| 1 |
El mar, 03-05-2005 a las 22:54 +0200, Simon Stelling escribió: |
| 2 |
> Nice work, but do we really need that strong passwords? IMHO the |
| 3 |
> passwords generated by portage should be changed right after emerging |
| 4 |
> the piece of software anyway, although that might not be reality in many |
| 5 |
> cases :( I didn't have a closer look to the packages you listed, but i |
| 6 |
> hope those don't need a password for a unix account. Probably (just |
| 7 |
> guessing) they save the password in plain text anyway, so it wouldn't |
| 8 |
> matter that much... |
| 9 |
> |
| 10 |
> Perhaps I'm just totally wrong. |
| 11 |
|
| 12 |
Not, but you can't ensure that users will change them, you can warn |
| 13 |
them, you can bite them, you can kick them to do it, but they will end, |
| 14 |
maybe, not changing the generated password. |
| 15 |
|
| 16 |
Also, having the password in clear text has nothing to do with the |
| 17 |
permissions that allow or restrict a certain user from accessing it, |
| 18 |
that is, standard DAC (aka "standard Unix permissions") prevents users, |
| 19 |
at least theoretically, from accessing those files you don't want them |
| 20 |
to access, among that you can use MAC or any other more complex model to |
| 21 |
enforce such access restrictions (ie. by using SELinux, RSBAC... from |
| 22 |
Hardened Gentoo). |
| 23 |
|
| 24 |
(Check the dev-db/phpmyadmin bug regarding the wrong permissions of the |
| 25 |
SQL script with the password of the pma user). |
| 26 |
|
| 27 |
As less risks we assess, less responsibilities we have to manage |
| 28 |
regarding Q/A. |
| 29 |
|
| 30 |
Cheers, |
| 31 |
-- |
| 32 |
Lorenzo Hernández García-Hierro <lorenzo@×××.org> |
| 33 |
[1024D/6F2B2DEC] & [2048g/9AE91A22][http://tuxedo-es.org] |
| 34 |
|
| 35 |
-- |
| 36 |
gentoo-dev@g.o mailing list |
Archives