Ruby, Rails, Gems Redux Part II

Did a little research on the gem for MySQL and decided to try this:

root@slackbox:~/RAILS/RAILS/Snorby# locate mysql_config
/usr/man/man1/mysql_config.1.gz
/usr/bin/mysql_config
root@slackbox:~/RAILS/RAILS/Snorby# gem install mysql -- --with-mysql-config=/usr/bin/mysql_config
Building native extensions. This could take a while...
Successfully installed mysql-2.7
1 gem installed
Installing ri documentation for mysql-2.7...
Installing RDoc documentation for mysql-2.7...
root@slackbox:~/RAILS/RAILS/Snorby#

Now about my Snort architecture, I'm thinking all I'm gonna have to do is copy my Snort database over to Slackbox and then have my two Snort machines (one internal and one sensor at a datacenter) report to Slackbox....OR, have the Snort sensors report to BOTH the FreeBSD server AND Slackbox! I think the latter will work and it sounds like the better solution.

I'll be updating this post with my successes and failures most of the night, I suspect, or at least until I get good and pissed off. LOL!

=====

Update:

There's nothing like backing up an 83MB database file on old hardware:

Starting: 6:31PM up 23 days, 19:27, 4 users, load averages: 2.89, 2.94, 3.13

Ending: 6:33PM up 23 days, 19:29, 4 users, load averages: 5.88, 3.98, 3.51

While I'm sure that's incomparable to an enterprise database, at one point, I thought the old dell system would lock up.

I also was trying to do this via phpMyAdmin on both machines, but I didn't know the dbase size was that large (4 yrs of sniffing data). phpMyAdmin on the BSD box would say it was finished exporting but I'd check the filesize and it was different each time (did it like 4 times before I decided to go commandline. phpMyAdmin kept giving me a filesize of between 20M and 40M. It must've been choking out. I optimized the dbase, also, so it was more than likely larger than 83MB.

=====

Update:

Had to upgrade MySQL, as my 83MB file wouldn't import into Slackbox's MySQL server. 30 seconds into the import, the import would lock up or die. Apparently, it's a known issue with MySQL's lower versions.

Anyways, after the import and creation of new MySQL users, I had to edit Snorby's config/database.yml file, specifically the development part. The reason:

root@slackbox:~/RAILS/RAILS/Snorby# script/server -p 11001
=> Booting WEBrick
=> Rails 2.3.2 application starting on http://0.0.0.0:11001
=> Call with -d to detach
=> Ctrl-C to shutdown server
[2009-07-17 21:48:14] INFO WEBrick 1.3.1
[2009-07-17 21:48:14] INFO ruby 1.8.6 (2007-03-13) [i486-linux]
[2009-07-17 21:48:14] WARN TCPServer Error: Address already in use - bind(2)
[2009-07-17 21:48:14] INFO WEBrick::HTTPServer#start: pid=3848 port=11001
/!\ FAILSAFE /!\ Fri Jul 17 21:48:17 -0400 2009
Status: 500 Internal Server Error
Can't connect to MySQL server on 'no_not_use' (111)

It's still not clear to me why I had to edit it, but I did because the production portion was populated with the proper credentials but I was still receiving the above error..."Can't connect to MySQL server on 'no_not_use'". When I did it, I stopped getting that error.

*** I found why I was getting the MySQL error. The config/database.yml development entry has 'mysql' for the database entry. It should be 'no_not_use'. I've edited this to what is was originally supposed to be and changed everything back to 'no_not_use'. I no longer get the error when using the production settings. ***

Also, notice that I ran in what I want to call 'debug mode' because I wanted to see what was hanging up the connection.

So, now, after some editing and fiddling, I get the following in 'debug mode':

root@slackbox:~/RAILS/RAILS/Snorby# script/server -e production -b 10.150.1.106 -p 11001
=> Booting WEBrick
=> Rails 2.3.2 application starting on http://10.150.1.106:11001
=> Call with -d to detach
=> Ctrl-C to shutdown server
[2009-07-17 21:55:37] INFO WEBrick 1.3.1
[2009-07-17 21:55:38] INFO ruby 1.8.6 (2007-03-13) [i486-linux]
[2009-07-17 21:55:38] WARN TCPServer Error: Address already in use - bind(2)
[2009-07-17 21:55:38] INFO WEBrick::HTTPServer#start: pid=3915 port=11001


Processing UserSessionsController#new (for ::ffff:10.150.1.106 at 2009-07-17 21:55:40) [GET]
Parameters: {"action"=>"new", "controller"=>"user_sessions"}
Rendering template within layouts/application
Rendering user_sessions/new

ActionView::TemplateError (undefined method `login' for #) on line #8 of app/views/user_sessions/new.html.erb:
5: <% form_for @user_session, :url => user_session_path do |f| %>
6: <%= f.error_messages %>
7: <%= f.label :login %>

8: <%= f.text_field :login %>

9:

10: <%= f.label :password %>

11: <%= f.password_field :password %>


app/views/user_sessions/new.html.erb:8
app/views/user_sessions/new.html.erb:5
/usr/lib/ruby/1.8/webrick/httpserver.rb:104:in `service'
/usr/lib/ruby/1.8/webrick/httpserver.rb:65:in `run'
/usr/lib/ruby/1.8/webrick/server.rb:173:in `start_thread'
/usr/lib/ruby/1.8/webrick/server.rb:162:in `start'
/usr/lib/ruby/1.8/webrick/server.rb:162:in `start_thread'
/usr/lib/ruby/1.8/webrick/server.rb:95:in `start'
/usr/lib/ruby/1.8/webrick/server.rb:92:in `each'
/usr/lib/ruby/1.8/webrick/server.rb:92:in `start'
/usr/lib/ruby/1.8/webrick/server.rb:23:in `start'
/usr/lib/ruby/1.8/webrick/server.rb:82:in `start'

Rendering /root/RAILS/RAILS/Snorby/public/500.html (500 Internal Server Error)

The thing is, I see nothing in the web browser, but:

We're sorry, but something went wrong.

