Smolt database is broken

[Christopher Aillon]

On 11/20/2007 12:52 PM, Alan Cox wrote:
> On Mon, Nov 19, 2007 at 11:32:03PM -0500, Yaakov Nemoy wrote:
>> That is in fact a possibility.  It uses the output from
>> /proc/sys/kernel/random/uuid to work.  Is there a flaw with this
>> method that we know nothing about?  Should we use another?
> 
> Well libuuid is the other alternative. I'm a bit concerned about reports
> the kernel uuid generator isn't as random as it should be. I'd like to know
> more - such as what uuids come up a lot and chase that further.

I agree we should track this down in the kernel.  But even so, it is 
theoretically possible for the same UUID to come up multiple times. 
Shouldn't we be generating UUIDs on the smolt server itself and letting 
the client know what the UUID is, to better protect against duplicates?