I don't do this completely, but I'm migrating there, which is run everything on 
VM and do snapshots and full backups often.

That doesn't solve security issues with information compromise, but it does 
bypass cryptolock entirely.
If someone gets in and locks your information, you just roll back to before and 
tighten security.
Most VM management tools now days do this in a few clicks.

Security is another issue.

I really need to start using LONG username and passwords, which appears to be 
one of the best ways to defeat the brute force stuff.

From: Af [mailto:[email protected]] On Behalf Of Tyler Treat
Sent: Friday, September 30, 2016 5:31 AM
To: [email protected]
Subject: Re: [AFMUG] the future of internet security

So lots of things - I just woke up so bear with me:
Most are dominant on the Corp side, but are migrating to SMB.

2FA/MFA is big and getting bigger.  Most of your major players support native 
integration with a 2FA provider.   For example: log in to your company VPN with 
you AD credentials, Duo sends a push notification to your device that you 
accept before you're allowed in.  Same goes for Windows servers.
Yes, pretty much block all access that's not needed.  Allow VPN from the world 
if you have to.  Require access to anything else to be across the VPN.
Microsoft has "Direct Access" which implements a seamless behind the scenes 
tunnel back to the "inside" of your network, though it requires IPv6 on any 
internal resource you need to access.
  Lots of things changing in the Windows world.  Hybrid Cloud/On-Prem networks 
are starting to pop up, where you can be on net at the office with your local 
DC's, then talk to the great DC in the sky when you're on the road.
The key is that no matter where you're at, you're sending all your traffic 
through your corporate controls.
Mobile Device Management suites are doing the same thing for phones and tablets.
Citrix and / or VDI has its use case too, where you can keep all the data in 
your friendly confines, and only send pictures across the wire.  Admittedly 
more useful for helping with running "heavy" apps across the network for 
performance reasons, but there's a security play there as well.
SQL injection is happening all the time as well, so be protective of your 
Webapps.
Full disk encryption has it's place as well.
I was at a Sophos meeting the others day where they claim to be able to stop 
Cryptolocker etc with their new "CloudIntercept" product.  Certainly seems like 
a steep claim, but it's interesting.

Which leaves you.  You are the weakest link in all of this.  Lots of stats out 
there about the percentage of hacks linked to social engineering.  
Phishing/Whaling via email is a huge deal right now and the emails are getting 
better all the time.
Phone scams are still prevalent too.
At the end of the day, it's all time and money.  If you're targeted by someone 
who has one or both, they're going to get in some way.  Your goal should be to 
raise the bar high enough that the dumb easy shit doesn't compromise you and 
that you have an active, well rehearsed recovery plan for anything critical.


___________________________
Mangled by my iPhone.
___________________________
Tyler Treat
Corn Belt Technologies, Inc.

[email protected]<mailto:[email protected]>
___________________________


On Sep 30, 2016, at 1:22 AM, CBB - Jay Fuller 
<[email protected]<mailto:[email protected]>> wrote:

to be COMPLETELY ANAL i suppose you could deny ALL incoming access and permit 
ONLY the ip addresses you work from.
then it is more work whenever you change a connection but i am aware of that 
policy.
additionally, i suppose someone could spoof your ip.

I keep hearing other security holes are large enterprise environments.  you 
might lock the heck down on the corp network but then joe blow uses his company 
provided ipad with 4glte internet and connects right into something that isn't 
controlled by the internal network security policy and someone comes in that 
way....

robots, aren't they great?  Script kiddies....


----- Original Message -----
From: CBB - Jay Fuller<mailto:[email protected]>
To: [email protected]<mailto:[email protected]>
Sent: Friday, September 30, 2016 1:10 AM
Subject: Re: [AFMUG] the future of internet security


i truly believe about 80% of all hacking is social engineering - -

convince the company to let you in - - you're with the phone company to get 
faster internet.  let me in the server room.
i guessed my ex-girlfriend's cat's name.

i have no idea what is involved in hacking the DNC phone records or hillary's 
emails...

probably not social engineering.  weak passwords?
I know looking at a mikrotik log there are tons tons tons tons of random ips 
guessing root passwords...
i usually just disable those ports.