We've been notified about this issue and we'll take a look at it shortly.

Something else that is nagging me that I was trying to fiddle with is:

[2009-07-17 21:55:38] WARN TCPServer Error: Address already in use - bind(2)

There is only one Ruby service running and nothing is utilizing that port when I run Ruby. I'm ignoring it for now.

I would love to see what the WEBrick logs show, if there are any.

For now, its time to do some serious Googling and maybe hit up my three Ruby/Rails books.

Ruby, Rails, Gems Redux

I decided to use Slackware this time. I've had better luck.

My install already has Ruby 1.8.6 (the latest stable is 1.8.7, I believe).

Ran into an issue when following these instructions. Was supposed to do 'rake gems:install' but got a 'prawn' error

root@slackbox:~/RAILS/RAILS/Snorby# rake gems:install
(in /root/RAILS/RAILS/Snorby)
rake aborted!
no such file to load -- prawn

Fixed it by using 'gem install prawn'. After running that command, I was able to run the 'rake gems:install' without error.

Now I'm having a similar issue when running 'rake snorby:setup':

root@slackbox:~/RAILS/RAILS/Snorby# rake snorby:setup
(in /root/RAILS/RAILS/Snorby)
Setting Up Snorby Database.
!!! The bundled mysql.rb driver has been removed from Rails 2.2. Please install the mysql gem and try again: gem install mysql.
rake aborted!
no such file to load -- mysql

Running 'gem install mysql' give me a BUNCH of errors:

root@slackbox:~/RAILS/RAILS/Snorby# gem install mysql
Building native extensions. This could take a while...
ERROR: Error installing mysql:
ERROR: Failed to build gem native extension.

/usr/bin/ruby extconf.rb
checking for mysql_query() in -lmysqlclient... no
checking for main() in -lm... no
checking for mysql_query() in -lmysqlclient... no
checking for main() in -lz... no
checking for mysql_query() in -lmysqlclient... no
checking for main() in -lsocket... no
checking for mysql_query() in -lmysqlclient... no
checking for main() in -lnsl... no
checking for mysql_query() in -lmysqlclient... no
*** extconf.rb failed ***
Could not create Makefile due to some reason, probably lack of
necessary libraries and/or headers. Check the mkmf.log file for more
details. You may need configuration options.

Provided configuration options:
--with-opt-dir
--without-opt-dir
--with-opt-include
--without-opt-include=${opt-dir}/include
--with-opt-lib
--without-opt-lib=${opt-dir}/lib
--with-make-prog
--without-make-prog
--srcdir=.
--curdir
--ruby=/usr/bin/ruby
--with-mysql-config
--without-mysql-config
--with-mysql-dir
--without-mysql-dir
--with-mysql-include
--without-mysql-include=${mysql-dir}/include
--with-mysql-lib
--without-mysql-lib=${mysql-dir}/lib
--with-mysqlclientlib
--without-mysqlclientlib
--with-mlib
--without-mlib
--with-mysqlclientlib
--without-mysqlclientlib
--with-zlib
--without-zlib
--with-mysqlclientlib
--without-mysqlclientlib
--with-socketlib
--without-socketlib
--with-mysqlclientlib
--without-mysqlclientlib
--with-nsllib
--without-nsllib
--with-mysqlclientlib
--without-mysqlclientlib


Gem files will remain installed in /usr/lib/ruby/gems/1.8/gems/mysql-2.7 for inspection.
Results logged to /usr/lib/ruby/gems/1.8/gems/mysql-2.7/gem_make.out

Grrr...!!!

But, I'm a lot closer this time than last time. I'll sort it out either tomorrow night or this weekend.

Oh, and one more thing. Maybe this is more complicated than it has to be, because I've already got BASE running on a server who's internal IP is 10.150.1.103 (FreeBSD on a Dell server/workstation). The MySQL server is also on that box. Maybe I don't actually need the MySQL gem? Can I leverage the actual database on the FreeBSD box and maybe skip some steps? We'll find out, otherwise, I'm going to have to maybe copy the database over to the Slackware machine so I can test. Yeah, I really wanted Snorby on the FreeBSD box but for some reason I'm more comfortable with Slackware.

Rails, Ruby, Gems...PITA

I spent the evening trying to get Snorby installed.

I've sporadically messed with Ruby on Rails before, actually getting it installed and playing a bit with it before moving on to other things.

Now, I've been hindered by an out-of-date Rails install. I tried to install Snorby and every step I have to take two steps backward. I ended up reinstalling to the latest version that FreeBSD (yeah, doing this on a BSD install, as it seems easier to install this way), but that version wasn't high enough.

