19 replies
by danlm
I have issues with people trying to ssh into my home machine all the time.... What I look for, after I firewall the pricks is........ Was a successful log in made? With that I have the ip address of the log in.
Won't the RAS show you the same? Ie, if an actual successful log in occurred with that use rid???
dan
by dbell
by danlm
I get slammed constantly on my home machine for brute force attempts at ssh. The only time I contact the people is when the IP is based in the united states. And then I provide a copy of the log to back up what I am notifying them of. I always contact their abuse email...
Unfortionally, I have only ever had one company get back to me.... But, I'll be damned if I'll make it easy on people like that. You've gone to this much trouble trying to insure your ok, whats 1 step more to possibly make life miserable for the twit that did this.
Dan
by marathoner
I had ip addys from rr hammering on me too.
Luckily they only annoyed us, didn't really hurt anything.
by neon samurai
The previous posts covered question 1; did they actually get in. You should have some standard idea of step to follow, check logs for successful login, where did the intruder go and what did they touch, how did they get in and how can it be corrected.
Ideally, if this is a business, you should have a standard investigation and documentation process setup. It's more responsible for a business to say "someone almost got through, here's what we did to be proactive for next time." Local law should be notified and the remote ISP notified incase the user is a known troublemaker. Even if your going to take the older way of thinking and skip notifying the public or law; document everything and adjust your security settings accordingly. It would also be worth contacting the Hacker Profiling project (sorry, no website link) as they are compiling a database of such attempts.
In a home setting, it's more your call. Was the breach big enough to warrent law enforcement? Was it large enough or repeated frequently enough to warrent notifying the registered owner of the IP address or there ISP? Are you better to simply adjust your security to cover the now discovered whole?
I had someone banging on my FTP server, in that case, it was more fun to turn my security apps back at them; I always did wonder what the look on there face was when they realized they got profiled by the mark (win2k box in North Korea according my trace).
by danlm
My experience has been with my home server(FreeBSD), I get attempts every day. They no longer can do brute force attacks because of firewall settings and personal scripts I wrote. 3 times and your out. If I firewall myself from work, I wait till I get home and remove it.
But, even as a home user I have established procedures. Automatic checking of my logs daily, listing off attempts. This is with who is information. Hell, I can tell you how many times a specific ip has tried over a period of time.
If it happened once, it's going to happen again. Lessons learned, be prepared so you can followup quickly.
Dan
by ctmcswain
by danlm
The OS is FreeBSD 6.1, and they ported over a latter version of the PF packet filter from OpenBSD. PF has throttle support.
[i]
pass in on $intf_in proto tcp to $intf_in port ssh flags S/SA keep state \
(max-src-conn 10, max-src-conn-rate 5/20, overload flush global)
pass in on $intf_in proto tcp to $intf_in port ftp flags S/SA keep state \
(max-src-conn 5, max-src-conn-rate 5/20, overload flush global)
[/i]
Those two lines in my pf.conf will stop most brute force attacks. More then 5 log in attempts in 20 seconds from the same ip is firewalled.
I then wrote a shell script to parse my auth.log, and firewall stuff identified in that. I have a Perl script that looks through my ftp log looking for these twits. These scripts get run every two minutes.
At the end of the day, I update a MySQL db I built for analysis purposes.
Other then the pf firewall, everything is home grown. I get slammed, this is no lie. I had 3 email notifications yesterday that brute force attempts were attempted against my ftp. This is the norm, no word of lie.
Dan
by dr dij
security mgmt software reads the logs and filters out false positives generally, and shows relations between events you might not realize were connected;
it can start incidents and collect all info; they can either shut out the intruder or ask you if you want to.
and others are right, procedures should be setup
if you're too small a company, there are managed providers who will monitor
Do not investigate without reporting it. File a report with your senior management and if you have a corporate policy in place, this will be passe on to the authorities. There have been many cases where sys admins have investigated and later the evidence failed to stand up in court since the "state of the system has been altered" and "it is no longer possible to vouch for the integrity of the system", whatever that means.
But if you need to raise hell to get this reported, do so.
by frvr
by dbell
by dbell
by danlm
This is my home machine and is set up completely how I want it. You have a Cisco firewall, which I believe is hardware. That's much better. What I described is like me running Norton on my windows machine. It's all software. I just have more control over it, that's all.
Dan
by balthor
by stimpi
Disable RAS in services
change reg policies to null
Port scan your self, stuff like that
If I had an IP I could find the weeknes
J
by packetracer
Contacting the authorities will do nothing for you. It used to be that they won't move a finger unless you have suffered at least $5,000 in damage. That may have changed, but... Either way you have to be certain that you've been broken into before you'll get any help from the authorities.
You're better off contacting a security company (maybe you have one that you already work with?). They will help you figure out if your servers have been compromised... and will help you prevent a future security problem.
As far as the login attempts, there are a few things here that you should have done to mitigate any issues:
1) The user's account should have been disabled as soon as she left the company. You did do that, right?
2) Your RAS server is patched with the latest patches, right?
3) Going back in time through the RAS logs. Ideally you kept at least a month worth of your RAS logs. Not many people do that, but you had the foresight to do that, you can go back to see when did the attempts start. They could have been happening while the employee was still with the company.
4) Firewall logs - all models of the Cisco PIX have logging! You will need to setup a syslog server to start receiving the logs from the PIX.
5) Make sure that your employees pick secure passwords and change them once every few months. Can be done through Group Policy. You can also run free cracking tools to check the strength of current passwords.
6) If possible try to figure out how the ex-employee's user name/password was stollen. Did she always login remotely from her own computer? Let her know that she may have malware/keylogged running on her computer.
Hope that helps!
by dbell
1. The user was disabled the day she left and permanently deleted a few months later.
2. According to the Microsoft Baseline security analyser all the patches are current.
3. We don't have that much history, but that was the only attempt I see in the last month.
4. I will have to have a word with our firewall consultant about that.
5. The group policies have always been enabled with password length and complexity requirements. I haven't tried any password crackers.
6. I suspect that the hacker got her name from a website or from directory harvesting spam but I can't be sure. No other suspicious attempts have been made to access RAS under any username.
You're right about the authorities, they were'nt very interested.
by dbell
This discussion is closed to new replies.
Trending Windows Forum posts
Sort by Trending