New NetSpectre Attack Can Steal CPU Secrets via Network Connections

On 28/07/18 18:45, Anonymous wrote:
Nomen Nescio laid this down on his screen :
In article
William Unruh wrote:

On 2018-07-27, Anonymous wrote:
https://www.bleepingcomputer.com/news/security/new-netspectre-attack-can-steal-cpu-secrets-via-network-connections/


Except it can't. (or rather at the rate of about 1 bit per hour,
which of course is useless if the cache actually changes in that
time.. And if you have defened yoursef against spectre, this is useless.

I love scary computer stories at bedtime.

I love the sell of infected computers in the morning.
^^^^
Thats no way to talk about Windows!

--
Renewable energy: Expensive solutions that don't work to a problem that
doesn't exist instituted by self legalising protection rackets that
don't protect, masquerading as public servants who don't serve the public.

On 29/07/18 06:19, Nomen Nescio wrote:
In article
The Natural Philosopher wrote:

On 28/07/18 11:43, Melzzzzz wrote:
On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication with web
server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

Nope, they dont.

That's incorrect. Without a network listener, i.e. "app" in
this new world of minimalistic linguisticism, the attack can't
even begin.

Ah. semantics. They dont need a *special* 'app' to be installed. The
standard stuff will do. (apache etc)


--
"When a true genius appears in the world, you may know him by this sign,
that the dunces are all in confederacy against him."

Jonathan Swift.

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to be the
safest servers out there.

On 29/07/18 11:03, Anonymous wrote:
Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

Â*This simply doesn't make sense.Â* Linux for years has proven to be the
safest servers out there.

Well if you know what they are running, and on what CPU, thats all AIUI
you need to know.

Performance of known code on known hardware allows inferences about its
internal data to be made.





--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

On 29/07/18 20:40, Andreas Kohlbach wrote:
On Sun, 29 Jul 2018 05:03:50 -0500, Anonymous wrote:

Nomen Nescio expressed precisely :

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to be the
safest servers out there.

Servers have nothing to do with Linux. You can probably have bad (or not
patched) servers on Linux as good ones. So you can on Windows, Mac...

AIUI all you need to know is what processor and what code is running on
the server and then the fine details of network timings in reponse to a
probing attack reveal something about what should be secret.

That its probably a lot simpler to hand the sysdamin a USB drive and
$1,000,000 to get the pentagon's passwords is noted.

But all those foreign intelligence agnencies have to do SOMETHING for a
living



--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

On 2018-07-28, Nomen Nescio wrote:

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

At the rate it produces results, I'll be dead, buried and won't
give a ****.

I'm in charge of several servers on three different 1Gb/s (or faster) connections
to the public internet, so I'm a bit more worried, I'd better make
sure I've rebooted them with a recent (ie presumed safe) kernel before the SSL key
can be exfiltrated.

--
ت

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to be
the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

On 30/07/18 15:06, Anonymous wrote:
Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

Â* This simply doesn't make sense.Â* Linux for years has proven to be
the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

Â*OS/400 must really be good.Â* I have never heard of it.


Wasn't/isn't that IBM mainframe?



--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

In comp.os.linux.misc The Natural Philosopher wrote:
On 30/07/18 15:06, Anonymous wrote:
Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

* This simply doesn't make sense.* Linux for years has proven to be
the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

*OS/400 must really be good.* I have never heard of it.


Wasn't/isn't that IBM mainframe?

No IBM m/f is Z/OS or Z/VM. OS/400 runs on the AS-400 system which is
more like an old-time mini-computer. It's designed for a small to
mid-range shop that couldn't support a m/f.

Jerry Peters wrote:
In comp.os.linux.misc The Natural Philosopher wrote:
On 30/07/18 15:06, Anonymous wrote:
Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used to
conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg communication
with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

Â* This simply doesn't make sense.Â* Linux for years has proven to be
the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

Â*OS/400 must really be good.Â* I have never heard of it.


Wasn't/isn't that IBM mainframe?

No IBM m/f is Z/OS or Z/VM. OS/400 runs on the AS-400 system which is
more like an old-time mini-computer. It's designed for a small to
mid-range shop that couldn't support a m/f.

It's called IBM i nowadays.

https://en.m.wikipedia.org/wiki/IBM_i

--
Neodome

After serious thinking Nomen Nescio wrote :
In article
Anonymous wrote:

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell
wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used
to conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg
communication with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to
be the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

probably because you're never operated or worked on a real
computer.