Now, I'm installing the latest by source. I'm as far as I've ever been tonight, which is good because I'm running out of time. I'm installing the Gems at the moment and the install is agonizingly slow (doing this on a dual proc 450MHz machine). It appears most of this are documentation installs. :/

Maybe I can get this done and still be able to get a bit of sleep before I've to get up for work.

So far, see below .txt file...looks good so far:

http://wigglit.ath.cx/ruby.txt

AAARGH!!

[root@delly /usr/local/www/data/Snorby]# rake snorby:setup
(in /usr/local/www/data/Snorby)
Missing these required gems:
javan-whenever

You're running:
ruby 1.8.5 at /usr/local/bin/ruby18
rubygems 1.3.1 at /root/.gem/ruby/1.8, /usr/local/lib/ruby/gems/1.8

Run `rake gems:install` to install the missing gems.

Almost there but I'm out of time...will continue tomorrow.

Main Slackbox (named slackbox) back online

I finally had time to figure out what was going on with my Slackware tower. It has been down for so long I forget when it actually started having issues. I believe it began having problems since the beginning of 2009.

The culprit? Either the SATA drive that I installed in it awhile back or the SATA controllers on teh board. It's difficult to tell without swapping the SATA drive out for another. I just disabled it (the drive) for now. It works fine without it connected...that tells me its the drive and not the controller.

I began by trying to boot it up by using a Ubuntu LiveCD. It wouldn't boot up and showed many ATA-based errors in the logs. I tried a different Ubuntu LiveCD (v8.10, I believe)...still, same issue.

It was then that I started focusing on the SATA drive. I just unplugged it and tried to reboot...got a reboot and the system has been running for about a week without any issues.

So, I lost a drive. It's not a big deal, as that drive was hosting Windows, I believe. Weird, because that drive is rather young. I believe its a WD (I have seriously bad luck with that brand). I can do without replacing that drive for now, though.

Youch! Freenode bans Mibbit.com connections

New freenode webchat (and why to use it)

OW.

I remember awhile back, the server opers would frown upon banning Mibbit users. I guess they had a change of heart due to abusers using Mibbit to dodge bans. I remember having to place broad bans on Mibbit on a temp basis, but it was always temporary bans that I placed. I even became a Mibbit user. I hadn't joined the server using Mibbit in about a month so something happened recently for Freenode to lock out all Mibbit connections.

Oh well.

For now, Freenode is hosting its own web-IRC client: http://webchat.freenode.net/

Researching and found an old flamefest spark

Reference:

http://mythtv.beirdo.ca/ircLog/channel/1/2008-07-14

Summary: At LQ.org, there was a discussion on the security forums on how vulnerable Linux was to attacks/malware. Someone didn't like what was being discussed because of typical Linux zealotry. What happened on LQ's forums spilled over into ##slackware on IRC. Dagmar, the instigator of a LOT of bad things that used to happen in ##slackware got perm banned by me. Later, documented in the link above, he is his typical self, not even attempting to objectively explain what the whole thing was about, pretty much slandering me about how flawed my thoughts are on the whole thing and is worrying that I'll propagate bad information.

Let me explain some things about myself. I'm an IT security engineer. I don't just mess with routers and I'm not some glorified network engineer. I'm a senior consultant. I not only consult, I'm able to find "needle-in-the-haystack"-type info using packet-level analysis. Most of what I do requires that I be a jack-of-all-trades in network engineering, but my specialty is security. I'm proficient in utilizing many industry-leading security tools, both freeware and commercial software. I work at a very large ISP/telecom within a large managed security services team. I am THE lead of a government security operations center. We manage well over 100 customers' security posture via firewalls, NIDS, HIDS, and IPS appliances, using ArcSight, an aggregation and correlation tool that is fast becoming the standard in security event monitoring.

Every day, we see machines being compromised...this is nothing new. The compromises span every mainstream OS. This includes Linux. Whether it is kernel level or application level is not the argument. The argument is that Linux is not as rock-solid as everyone makes it out to be. Sure, it has more safeguards than Windows-based systems, but it is still susceptible to application-level exploits. Whether this is a coder issue or PEBKAC/user/admin issue is besides the point.

People need to stop thinking that just because they are running Linux, they are safe. That is NOT the case. This is not paranoia speaking. It is from seeing such things happen on a daily basis during security event monitoring. Due to applications such as PHP-Nuke, it is becoming more difficult to secure back end applications. It is much harder to stop SQL injection than it is to stop SSH brute-forcing, for instance. This isn't the only issue, though. The issue is the perception that because Linux code is open and free, the code base is free of vulnerabilities. That is NOT the case. Also, many people think that a majority of the cracker focus is on Win32 because MS has a majority of the market share. That also is NOT the case. That is a big assumption. milw0rm and other such sites document many *nix-based vulnerabilities, along with Bugtraq at Securityfocus track all vulnerabilities. Sometimes, people justify Linux because its security model is better focused than Win32 systems. It is, but that does not mean that Linux is rock-solid. It has its own faults, whether it is the user, the admin, or the software developer (or even kernel developer).

Dagmar has a habit of blocking out people's opinions and sometimes beating people down with his own. Dagmar thinks he knows security more than anyone else when he's just a developer. I see attacks every day on all types of machines. Some of the attacks are successful. I doubt that Dagmar sees those. Dagmar need not worry about me "propagating" untruth, because what I say IS the truth. All you have to do to see the truth is to research and not be blind to other opinions.

