thanks for the link, so all addresses are affected regardless of the client (due to them being based off the same RNG weakness), has this weakness not been resolved in newer versions of OpenSSL? if so could it be worth upgrading OpenSSL and recompiling the wallet or have i completely missed the point?

in the mean time, services that use public wallets would have to generate new addresses (to send and receive money periodically?). whats stopping someone going after the public key of an exchange like cryptsy, what would a service like this do to mitigate this issue?

Quote from: BurtW on January 29, 2014, 08:57:06 AM

Please see:

https://bitcointalk.org/index.php?topic=437220.msg4813821#msg4813821

it would be great to identify whether your key is particularly weak and more susceptible and to also identify the risks of services using the same wallet to send and receive payments.. are they going to have to change their addresses regularly from now own just to avoid this attack?

Quote from: toknormal on January 29, 2014, 09:17:53 AM

To put this into perspective:

[1] - imagine the bitcoin address space is ALL the sand grains on planet earth (it's actually much bigger than that I think but this is easier to visualise)

[2] - imagine going to a particular spot in some country with a magnifying glass and identifying a particular sand grain

[3] - now move out from that sandgrain and identify the 5 sand grains **touching** the one you spotted with your magnifying glass. These are the "weak address" sandgrains

[4] - now imagine an astronaut orbiting the planet who lands at some random location and picks themselves a random sandgrain at their landing spot

Now you can see that the chance of collision with one of the 'weak addresses' is almost the same as the chance of collision with the primary address = no weakness at all.

Forget about it. The issue is of theoretical interest only.