IBM main frame software engineer.

Nomen Nescio expressed precisely :
In article
Anonymous wrote:

After serious thinking Nomen Nescio wrote :
In article
Anonymous wrote:

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell
wrote:
Melzzzzz writes:
I still can't see how anything related to network can be
used to conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg
communication with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to
be the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

probably because you're never operated or worked on a real
computer.

IBM main frame software engineer.

I'm familiar with you IBM guys. I had to give a bunch of "IBM
Software Engineers" classes on TCP/IP and SMTP because as a
group, you couldn't figure out how to format email messages and
drop them in a pickup folder to be sent. TCP/IP routing and DNS
was a mystery that sent some of you to retirement because you
couldn't make the transition from SNA.

Like a liberal democrat, you lump everyone into a corporate bundle.
As for me, I was an IBM main frame software engineer, not an IBM
enployee.

Nomen Nescio writes:

In article
Anonymous wrote:

After serious thinking Nomen Nescio wrote :
In article
Anonymous wrote:

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell
wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used
to conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg
communication with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to
be the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

probably because you're never operated or worked on a real
computer.

IBM main frame software engineer.

I'm familiar with you IBM guys. I had to give a bunch of "IBM
Software Engineers" classes on TCP/IP and SMTP because as a
group, you couldn't figure out how to format email messages and
drop them in a pickup folder to be sent. TCP/IP routing and DNS
was a mystery that sent some of you to retirement because you
couldn't make the transition from SNA.

Really, you know us all?

Retired MVS maven _and_ Linux expert.
Lots of other stuff too numerous to list.
Software developer since 1964.

How do I know you are an idiot?

--
Dan Espen

Anonymous writes:

Nomen Nescio expressed precisely :
In article
Anonymous wrote:

After serious thinking Nomen Nescio wrote :
In article
Anonymous wrote:

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell
wrote:
Melzzzzz writes:
I still can't see how anything related to network can be
used to conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg
communication with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to
be the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

probably because you're never operated or worked on a real
computer.

IBM main frame software engineer.

I'm familiar with you IBM guys. I had to give a bunch of "IBM
Software Engineers" classes on TCP/IP and SMTP because as a group,
you couldn't figure out how to format email messages and drop them
in a pickup folder to be sent. TCP/IP routing and DNS was a mystery
that sent some of you to retirement because you couldn't make the
transition from SNA.

Like a liberal democrat, you lump everyone into a corporate bundle.
As for me, I was an IBM main frame software engineer, not an IBM
enployee.

Still early, but stupidest thing I've read so far today.

--
Dan Espen

In article
Nomen Nescio wrote:

In article
Dan Espen wrote:

Nomen Nescio writes:

In article
Anonymous wrote:

After serious thinking Nomen Nescio wrote :
In article
Anonymous wrote:

Nomen Nescio formulated the question :
In article
Anonymous wrote:

Nomen Nescio expressed precisely :
In article
Melzzzzz wrote:

On 2018-07-28, Richard Kettlewell
wrote:
Melzzzzz writes:
I still can't see how anything related to network can be used
to conclude that some data is from cache or memory.
Care to explain how that can be concluded from eg
communication with web server?

The paper is not hard to find:
https://misc0110.net/web/files/netspectre.pdf

I have skimmed over paper, yet I see they need server side
application...

All they have to do is run linux and let the leakage begin.

This simply doesn't make sense. Linux for years has proven to
be the safest servers out there.

Wrong.

OpenVMS and OS/400 both put linux to shame.

OS/400 must really be good. I have never heard of it.

probably because you're never operated or worked on a real
computer.

IBM main frame software engineer.

I'm familiar with you IBM guys. I had to give a bunch of "IBM
Software Engineers" classes on TCP/IP and SMTP because as a
group, you couldn't figure out how to format email messages and
drop them in a pickup folder to be sent. TCP/IP routing and DNS
was a mystery that sent some of you to retirement because you
couldn't make the transition from SNA.

Really, you know us all?

Retired MVS maven _and_ Linux expert.
Lots of other stuff too numerous to list.
Software developer since 1964.

How do I know you are an idiot?

Retired means not gifted enough to continue.

--
Ex-Banyan/Novell/Citrix/CiscoUCS/EMC/Extreme, currently 3PAR
engineer happily rolling IBM products to the scrapyard because
they are too complicated, too expensive, and designed from the
start to be nothing but over-complicated lifetime revenue traps
for hardware and software services.