Dagmar also stalked. After the IRC discussion, he began to frequent the LQ security forums and respond to every thread I posted to. He was hardly ever in those forums before then. I noticed this immediately (and also checked). I didn't mind this, but when it spilled back over into IRC, I tired of it and wanted it ended...it really had no place in ##slackware and I was fed up with his attitude about the whole thing. I don't suffer drama very well.

Now, Dagmar has been banned several times before for the lack of tact in the way he 'helped' people in ##slackware. He was walking a thin line to begin with. Those with operator status in ##slackware acknowledge that he is knowledgeable, but that is not grounds for him to be dismissed as an abusive ##slackware visitor. Sure enough, he did the same thing with a channel operator (me) and I banned him. I also discussed it with the other operators. The consensus was that he stay banned since his history of being banned was substantial.

That was why he got banned...not because his views went against my own, but because he started regressing back to his former self and became abusive. He did the same in the LQ.org forums, but I was able to filter his posts from my normal views. As an operator at Freenode.net, I can't and shouldn't filter any visitor from my views in ##slackware, so my only option was to ban him, and like I said before, he'd his own infamous nature that was going against him.

As a security consultant, I'm certainly not going to keep my thoughts quiet about what I think is a disservice to my favorite operating system. I certainly know more than someone who is not a security consultant about IT security...its what I get paid to do and its what I've been doing for years. It's the same as a person who has built his own car, vs. someone who works as a senior Mercedes mechanic.

As much as I can, I tell people that there is NO secure OS. It is only as secure as the admin makes it, and even if the admin puts 100% resources into hardening the box, it will never be 100% secure. The LQ security forums is itself proof that Linux systems get compromised more than most people think. 2-3 times a week, someone reports they've been compromised. There's even 4 threads on Linux-based vulnerabilities:

Kernel Vulns
Mozilla Firefox Vulns
The Problem with PHP Application Security
Failed SSH Login Attempts

I can post a ton of other links but why do this when there is Google?

No further issues with gOS so far

It has been a very good experience, so far.

In fact, I've also been leveraging Thunderbird...this is a first time for me (since Netscape Mail back in the 90s). It is very robust!

I've also aliased a few commands that I tend to use alot, mainly ssh commands that I use on remote hosts.

I've also found some decent background images that I've scaled (using Gimp) to 1024x600.

I've also been conducting my typical security audits (BASE and iptables and web server log perusing).

I've not used my Mac in like 3 weeks! I don't know if that's a good or bad thing (probably bad for the battery).

gOS v3.1 installed on Dell Mini 9

I took the plunge and installed gOS v3.1 Gadgets onto the Dell Mini 9.

The install went flawlessly.

The issues I have discovered so far:

1. Wireless would not work. I followed the instructions located at http://gosforums.org/viewtopic.php?f=21&t=48&p=203&hilit=broadcom#p223. I applied this fix (when I was using a cat5 connection) and it worked, so the wireless non-functionality is no longer an issue.

2. The Mini won't suspend when I close the lid. I can manually suspend, though. I'll hunt for a fix and apply it later.

I also just noticed that a swap partition was created and configured for use (automatically, when installing gOS). I already have a gig of physical RAM and I don't want to burn out my SSD card, so I'll disable it for now and consider a workaround if/when I need it.

Overall, this is a pretty solid distro and it is pretty cool to be able to use Google tools (this will save drive space and conserve the limited resources this machine has). The Gadgets can also be used offline, so I won't need to be connected to use them...now, that's cool.

Some screenshots:






EDIT:

Fixed the sound issue by following Step 4 of "Installing Ubuntu 8.10 on the Dell Inspiron Mini 9".

I turned off the swap partition by editing out the swap entry in /etc/fstab. I'll test to see how this impacts my install of gOS before removing the partition.

Power adapters and other news

Wouldn't it be nice if the laptop empire joined forces and standardized laptop power adapter cords? I've several laptop and each requires a dedicated cord. I forgot my cord for my Mini today...fortunately, it has a decent charge, but I now have to watch consumption of power. Good thing my Macbook is in my truck. May have to go get it!

Other news:

I'm still trying to script FW log parsing. I've pretty much nailed it for my BSD machine, but will need to edit what I have, as the script parses and adds IPs to a block list. Sometimes, I just want to parse and find the top 10 offenders. I also still need to do this on my Linux machines. I've a script that parses (for Linux) but it is very rudimentary. And, I still want to port the script to Perl or Python.

tcpdump, Dell Mini, and BASE

So, I'm wondering why tcpdump is missing from the default install of my Dell's Ubuntu...doesn't make sense. I was having issues with getting my wifi card associated with my WAP and wanted to see the packets leaving the wireless interface, so I tried to bring up tcpdump, but it wasn't available. I actually had to hook a cat5 cable to the Mini to get this package, just to troubelshoot. I noticed the same thing with Suse about a year ago.

Apparently, tcpdump was created on the permissive free software license, per Wikipedia. I don't know if this is actually GPL or a derivative of GPL. The manpage doesn't mention what license tcpdump falls under and I'm sometimes wary of Wikipedia, as I like to find the facts on my own to validate (or invalidate) internet claims.

I'll research this and post my findings here.

On another note, I found a very cool bag for my Dell Mini, at Dell's website. I'll try to post pics and a link soon (from my Macbook, as the Mini's keyboard slows me down a bit).

