4 ms·
Whose responsibility is it to prevent this? USCYBERCOM? http://en.wikipedia.org/wiki/United_States_Cyber_Command http://en.wikipedia.org/wiki/United_States_Cyb
by dm2 12y ago
Whose responsibility is it to prevent this?
USCYBERCOM? http://en.wikipedia.org/wiki/United_States_Cyber_Command http://en.wikipedia.org/wiki/United_States_Cyber_Command
The NSA should also play a role in defending the US from attacks via the internet.
The hackers must have used a US based VPN, because it would be retarded to allow connections directly from China to access any US DoD systems, right?
- Someone1234 12y agoThe department itself is responsible for preventing this. Cyber Command shouldn't be operating on US soil (or constitutional reasons), at least outside of "training exercises" and securing inter-DoD systems (of which the federal government is not). The NSA can help other departmental and non-departmental organisations secure their system either through consulting or potentially even pen testing (that's legal and part of their remit). The FBI has also taken an on-again/off-again role in trying to secure inter-US systems but they don't really have the budget or remit to be doing that fully. > The hackers must have used a US based VPN, because it would be retarded to allow connections directly from China to access any US DoD systems, right? As you yourself eluded to, blocking connections based on GeoIP is utterly utterly pointless. The internet is designed such that you can route around those types of blocks fairly trivially (e.g. proxies, VPNs, foreign servers, private interconnect agreements, et al).
- dragonwriter 12y ago> Cyber Command shouldn't be operating on US soil (or constitutional reasons) Could you elaborate the specific constitutional bar to Cyber Command operating on US soil?
- Someone1234 12y agoThis Yahoo! Answer (of all places!) does an amazing job of answering this: https://answers.yahoo.com/question/index?qid=20090401073222AAzQsmX https://answers.yahoo.com/question/index?qid=20090401073222A...
- dm2 12y agoI'd argue that the intention for that is to protect against using military equipment against civilians. (which is already being done in some cases, MRAPs, automatic weapons, body armor, SAMs) Using the military to protect from foreign attacks is exactly what it's suppose to be used for. What's the harm of protecting US govt employees and even civilians and businesses from internet attacks with agencies that are technically part of a military branch? I'm not saying send troops after anyone, if someone within the US hacks something, either pass it to the FBI or just fix the security hole and move on. From the page you linked to: "Active participation in civilian law enforcement, such as making arrests, is deemed to be a violation of the Act, while taking a passive supporting role is not. Passive support has often taken the form of logistical support to civilian police agencies." So basically that doesn't apply to internet and telecommunications protection. I'm not suggestion that USCyberCom hacks US citizens or businesses, but they possess the capabilities to help protect them which in turn strengthens their own security.
- Someone1234 12y agoThe way Cyber Command operates is quite offensive (e.g. attacking servers used by "enemy" "combatants"). The NSA's job is to secure US systems, the Cyber Command's job is to go attack foreign country's systems. There is definitely some overlap there (particularly with cyber Command working to improve DoD security). I'd say that ultimately the NSA has a far bigger/better legal right to be doing that kind of thing than the Cyber Command does (and also the law wouldn't be super-clear-cut if the Cyber Command had to attack a US-hosted server, network, or client). Or to use a really bad analogy: You could have the US Army defend Washington DC instead of the Capitol Police, however most would be more comfortable with the Capitol Police and there are far less potential legal hurdles also. Everything is fine if all the US Army did all day was stand outside of buildings, but as soon as they literally have to shoot someone all hell is going to break lose (or in our case as soon as Cyber Command had to act in an offensive way against a US company, citizen, or similar).
- dm2 12y agoFrom the wikipedia article on USCYBERCOM: "In July 2011 Deputy Defense Secretary William Lynn announced in a conference that “We have, within Cyber Command, a full spectrum of capabilities, but the thrust of the strategy is defensive.” “The strategy rests on five pillars, he said: treat cyber as a domain; employ more active defenses; support the Department of Homeland Security in protecting critical infrastructure networks; practice collective defense with allies and international partners; and reduce the advantages attackers have on the Internet.”" Their directors have emphasized that while it will have extensive attack capabilities that it will also provide support to all departments to help strengthen system security across the board. Banks, hospitals, government departments not relating to the military, utilities (power/water), are all critical to the nation and we must use every asset we have to protect them. I'd argue that any analogy involving armed soldiers or deadly equipment is very different from defensive cyber warfare. Cyber Command should NEVER act offensively against a US company or citizens, no government organization should in my opinion unless a similar level of physical violence is threatened or preformed by those citizens. If I was in charge I would allow all US citizens to attack US government systems as long as they responsibly disclose the security vulnerabilities and do not distribute any stolen information (and as long as they don't cause a service interruption), without fear of prosecution.
- dm2 12y agoIs it reasonable to expect every department, organization, and agency to have flawless security? Not necessarily blocking based on IPs, but requiring VPNs to access these networks, which would be a requirement for another level of authentication and another point of monitoring who accesses the systems.
- Someone1234 12y ago> Is it reasonable to expect every department, organization, and agency to have flawless security? No. But it is no more reasonable to assume that a single department or organisation would have a remit to secure every computer system of the US government. > Not necessarily blocking based on IPs, but requiring VPNs to access these networks, which would be a requirement for another level of authentication and another point of monitoring who accesses the systems. You mean employees require a VPN to access these systems or the "bad guys?" VPN access for internal networks is fairly standard already. A handful of these kind of incidents originate when an employee infects their machine with malware which then "calls home" and gives others a backdoor way into the network (which can then be sold on or utilised).
- jdong 12y agoUSCYBERCOM is only responsible for .mil networks. DHS is in charge of the security of .gov networks.
- r00fus 12y agoWell great. Does DHS have the appropriate competency to handle this kind of stuff?
- dm2 12y agoTo protect US (non-military) systems from foreign intelligence agencies and hackers? Probably not. I can't imagine many security researchers would want to work for the DHS unless the pay was enormous.
- mpyne 12y agoIndeed, it's a known problem area for the Federal government. http://www.rand.org/pubs/research_reports/RR430.html http://www.rand.org/pubs/research_reports/RR430.html