, as changing ac calls for the "admin" role (which was
never touched). that seems like a big bug.

the user does not get a meaningful error message, and lenya throws an
exception. that is perhaps a small bug, but it needs to be fixed.

after the entire transaction (which only touched the *edit* role), user
lenya can no longer *view* the page. "world" should still be able to
view the page, and lenya is "world". plus the *view* role was never
touched as well in the entire test case. that looks like another mother
of a bug.

and even if this all is due to some misconception of mine (which might
well be the case), it would be a lot more polite to point out my errors
in understanding and give me a chance to close the bug than marking it
as invalid without giving it more that a passing glance. i may be alone
in this, but for me, ac is serious stuff. if i'm totally not grokking
it, that's at least a serious documentation issue.





-- 
"I don't need backups. I need restore!" - Trad.

--
Jörn Nettingsmeier, EDV-Administrator
Institut für Politikwissenschaft
Universität Duisburg-Essen, Standort Duisburg
Mail: pol-admin@uni-due.de, Telefon: 0203/379-2736


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Thorsten Scherler <th...@wyona.com>.
On Tue, 2006-09-26 at 17:50 +0200, Jörn Nettingsmeier wrote:
> before we start a REOPEN war on bugzilla, let's discuss this here first.
> 
> imho an exception caused by a normal gui operation is always a bug.
> imnsho a "deny" credential for the role "edit" that leads to a loss of
> "visit" for the user admin is a blocker bug.

The thing is a user can have more then one role. Lenya e.g. has the role
"edit, admin, visit", now we test whether we find credentials for this
roles, if there is a credential (with high priority) which deny access
for editor then lenya get locked out. To allow lenya and not any other
editor have a look at our documentation. 

salu2

> 
> bugzilla@apache.org wrote:
> > ------- Additional Comments From thorsten@apache.org  2006-09-26
> > 15:36 ------- If you switch the rights then it will work again.
> 
> please elaborate.
> 
> > I do not consider this a bug, but a feature. Proofs that the ac is
> > working in locking out the ones that you define.
> > 
> > If you did not got locked out then we need to worry.
> 
> either you have totally misread my bug report or i'm totally at sea
> about how ac is supposed to work.
> 
> 

-- 
Thorsten Scherler
COO Spain
Wyona Inc.  -  Open Source Content Management  -  Apache Lenya
http://www.wyona.com                   http://lenya.apache.org
thorsten.scherler@wyona.com                thorsten@apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: DO NOT REPLY [Bug 40609] - tab.acAuthoring throws exception and corrupts ac data

Posted by Jörn Nettingsmeier <po...@uni-duisburg-essen.de>.
before we start a REOPEN war on bugzilla, let's discuss this here first.

imho an exception caused by a normal gui operation is always a bug.
imnsho a "deny" credential for the role "edit" that leads to a loss of
"visit" for the user admin is a blocker bug.

bugzilla@apache.org wrote:
> ------- Additional Comments From thorsten@apache.org  2006-09-26
> 15:36 ------- If you switch the rights then it will work again.

please elaborate.

> I do not consider this a bug, but a feature. Proofs that the ac is
> working in locking out the ones that you define.
> 
> If you did not got locked out then we need to worry.

either you have totally misread my bug report or i'm totally at sea
about how ac is supposed to work.



-- 
"Open source takes the bullshit out of software."
	- Charles Ferguson on TechnologyReview.com

--
Jörn Nettingsmeier, EDV-Administrator
Institut für Politikwissenschaft
Universität Duisburg-Essen, Standort Duisburg
Mail: pol-admin@uni-due.de, Telefon: 0203/379-2736

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org