Lastly, I somehow broke access to my MySQL database, so now my snort sensors won't report to it. It's been down for about 2 weeks and I don't have the time to fix it. I'm going on vacation for my birthday and hope to have some personal (ie, QUIET) time to myself to fix this. I'll be visiting my parents for my birthday this weekend and will see about shelling in to fix it remotely.

Worked on...

Reconfigured mnwclient so that I can provide FW logs to MyNetworkWatchman (which is similar to Dshield).

I'd much rather get Dshield working on the Linode but for some reason, I been having difficulties using their supplied clients. I'll continue to work on it, as I had it working prior to the last Linode upgrade.

With that in mind, at some time I'm going to have to upgrade the Linode from v12.0 to v12.2.

Night...

What I've installed on the Mini so far...

So far, I've installed the following on my Dell Mini:

tcpdump (IMO, this is a mandatory package...should've been installed by default)
xchat
gdesklets (I'm going to remove this because desktop space is at a premium)

That's it, so far. Surprisingly, I find myself not needing much more than what apps are already installed. Then again, I've had the Mini for less than a week. :)

EDIT: also installed 'locate' but the install didn't include the updatedb.conf file...working on that now.

My Mini is here!

My Dell Mini has arrived. Went to pick it up at the local Fedex facility last night.

It came in a LITTLE box. Basically, it was just the netbook and the adapter that were in the box, with a few CDs and documents.

I took it out and plugged it up and the battery was almost fully charged already, but I charged it anyways.

The keypad is small and my big hands don't help much but the unit itself is pretty slick.

The tech specs are below:

32gb SSD drive
1.3mb vidcam
1gb RAM
Ubuntu Linux OS
Black in color

I opted for no bluetooth but now I'm thinking maybe I should've went that route (although it would've jacked up the price.

I now need a case. For now, I'm carrying it in my work laptop's bag.

My immediate problem is that it won't connect to any wireless networks, either at home or at work. I did a quick search and this appears to be a known issue. I'll have to delve more into this. When I attempt to have it automatically apply the wireless setting and check those settings, the settings are blank (no IP or DNS). Manually applying them doesn't help.

So, I'll be occupied the next few days... :)

Dell Mini 9 and system upgrades

I just ordered a Dell Mini 9.

It should be here next week. I purchased the 16GB SSD upgrade and webcam. It is also a Windows XP model. Later, I'll more than likely either install Linux on it (or OS X), although this will require a USB optical drive.

The plan is to also upgrade the SSD to 32GB (a $119 purchase at My Digital Discount).

More than likely, I'm going to sell my Macbook (and maybe use the money to get an iMac).

As for my dead motherboard, it looks like I'm looking at this. I've decided to go AMD Phenom. Why? Because, while I want to upgrade to quad core while I can, I don't want to go Core i7 because I'd have to spend butt-loads on the CPU and motherboard in addition to DDR3 RAM...too much upgrading. At this point, I don't even want to go Intel. While Intel has better bang for the buck regarding CPU power and robustness, I'm extremely loyal to AMD. There's nothing wrong with AMD's products if you don't benchmark (and I'm not one to hang his every decision on raw stats alone). I should have the parts within a month...then I can game again. This will leave me with an unused Pentium D 830 (dual core), though. That means I'll more than likely buy a cheaper motherboard just to utilize that CPU in one of my older systems...it is a nice CPU with good power.

EDIT: I upgraded my Mini order because they were offering a $50 discount on Presidents Day. It will have Ubuntu and it will also have a 32GB SSD. More than likely I'm going to send it back, though, since Asus has a better buy, sans the SSD, but also has BT.

About my mainboard and CPU upgrade. I went AMD. I bought a Phenom 940 with a Foxconn mainboard. It is up and running now. It is FAST and I've yet to see all 4 cores maxed out! I also bought another mainboard for the Pentium CPU. That's a project for another day, though.

Foxconn A79A-S AM2+/AM2 AMD 790FX ATX AMD Motherboard

AMD Phenom II X4 940 Deneb 3.0GHz Socket AM2+ 125W Quad-Core Black Edition Processor Model HDZ940XCGIBOX

System dead

Ref: http://slackfiles.blogspot.com/2007_02_18_archive.html

OK. Apparently when the Ultra PS died, it may have taken the motherboard with it. The machine has been blue-screening ever since the PS replacement, just not as frequently as before.

I also noticed that the HSF on the Northbridge isn't working. When searching Google for issues with the Northbridge, I found that this is happening to others and that it actually signifies a motherboard failure. The Northbridge gets HOT and I'm assuming that over time, the heat has killed the motherboard.

This is the time to upgrade, I guess. I'll more than likely buy a new motherboard/CPU combo and it'll more than likely be a quad-core CPU that I'll be getting (Phenom):

(Intel) http://www.newegg.com/Product/ComboDealDetails.aspx?ItemList=Combo.152153

or

(AMD) http://www.newegg.com/Product/ComboDealDetails.aspx?ItemList=Combo.150323

EDIT - I ended up purchasing a

Foxconn A79A-S AM2+/AM2 AMD 790FX ATX AMD Motherboard

and a

Retail AMD Phenom II X4 940 Deneb 3.0GHz 4 x 512KB L2 Cache CPU

and basically rebuilt this system...it is SMOKING. I was gaming and installing game patches for another game AND XP updates, and only two cores were being used while two were idle!! I put 3GB of RAM into the system (I almost put 4GB but then realized that XP 32-bit would only see 3GB of it...time for a 64-bit Windows OS install, I think). I really should install some high-grade RAM, though, as I think I only have 667mhz installed.

