Encrypted OAuth1 all access token and secret with system salt for now, will replace this with use...

Encrypted OAuth1 all access token and secret with system salt for now, will replace this with user suplemental key once that gets exposed from cryptohome.

Also fixed crash on re-adding the same user to the system.

BUG=chromium-os:18634, chromium-os:18551
TEST=open /home/cronos/user/Preferences and make sure oauth token and secert are encrypted now
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=96031

[Chris Masone, 2011-08-08 19:41:12]

On 2011/08/07 19:17:19, zel wrote:

it seems heavyweight to have cryptohomed export key material for you to use
here.  Is tehre no symmetric key in the user's nssdb you could grab and use? 
or, could you generate one and use that?

[zel, 2011-08-08 20:17:41]

The supplemental user key usage id wad@'s idea. In the absence of it, I am
using the salt as the key right now.

On Mon, Aug 8, 2011 at 12:41 PM, <cmasone@chromium.org> wrote:

> On 2011/08/07 19:17:19, zel wrote:
>
> it seems heavyweight to have cryptohomed export key material for you to use
> here.  Is tehre no symmetric key in the user's nssdb you could grab and
> use?
> or, could you generate one and use that?
>
>
>
http://codereview.chromium.**org/7584026/<http://codereview.chromium.org/7584...
>

[Will Drewry, 2011-08-08 21:04:25]

Yeah -  nssdb seems like a better idea since the strength is really just in
separation - not any specific enforcements.

On 2011/08/08 20:17:41, zel wrote:
> The supplemental user key usage id wad@'s idea. In the absence of it, I am
> using the salt as the key right now.
> 
> On Mon, Aug 8, 2011 at 12:41 PM, <mailto:cmasone@chromium.org> wrote:
> 
> > On 2011/08/07 19:17:19, zel wrote:
> >
> > it seems heavyweight to have cryptohomed export key material for you to use
> > here.  Is tehre no symmetric key in the user's nssdb you could grab and
> > use?
> > or, could you generate one and use that?
> >
> >
> >
>
http://codereview.chromium.**org/7584026/%3Chttp://codereview.chromium.org/75...>
> >

[zel, 2011-08-09 03:13:27]

since we are planning to keep this out of M14, I would really like to land this
on TOT right now and then iterate over the encoding key choice issue in the next
few days.

so, PTAL

[Chris Masone, 2011-08-09 03:30:28]

On 2011/08/09 03:13:27, zel wrote:
> since we are planning to keep this out of M14, I would really like to land
this
> on TOT right now and then iterate over the encoding key choice issue in the
next
> few days.
> 
> so, PTAL

Won't changing the binary->hex mapping (lowercase to upper) break all existing
logins?

[zel, 2011-08-09 04:13:24]

whops, sorry - i forgot to upload the latest and greatest.

PTAL

On Mon, Aug 8, 2011 at 8:30 PM, <cmasone@chromium.org> wrote:

> On 2011/08/09 03:13:27, zel wrote:
>
>> since we are planning to keep this out of M14, I would really like to land
>>
> this
>
>> on TOT right now and then iterate over the encoding key choice issue in
>> the
>>
> next
>
>> few days.
>>
>
>  so, PTAL
>>
>
> Won't changing the binary->hex mapping (lowercase to upper) break all
> existing
> logins?
>
>
>
http://codereview.chromium.**org/7584026/<http://codereview.chromium.org/7584...
>

[Chris Masone, 2011-08-09 04:19:32]

LGTM w/nit

http://codereview.chromium.org/7584026/diff/11002/chrome/browser/chromeos/log...
File chrome/browser/chromeos/login/parallel_authenticator.h (right):

http://codereview.chromium.org/7584026/diff/11002/chrome/browser/chromeos/log...
chrome/browser/chromeos/login/parallel_authenticator.h:244: std::string
UserSupplementalKeyAsAscii();
It'd be good to file an issue for converting this to a key stored in NSS, and
then put a TODO here