Security / Privacy
==================
Binkp secure encryption for all hubs.
Better privacy.
SSH officially supported.
SSH for specific echos.
# More discussion needed around these points. It's only as strong as weakest link and echomail may not have been designed with privacy in
mind. How best to enforce an echomail area only available via SSH?
# We could choose to 'secure' the network using something like ZeroTier
# We can offer echos and netmail but not privacy
There are several aspects where the current practice in fsxNet and the
BBSs connected to it are not compatible with the GDPR in the EU
(General Data Protection Regulation) (I guess there are other
countries with strict privacy laws that might apply too).
# We could choose to 'secure' the network using something like ZeroTierI used ZeroTier and it's quite easy to setup and works, but I dislike the idea to use a commercial provider for the basic infrastructure. FTN is DIY.
I don't really understand how european laws are enforcable in
non-european nations? If the BBS was in europe, sure, they must comply to european laws, but if a BBS is in another country.. do we have
international agreements to honour GDPR laws? Am I going to get
extradited from Australia if a European user logs into my BBS?
There are several aspects where the current practice in fsxNet and the
BBSs connected to it are not compatible with the GDPR in the EU
(General Data Protection Regulation) (I guess there are other
countries with strict privacy laws that might apply too).
I don't really understand how european laws are enforcable in
non-european nations? If the BBS was in europe, sure, they must comply to european laws, but if a BBS is in another country.. do we have international agreements to honour GDPR laws? Am I going to get
extradited from Australia if a European user logs into my BBS?
I don't see any need to block europeans from fsxnet / BBSing, it's up to them to comply with their own laws. What's to stop a european from
logging into a BBS via a proxy even if we did block them all out?
Ok, now say we care about the GDPR, how do we comply? is it simply a
matter of having a privacy policy?
Personally, I don't care. I'm not in europe, I'm never going to europe,
and I'm kind of offended that europeans think they can enforce their moronic laws on the entire world?
So you don't know the GDPR, but you know it is a moronic law? I wonder
how a non-moronic law would look like and work.
- don't store and process personal data that are not technical
essential
- get informed consent for the storage and processing of personal data
in advance
- don't make optional (non-essential) personal data a condition (as in non-optional) for using the service
- don't leak / transmit personal data to third parties (without
informed consent)
# We could choose to 'secure' the network using something like
ZeroTier
I used ZeroTier and it's quite easy to setup and works, but I
dislike the idea to use a commercial provider for the basic
infrastructure. FTN is DIY.
You dont have to use "a provider" with ZeroTier.
I run a ZeroTier network that is independant of "zerotier" (the provider) itself.
While you may argue that you "find" me through their root server (which
is the default) - it doesnt "have" to operate that way. I can populate a "moon" that you "orbit" around (their terms, not mine) so that zerotier
can be turned off and our connection still works.
I know ZeroTier were working on personal "roots" so that this moon thing has a less of a value (and they are no longer a sudo dependancy). (I
havent kept up with it recently though.)
The other good thing, with ZeroTier, you dont necessarily provide anybody on the network (who needs to be authorised if it is configured to do so), to see everything on all ports. You can firewall it to a certain extent
(at the network layer), such that only specific ports are permitted on
the network. (I did setup the FSX zerotier network this way.) (You could also have your own running firewall as well if you wanted.)
Is it completely independent?
Wikipedia tells me: "Virtual networks are created and managed using a ZeroTier controller. Management is done using an API,
proprietary web-based UI (ZeroTier Central), open-source web-based or CLI alternative. Using root servers other than those hosted by
ZeroTier Inc. is *impeded* by the software's license.
Can I configure the ports or has the admin the power to change the rules at will?
Is it possible to use ZeroTier in a really decentralized way?
Another incredibly powerful feature of ZeroTier is the ability to tap the entire network regardless of how widely distributed its
nodes are. Using the tee ability within a flow rule essentially copies every frame sent/received by nodes on the network and sends it
to a node of your choice such as an IDS or full packet capture solution such as Moloch.
from: https://blog.reconinfosec.com/locking-down-zerotier/
see also: https://www.zerotier.com/2016/08/31/capability-based-security-for-virtual-networks/
headline "Global Rules and Security Monitoring"
Is there a way to prevent this?
It's still kind of centralized (your moon).
While still a "VPN" - it is still semi public, so you still have obligations. Their are people you dont know on the network - but not
*anybody* - the network "admin" can choose to "authorise" (or not) those requesting to join it.
So in the case of a
Any VPN has to have some sort of a hub. Even ZeroTier. At least with OpenVPN it's open source, and we could customize it to how we
see fit and we need
not announce which port or which protocol type we decide to use.
The root nodes in this case would be hubs. There needs to be a central point within each network to host and serve the proper
security certs. Even with OpenVPN, a point/node would still be able to see another point/node within the private IP network. That
The root nodes in this case would be hubs. There needs to be a central point within each network to host and serve the proper
security certs. Even with OpenVPN, a point/node would still be able to see another point/node within the private IP network.
That
So no.
Like web serving - the DNS server has nothing to do with the SSL exchange that occurs when you "A" and the server "C" when you are
browsing a secure website.
So in the case of a "network" setup for "fsx" - the network admin would authorise nodes to access the "fsx" network (I would suggest based on
their application to join the network) - and de-authorise them when they leave the network.
We are still strangers here, but we are a list of known strangers and we can identify who is doing something in appropriate on the network and
take action if that is deemeed the right response.
But at the same time,
our conversations and traffic is encrypted from the outside world.
Anybody outside of the network cant get to our systems and do stuff
(which is the script kiddies reference I made when I started this thread).
Another incredibly powerful feature of ZeroTier is the ability to
tap the entire network regardless of how widely distributed its
nodes are.
Is there a way to prevent this?
I dont see this as an issue, it would be no differnet to tcpdump -ni eth0:
Anybody outside of the network cant get to our systems and do stuff
(which is the script kiddies reference I made when I started this thread).
So you propose everything should happen within the VPN? No open BBS / binkp ports to the real Internet?
So in the case of a "network" setup for "fsx" - the network admin would authorise nodes to access the "fsx" network (I would suggest based on
their application to join the network) - and de-authorise them when they leave the network.
-1
We are still strangers here, but we are a list of known strangers and we can identify who is doing something in appropriate on the network and
take action if that is deemeed the right response.
-1
there are other ways for encryption, which fit the FTN model better.
Another incredibly powerful feature of ZeroTier is the ability to
tap the entire network regardless of how widely distributed its
nodes are.
Is there a way to prevent this?
I dont see this as an issue, it would be no differnet to tcpdump -ni eth0:
I was not aware that you can monitor all of my fsxnet traffic with a tcpdump on your side.
For a corporate network this is obviously a feature, but in our use case I would call it a security flaw.
Sysop: | CyberNix |
---|---|
Location: | London, UK |
Users: | 22 |
Nodes: | 10 (0 / 10) |
Uptime: | 01:28:23 |
Calls: | 898 |
Files: | 4,585 |
Messages: | 686,776 |