Oh - my mistake, I see, again when I am not talking up to your official mind, I 
am wrong... :). 

These are just about one half of the problem:
http://openerp.com/forum/topic7743.html
http://openerp.com/forum/topic6389.html
http://openerp.com/forum/topic2449.html
http://openerp.com/forum/topic4213.html
http://openerp.com/forum/topic3725.html
http://openerp.com/forum/topic896.html

I tried to secure OpenERP, not to allow one user to read info not intended him 
to see. There are 2 things that should be done - remove access to the menu's 
and to the models that user should not access. But there is one model that 
could not be restricted - res.users. If we restrict this medel to let's say 
admin, no one else are able to log in. And this is where still plaintext 
password lives.

I am very sorry not to find anymore the topic where Fabien himself have 
proposed a patch - as I remember more or less working python code snippet to 
replace a password field with hashed result. But anyway these bugs are well 
known, but are not implemented. (I do not know of 5th version, have not tried 
yet.)

Ok, I tried to build a small perl script (I like this language, and consider it 
more mature), which logs in as unprivileged user through XML-RPC connection and 
reads unrestricted res.users model. And this is where you can find any and 
every login information.

This is true and tried by myself.

Why I have not submitted a patch for this? There are several reasons for this, 
and you can find them in my other posts. Anyway there are not only me who 
thinks, that there should be paid more attention to maturity and stability not 
only new features.

Recently I discovered a fork for this system called (could be found on 
tryton.org) Tryton, which aims to fix these issues. This is their policy, do 
not know how well they are at implementing it, have not tried it yet, although 
they have a release.

But to me this is clear result for lack of respect and attention to an ordinary 
contributor, remember German translations and contributions. Ah well, they were 
wrong too. :)

Though this absolutely is not good to divide force.

P.S. It should be better if such advocates as gegard would spend a little more 
time to test not just blindly oppose. Just think, who would spend time to 
discover things like this, if one would be "fake"?

Good day!
-sraps




-------------------- m2f --------------------

--
http://www.openerp.com/forum/viewtopic.php?p=27254#27254

-------------------- m2f --------------------


_______________________________________________
Tinyerp-users mailing list
http://tiny.be/mailman/listinfo/tinyerp-users

Reply via email to