----- Original Message -----
From: CBB - Jay Fuller<mailto:[email protected]>
To: [email protected]<mailto:[email protected]>
Sent: Friday, September 30, 2016 1:00 AM
Subject: the future of internet security


Travis brings up a very good point with his randomware query.....and i caught 
the tail end of a discussion with i think it was cisco on some tv channel that 
i thought also raised some very good points.

One thing I always like about trade shows and user group get togethers is 
conversations about topics like this.

on tv the point was made that the internet is no longer coming "into one 
business" or a clearly defined demarcation point.
with the advent of internet of things and distributed work environments there 
is no longer just one place to look at securing.

as i thought about this it became very clear to me.  for example, even our 
company is pretty distributed.  we have up to five office staff that work from 
home at any given point in time - myself included.  we use web browsers and the 
powercode WISP management system - that does reside from within our network - 
but on a public IP address.  At my house (and at least two of our techs) they 
are not within our network footprint or the trees are so thick there is limited 
to no service here.  so my connection comes in off a cable modem.  I also have 
a voip phone - going back to our office - and our netsapiens voip switch.

in the near future we will probably have voip phones, remote access to our 
office (i do have a mikrotik with a permanent vpn connection to the office, but 
the aforementioned devices are all public ip accessible), if i happen to be 
visiting my parents or at a location with nothing more than a windows pc, i can 
access powercode or plug my phone in and hopefully not have to create a vpn 
connection or a remote desktop connection or remember to clean those 
credentials off.

in fact, even more annoying is a lake rental property we have with a pretty 
smart tv.  i've watched some netflix out there inbetween renters but when i 
leave and come home 'cause it's rented, you better be sure you clean your 
netflix account off that smart tv...

what about home security systems?  connected homes?  yes, i suppose they all 
login to the cloud and authenticate - hopefully by something more than just a 
username and a password.  would be nice if they authenticated by serial number 
or something - (like a mac address?) but maybe we access those with our smart 
phones, over our verizon 4glte connection, and not our home wifi, or maybe my 
dad's ipad - which is provided by his company and has company software on it - 
and probably shouldn't access my home automation stuff.

heaven forbid if my neighbor somehow accidentally connects to my home 
automation software instead of theirs - (hah, surely not)

one thing i have seen a lot more of and am probably good with - is this two 
factor authentication.   our rental management portfolio (homeaway) requires 
confirmation of both an email alert and txt alert to your phone before you can 
login (at least on a new device) - which i'm sure is set by a cookie.  i've 
seen this in banks too - so it's obviously more secure - but i also saw it once 
(i think) on ebay or paypal - - where someone registered a disposal phone to my 
account - and then somehow used that to authenticate fraudulently - that took 
some tracking down.

an associate recently notified me they wanted to change from using yahoo as 
"free email"  (i told them it sold to verizon anyway, not a bad thing to look 
for something else) - to something else cause it now required a cellphone to be 
registered.  i know from running a wisp about 40% of my customer base has a new 
cellphone number every three months.   must be the dating folks...

i suggested gmail - they laughed, said they didn't like gmail's clunky 
interface - - i have to admit i have been annoyed by it.  i have one company on 
gmail's domain and spam gets "eaten" then never delivered - even if it is 
legitimate mail.  after about three months of training the system i think it is 
finally delivering what it is supposed to - - but who knows.

ok, i've rambled.  but i think it's a good topic for discussion.  what does 
security --  not network security (cause we're on like 15 different networks, 
no?) - - what does secuity look like moving forward?

is the two-factor authentication becoming standard?  what is better?  Where do 
we go from here?

Good day. :)


Reply via email to