1998 attack that messes with sites’ secret crypto keys is back in a big way

[StefanM]

A surprisingly big number of top-name websites—Facebook and PayPal among them—recently tested positive for a critical, 19-year-old vulnerability that allowed attackers to decrypt encrypted data and sign communications using the sites' secret encryption key.

The vulnerability in the transport layer security protocol for Web encryption was disclosed in 1998 when researcher Daniel Bleichenbacher found it in the TLS predecessor known as secure sockets layer. A flaw in the algorithm that handles RSA encryption keys responded to certain types of errors in a way that divulged potentially sensitive information. With enough specially formed queries, attackers could exploit the weakness in a way that allowed them to decrypt ciphertext even when they didn't have the secret decryption key. SSL architects responded by designing workarounds that suppressed the error messages rather than removing or rewriting the faulty RSA algorithm.

Full story at Ars Technica

 

[RejZoR]

What's with surfacing of security flaws dating decades back? First Spectre/Meltdown and now this? WTF?

 

[natr0n]

Everyone is like "Oh know im going to lose some performace yet again"

 

[R-T-B]

What's with surfacing of security flaws dating decades back? First Spectre/Meltdown and now this? WTF?

Spectre and Meltdown may be decades old, but they are relatively newly discovered. This has been KNOWN for 19 years, so no excuse.

That said, as bad as this is, someone still needs to tap your line and break the password hash between you and paypal to actually hijack an account, so nothing much meaningful will probably happen from this. It still needs to be patched immediately.

 

[HTC]

Spectre and Meltdown may be decades old, but they are relatively newly discovered. This has been KNOWN for 19 years, so no excuse.

That said, as bad as this is, someone still needs to tap your line and break the password hash between you and paypal to actually hijack an account, so nothing much meaningful will probably happen from this. It still needs to be patched immediately.

But it hasn't been fixed for 19 years: instead, a workaround was implemented.

The vulnerability in the transport layer security protocol for Web encryption was disclosed in 1998 when researcher Daniel Bleichenbacher found it in the TLS predecessor known as secure sockets layer. A flaw in the algorithm that handles RSA encryption keys responded to certain types of errors in a way that divulged potentially sensitive information. With enough specially formed queries, attackers could exploit the weakness in a way that allowed them to decrypt ciphertext even when they didn't have the secret decryption key. SSL architects responded by designing workarounds that suppressed the error messages rather than removing or rewriting the faulty RSA algorithm.

Since the problem was never fixed, it was "just waiting" for another way to be taken advantage of, and that way was just found, it seems.

 

[TheoneandonlyMrK]

But it hasn't been fixed for 19 years: instead, a workaround was implemented.



Since the problem was never fixed, it was "just waiting" for another way to be taken advantage of, and that way was just found, it seems.

And shows in crystal clarity just what can be done at times to fix issues and also just how fixed they are in reality.

 

[R-T-B]

But it hasn't been fixed for 19 years: instead, a workaround was implemented.

A workaround that should be effective if actually implemented. My understanding is the above sites are simply configured badly.

 

[HTC]

A workaround that should be effective if actually implemented. My understanding is the above sites are simply configured badly.

But, if it was indeed fixed, this could not have happen, no?

By leaving the issue unfixed, they exposed themselves to further problems, but the real problem is that it took 19 years for them to notice.

 

[R-T-B]

But, if it was indeed fixed, this could not have happen, no?

By leaving the issue unfixed, they exposed themselves to further problems, but the real problem is that it took 19 years for them to notice.

My understanding is it was fixed by simply suppressing errors that could leak data in the default config. This does not prevent idiot admins from overriding the default config to expose those errors again.

 

[HTC]

My understanding is it was fixed by simply suppressing errors that could leak data in the default config. This does not prevent idiot admins from overriding the default config to expose those errors again.

According to Ars Technica's article (see below), they specifically did not fix the problem and instead used workarounds to suppress the issue.

SSL architects responded by designing workarounds that suppressed the error messages rather than removing or rewriting the faulty RSA algorithm.

 

[AsRock]

What's with surfacing of security flaws dating decades back? First Spectre/Meltdown and now this? WTF?

O come on, one reason i stopped watching the TV, news channels make it olook like a break out of crap but in realility it's much more often than they make out.

 

[R-T-B]

According to Ars Technica's article (see below), they specifically did not fix the problem and instead used workarounds to suppress the issue.

That is essentially what I just stated. Semantics perhaps, as it's still broken.

 

[HTC]

That is essentially what I just stated. Semantics perhaps, as it's still broken.

A workaround is not a fix, as far as i'm concerned.

To give a somewhat crude analogy:

If a water heater suddenly starts not enabling more heat (more gas usage), you should not "fix it" by reducing the amount of water, thus heating the water more. Instead you should fix the heating problem. This kind of workaround seems to be the thing that happened with this exploit.

 

[R-T-B]

Fair point.