I came across a neat little
command that will
allow you to SSH through an http-proxy. Useful for when you're at a
library or elsewhere and need to make an outbound SSH connection and the
only thing stopping you is a proxy.
Features of connect.c are:
- Supports SOCKS (version 4/4a/5) and https CONNECT method.
- Supports NO-AUTH and USERPASS authentication of SOCKS
- Partially supports telnet proxy (experimental).
- You can input password from tty, ssh-askpass or environment variable.
- Run on UNIX or Windows platform.
- You can compile with various C compiler (cc, gcc, Visual C, Borland
C. etc.)
- Simple and general program independent from OpenSSH.
- You can also relay local socket stream instead of standard I/O.
Posted by Marcin on Saturday, June 9, 2007 in
Security.
Mikko @ F-Secure made a
post
on their blog about whether or not law enforcement organizations should
be permitted to utilize security tools and hacking techniques in
investigations that got me thinking. To me the answer to this question
is very clear -- NO WAY JOSE! -- not unless proper oversight can be
implemented and safe guards to protect our privacy are devised.
EFF, help us on this one!
Given that police have been abusing laws made to combat terrorism to
violate the rights of US citizens that are not members of terrorist
organizations, I think that opening the gates to allowing the use of
hacking techniques and malware invites a whole host of other problems.
The first problem that I have with this is the cost of training. Will
training police in computer security be cost-effective? I would be
leaning more towards saying no. Learning security techniques is not
something that they can create two week training courses on. Also, will
the cost of providing advanced security training to police provide
enough benefit to us to justify the spending?
The second problem that I have with this is the risk associated with
police using these tools and techniques without proper understanding of
the tools and their effects. Will Joe Blow Officer here in Phoenix know
what he's doing or will he be damaging my server I have at home while he
tries to install his spyware?
Another problem that I have with this is how will prosecutors be able to
prove that the data collected by investigators hasn't been tampered with
or fabricated. Without proper controls and oversight for these types of
operations officers with an axe to grind will be able to go penetrate
computers and plant fabricated evidence. Will companies be forced
through legislation to preinstall the software onto computers prior to
shipping them to consumers? and if so, will it be a felony to remove it?
Lastly, if we properly secure our systems against what I feel is a gross
violation of my rights to privacy, will there be legal ramifications?
Will I get sent to jail because I have a proper firewall and IDS? Will
they 'disarm' us of security tools like Britain did to their citizens
when taking firearms? Will security tools be outlawed without a license
to own them? We need to think about things like this and tell our local
government officials what we think before it's too late.
Posted by Casey on Tuesday, June 5, 2007 in
Politics,
Privacy and
Security.
I started working on a project that has no doubt, been done before. It's
something no one has publicly posted information on and it's not new --
something everybody wants yet every vendor says is impossible. The
problem with this project, is it can't be managed by IT security at a
"sysadmin level" in production. The only role IT security will have in
the project after rollout, is implementing new features. If it can't
serve the customers (site owners) exactly as they want at an abstracted
level, it's just another project to waste several weeks on.
Too often in security we'll do things just to please us -- to make us
feel good, but what value do we really add? That's what I learned this
first week of my internship.
Posted by Marcin on Saturday, June 2, 2007 in
Security and
Work.
Christopher Soghoian has an excellent remote vulnerability
disclosurereport
concerning Firefox Add-ons. More than several extensions from various
3rd parties are vulnerable to man-in-the-middle attacks.
Q: Who is at risk?
A: Anyone who has installed the Firefox Web Browser and one or
more vulnerable extensions. These include, but are not limited to:
Google
Toolbar,
Google Browser
Sync, Yahoo
Toolbar, Del.icio.us
Extension, Facebook
Toolbar, AOL
Toolbar,
Ask.com Toolbar,
LinkedIn Browser
Toolbar,
Netcraft Anti-Phishing Toolbar,
PhishTank SiteChecker.
Q: When am I at risk?
A: When you use a public wireless network, an untrusted Internet
connection, or a wireless home router with the default password set.
He provides two fixes to the problem:
- Disable extensions not obtained through https://addons.mozilla.org or
through an authorized SSL enabled website. (Add-ons obtained through
the official Mozilla add-on repository and/or secured, SSL enabled
sites are not vulnerable to MITM attacks)
- Disable automatic updates of Firefox Add-ons
I disable automatic updates for Firefox on my laptop through
Edit>Preferences>Advanced>Update tab (Tools>Options>Advanced>Update tab
for Windows users).
The thing that gets me about this vulnerability is it doesn't seem very
probable. Around 75% of my computer illiterate friends use Firefox, and
over 90% of them don't even know what an extension or add-on is, since
they were so used to IE and are accustomed to that level of
functionality. They still have the default news RSS feed in the bookmark
toolbar and all bookmarks are under the top directory). It would take an
attacker watching unsecured wifi networks or cafe hotspots all day long
to find someone vulnerable to exploit. Doesn't seem very economical to
me, when there are so many other vectors for attack.
Posted by Marcin on Thursday, May 31, 2007 in
Security.
Andrew Hay writes:
Dell & Google Secretly Installing Software to Make Money Off Your
Typos
- Those bastards, how is this business practice not illegal?
New Dell machines that include the Google toolbar as part of a
marketing agreement also include a secret program that redirects
non-url information typed into a browser window to a Dell-branded
page filled with ads. For example if you type in dogfood.cim,
instead of getting a browser error message, the secret Google
Address Redirector redirects the query to an ad-filled page of
search results.
They're simply redirecting invalid DNS requests to their own page. When
you use OpenDNS, they redirect mistyped queries to their own page as
well. OpenDNS is having a fit because it's competing with their revenue
stream,
powered by Yahoo!
To remove all installed software, leaving you with a clean system and
for those who don't wish to reformat their new computer, run PC
Decrapifier.
Posted by Marcin on Thursday, May 24, 2007 in
News,
Privacy,
Security and
Tech.