I also bought a

BIOSTAR TForce TP43D2A7 LGA 775 Intel P43 ATX Intel Motherboard

for to replace the motherboard in the Pentium-D system...I'll use this motherboard and CPU to rebuild my old Sempron 3000+ system (I'm starting to stack up 'older' motherboards and CPUs).

Recent Happenings

I've added 2 more GB to my gaming system (the Ultra that I bought from TigerDirect 2 yrs ago). The system now has a total of 4GB. It runs Windows XP, so its only seeing 3.25GB of the installed 4GB, as I'm running XP 32-bit. I need to either go to Vista or install XP 64-bit. I may just wait until the new version of Windows desktop is released.

I've also begun to (again) monitor my internal LAN traffic at home. I broke down my security systems this past Fall because I bought a house and had to move. I wanted to reinstall my external IDS but didn't want to run an ethernet tap yet, so I used the same machine to just log internal traffic for now. Later, I'll have this machine start a separate Snort process that will monitor external traffic via the tap.

I need to upgrade my Slackware boxes to v12.2. The current boxes are running v12.0, including my Linode (http://wigglit.ath.cx). I'll try to do this soon, although this will require upgrading from v12.0 to v12.1, then v12.1 to v12.2.

I'll also begin to attempt to convert all (which isn't many) my BASH scripts that manipulate text (FW log parsing scripts and such) to Perl. This is mainly as an exercise to force myself to learn Perl. I could also do Python. This is an educational 2009 goal for me. :)

Distributed SSH Brute Force Attempts, part 3

I wanted to be able to harvest the log data that the brute force attempts are generating, so I've decided to not move the SSH listening port. I'm also logging each event within my firewall logs. The particular machine I'm seeing the attacks on is a FreeBSD box (I should've mentioned that earlier) and I'm using PF as the firewall of choice.

I'd like to show you my listing of blocked IPs. I've been actively gathering them since approximately 11/17/2008. I was unhappy that I hadn't noticed the distributed attempts until November and wanted more trending data, so I reached into my SSH logs and parsed the files present with a very dirty script that added each unique IP to a PF table that is designed to block such activity. The script is below:

[root@delly ~]# cat IPscript
#/bin/bash

cd /var/log
bunzip2 pflog.*

cat /var/log/auth.log | grep sshd | grep -i 'invalid user' > /tmp/auth_IP_list_1
bzcat /var/log/auth.log.[01234567].bz2 | grep sshd| grep -i 'invalid user' >> /tmp/auth_IP_list_1
sed '/Failed keyboard-interactive/d' /tmp/auth_IP_list_1 > /tmp/auth_IP_list_2
awk '{print $10}' /tmp/auth_IP_list_2 > /tmp/auth_IP_list_3

tcpdump -nettttr /var/log/pflog > /tmp/fw_IP_list_1
tcpdump -nettttr /var/log/pflog.0 >> /tmp/fw_IP_list_1
awk '{print $9}' /tmp/fw_IP_list_1 > /tmp/fw_IP_list_2
nawk -F. '{print $1, $2, $3, $4}' /tmp/fw_IP_list_2 > /tmp/fw_IP_list_3
sed 's/ /./g' /tmp/fw_IP_list_3 > /tmp/fw_IP_list_4

cat /tmp/fw_IP_list_4 >> /tmp/auth_IP_list_4
cat /tmp/auth_IP_list_4 | sort -rn | uniq > /tmp/auth_IP_list_5
cat /tmp/auth_IP_list_5 | grep -v '64.62.231.220' > /tmp/auth_IP_list_6
cat /tmp/auth_IP_list_6 | grep -v '66.160.141.30' > /tmp/auth_IP_list_7
cat /tmp/auth_IP_list_7 | grep -v '10.150.1' > /tmp/auth_IP_list_8
cat /tmp/auth_IP_list_8
cat /tmp/auth_IP_list_8 | wc -l

pfctl -t bruteforce -T add -f /tmp/auth_IP_list_8

rm -rf /tmp/fw_IP_list_* /tmp/auth_IP_list_*

I apologize for the lack of comments in the script...as I said, it was an extremely dirty hack that required me to learn a tad of sed and awk. The script outputs the following file: http://wigglit.ath.cx/txt/bruteforce_IPtable

My logs don't actually go back that far. The FW logs go back to 14 Dec. The SSH service logs go back to 26 Nov. I probably have captured IPs reaching back to the beginning of November, though.

My FW policy prevents anything from entering the network unless specifically allowed (default deny policy). The reason I wanted to track the IPs was because my Denyhosts configuration wasn't catching most of these and it was working fine before this new trend occurred. I'm a security consultant and a researcher at heart, so I thought that tracking this would be cool. My home router has ports 22, 443, and 3306 exposed to the wild for the FreeBSD box. The FW only allows certain IPs in on those ports, though, so nothing will get in. The activity that doesn't get immediately blocked is blocked by the bruteforce_IPtable script that I run daily. The script is flawless and I will begin to have it run hourly via a cronjob. I'll also have it send an updated list to my website daily.

