RE: Rogue ip address 169.254.x.x ???
From: Phil H (PhilH_at_discussions.microsoft.com)
Date: 01/25/05
- Next message: Arthur Kater: "Hive based registry."
- Previous message: Johns: "Re: NDISUIO_QUERY_BINDING"
- Messages sorted by: [ date ] [ thread ]
Date: Tue, 25 Jan 2005 02:17:01 -0800
I now seem to have a fix to this problem ... which was actually caused by
another problem ...
The fix is to comment out the (seemingly automatic) addition of 'vmini' in
cesysgen.bat. We have no idea why it is there in the first place.
The bug is caused by the legacy-registry flush mechanism ... which reverses
the order of values beneath its keys at every flush ... which results in
vmini being the first adapter instead of the real adapter ... for every
_other_ flush !!!! Grrrrr !
If the 'order' value worked for 'HKEY_LOCAL_MACHINE\Comm\' then this would
be avoided, but it doesnt.
"Phil H" wrote:
> My problem is that after a while (for some reason I am chasing and getting
> nowhere) my code start sending out some of its udp frames from a rogue
> address (eg 169.254.196.148)
> Most frames use its real (dhcp) address (got with GetAdaptersInfo is
> 171.192.15.81)
> I am monitoring with ethereal (with all name resolution turned off !)
>
> The only place I can find this ip address is in lan90001/parms/tcpip there
> is a value 'AutoSubnet' which is 169.254.0.0 on my system.
>
> AND - I cant get the prob to occur on my debug system ( a pda image) - only
> a release image !
- Next message: Arthur Kater: "Hive based registry."
- Previous message: Johns: "Re: NDISUIO_QUERY_BINDING"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|