On 06/09/11 13:07, Adam Stokes wrote:
>> +
>> +const char *
>> +mh_os_uuid(void)
>> +{
>> + uuid_t u;
>> + static char s[MAXUUIDLEN];
>> + unsigned char __RPC_FAR *rs;
>> +
>> + UuidCreate(u);
>
> This function generates a new UUID each time it is called. Is that the
> intended behaviour? It doesn't seem consistent with the Linux
> implementation of mh_host_uuid().
The UUID isn't stored anywhere initially but we could always add that
portion to the registry?
At the very minimum it needs to be cached per-session (SrvAgent::setup()
calls it twice, for example). To get consistent behaviour with the Linux
implementation, I agree - the registry sounds like the right place for it.
cheers,
Zane.