The IPs within my block table number 565. The script parsed 115 from the logs tonight, but only added 1 IP. The norm is usually 2-3 daily. The rest of the IPs are from logs over the last month and a half.

Oh yeah, I've another script does a daily copy of the IPs that are added to the table, so I can at least quickly determine (using 'diff') what was added on a certain day. I can create a script that will show me what was added daily for a given timespan (day/week/month).

Distributed SSH Brute Force Attempts, part 2

OK, so I'm still curious about the distributed SSH brute force attempts I've been seeing.

I wanted to check all my logs and not just the most recent logfile. Looking at my /var/log directory:

-su-2.05b# ls | grep auth
auth.log
auth.log.0.bz2
auth.log.1.bz2
auth.log.2.bz2
auth.log.3.bz2
auth.log.4.bz2
auth.log.5.bz2
auth.log.6.bz2
auth.log.7.bz2

I tailed the auth.log file and grabbed a random IP:

Nov 24 21:53:59 delly sshd[75490]: Invalid user bryan from 170.56.255.20
Nov 24 21:54:00 delly sshd[75490]: error: PAM: authentication error for illegal user bryan from 170.56.255.20
Nov 24 21:54:00 delly sshd[75490]: Failed keyboard-interactive/pam for invalid user bryan from 170.56.255.20 port 43229 ssh2

I then wanted to check all the auth.log.* files, but was curious as to how I could check compressed files. I found that there's a command called bzgrep that allows one to grep compressed files, so I used the following command and came up with quite a few hits for the referenced IP over seven (7) log files:

-su-2.05b# bzgrep '170.56.255.20' auth.log.*

The results show 2-3 instances of log entries per login attempt, so I wanted to isolate each instance without having to use arcane sed and sort commands, so I used the following:

-su-2.05b# bzgrep '170.56.255.20' auth.log.* | grep 'Invalid user' | wc -l
19

So, this particular IP generated 19 log entries between 22 and 24 Nov.

That's not particularly good. I'd love to create a script that would break down all of these IPs' unique login attempts (and possibly block them). Sounds like a project, no? :)

I've a bit of time, so I did the following (non-scripted). I cat'd the auth.log file and collected a screens-worth of data:

-su-2.05b# cat auth.log | less
Nov 24 15:00:00 delly newsyslog[74001]: logfile turned over due to size>100K
Nov 24 15:00:49 delly sshd[74014]: Invalid user brand from 218.80.215.198
Nov 24 15:00:50 delly sshd[74014]: error: PAM: authentication error for illegal user brand from 218.80.215.198
Nov 24 15:00:50 delly sshd[74014]: Failed keyboard-interactive/pam for invalid user brand from 218.80.215.198 port 19051 ssh2
Nov 24 15:02:25 delly sshd[74017]: Invalid user brandee from 83.19.224.11
Nov 24 15:02:26 delly sshd[74017]: error: PAM: authentication error for illegal user brandee from dum11.internetdsl.tpnet.pl
Nov 24 15:02:26 delly sshd[74017]: Failed keyboard-interactive/pam for invalid user brandee from 83.19.224.11 port 50163 ssh2
Nov 24 15:03:48 delly sshd[74020]: Invalid user brandee from 194.224.118.61
Nov 24 15:03:48 delly sshd[74020]: error: PAM: authentication error for illegal user brandee from 194.224.118.61
Nov 24 15:03:48 delly sshd[74020]: Failed keyboard-interactive/pam for invalid user brandee from 194.224.118.61 port 6345 ssh2
Nov 24 15:05:21 delly sshd[74026]: Invalid user brandee from 90.176.233.222
Nov 24 15:05:22 delly sshd[74026]: error: PAM: authentication error for illegal user brandee from 222.233.broadband9.iol.cz
Nov 24 15:05:22 delly sshd[74026]: Failed keyboard-interactive/pam for invalid user brandee from 90.176.233.222 port 46108 ssh2
Nov 24 15:06:42 delly sshd[74029]: Invalid user branden from 125.77.106.246
Nov 24 15:06:42 delly sshd[74029]: error: PAM: authentication error for illegal user branden from 125.77.106.246
Nov 24 15:06:42 delly sshd[74029]: Failed keyboard-interactive/pam for invalid user branden from 125.77.106.246 port 46495 ssh2
Nov 24 15:10:59 delly sshd[74035]: Invalid user brandi from 122.224.128.222
Nov 24 15:10:59 delly sshd[74035]: error: PAM: authentication error for illegal user brandi from 122.224.128.222
Nov 24 15:10:59 delly sshd[74035]: Failed keyboard-interactive/pam for invalid user brandi from 122.224.128.222 port 42253 ssh2
Nov 24 15:12:27 delly sshd[74051]: Invalid user brandi from 59.125.200.51
Nov 24 15:12:28 delly sshd[74051]: error: PAM: authentication error for illegal user brandi from 3w.upcc.com.tw
Nov 24 15:12:28 delly sshd[74051]: Failed keyboard-interactive/pam for invalid user brandi from 59.125.200.51 port 14046 ssh2
Nov 24 15:15:23 delly sshd[74057]: Invalid user brandice from 62.112.222.88
Nov 24 15:15:24 delly sshd[74057]: error: PAM: authentication error for illegal user brandice from 3e70de58.adsl.enternet.hu
Nov 24 15:15:24 delly sshd[74057]: Failed keyboard-interactive/pam for invalid user brandice from 62.112.222.88 port 42127 ssh2
Nov 24 15:16:49 delly sshd[74060]: Invalid user brandice from 218.80.215.198
Nov 24 15:16:50 delly sshd[74060]: error: PAM: authentication error for illegal user brandice from 218.80.215.198
Nov 24 15:16:50 delly sshd[74060]: Failed keyboard-interactive/pam for invalid user brandice from 218.80.215.198 port 57929 ssh2
Nov 24 15:18:11 delly sshd[74063]: Invalid user brandice from 65.203.231.41
Nov 24 15:18:11 delly sshd[74063]: error: PAM: authentication error for illegal user brandice from 65.203.231.41
Nov 24 15:18:11 delly sshd[74063]: Failed keyboard-interactive/pam for invalid user brandice from 65.203.231.41 port 38395 ssh2
Nov 24 15:19:43 delly sshd[74066]: Invalid user brandie from 123.14.10.64
Nov 24 15:19:44 delly sshd[74066]: error: PAM: authentication error for illegal user brandie from 123.14.10.64
Nov 24 15:19:44 delly sshd[74066]: Failed keyboard-interactive/pam for invalid user brandie from 123.14.10.64 port 4925 ssh2
Nov 24 15:21:07 delly sshd[74072]: Invalid user brandie from 200.170.141.134
Nov 24 15:21:07 delly sshd[74072]: error: PAM: authentication error for illegal user brandie from 200-170-141-134.static.ctbctelecom.com.br
Nov 24 15:21:07 delly sshd[74072]: Failed keyboard-interactive/pam for invalid user brandie from 200.170.141.134 port 39979 ssh2
Nov 24 15:22:44 delly sshd[74088]: Invalid user brandie from 80.51.31.84
Nov 24 15:22:44 delly sshd[74088]: error: PAM: authentication error for illegal user brandie from 80.51.31.84
Nov 24 15:22:44 delly sshd[74088]: Failed keyboard-interactive/pam for invalid user brandie from 80.51.31.84 port 39453 ssh2
Nov 24 15:24:02 delly sshd[74091]: Invalid user brandon from 200.157.176.13
Nov 24 15:24:03 delly sshd[74091]: error: PAM: authentication error for illegal user brandon from 200.157.176.13
Nov 24 15:24:03 delly sshd[74091]: Failed keyboard-interactive/pam for invalid user brandon from 200.157.176.13 port 54638 ssh2

I then checked every IP for unique log entries within all of my ssh logs:

-su-2.05b# bzgrep '218.80.215.198' auth.log.* | grep 'Invalid user' | wc -l
6

-su-2.05b# bzgrep '218.80.215.198' auth.log.* | grep 'Invalid user'
auth.log.0.bz2:Nov 24 10:55:25 delly sshd[73084]: Invalid user bjorn from 218.80.215.198
auth.log.1.bz2:Nov 23 23:22:28 delly sshd[70023]: Invalid user bahari from 218.80.215.198
auth.log.3.bz2:Nov 23 11:18:54 delly sshd[66908]: Invalid user archibald from 218.80.215.198
auth.log.4.bz2:Nov 23 04:11:52 delly sshd[65051]: Invalid user amy from 218.80.215.198
auth.log.5.bz2:Nov 22 23:20:53 delly sshd[63465]: Invalid user alize from 218.80.215.198
auth.log.7.bz2:Nov 22 01:07:10 delly sshd[57652]: Invalid user claire from 218.80.215.198

-su-2.05b# bzgrep '83.19.224.11' auth.log.* | grep 'Invalid user' | wc -l
8
-su-2.05b# bzgrep '194.224.118.61' auth.log.* | grep 'Invalid user' | wc -l
11
-su-2.05b# bzgrep '90.176.233.222' auth.log.* | grep 'Invalid user' | wc -l
0
-su-2.05b# bzgrep '125.77.106.246' auth.log.* | grep 'Invalid user' | wc -l
9
-su-2.05b# bzgrep '122.224.128.222' auth.log.* | grep 'Invalid user' | wc -l
8
-su-2.05b# bzgrep '59.125.200.51' auth.log.* | grep 'Invalid user' | wc -l
5
-su-2.05b# bzgrep '62.112.222.88' auth.log.* | grep 'Invalid user' | wc -l
10
-su-2.05b# bzgrep '218.80.215.198' auth.log.* | grep 'Invalid user' | wc -l
6
-su-2.05b# bzgrep '65.203.231.41' auth.log.* | grep 'Invalid user' | wc -l
15

-su-2.05b# bzgrep '65.203.231.41' auth.log.* | grep 'Invalid user'
auth.log.0.bz2:Nov 24 14:56:28 delly sshd[73982]: Invalid user bran from 65.203.231.41
auth.log.1.bz2:Nov 23 22:32:01 delly sshd[69793]: Invalid user azra from 65.203.231.41
auth.log.1.bz2:Nov 24 02:04:43 delly sshd[70680]: Invalid user bartholemew from 65.203.231.41
auth.log.1.bz2:Nov 24 04:07:45 delly sshd[71474]: Invalid user beck from 65.203.231.41
auth.log.2.bz2:Nov 23 19:02:03 delly sshd[68866]: Invalid user aurora from 65.203.231.41
auth.log.2.bz2:Nov 23 20:16:18 delly sshd[69213]: Invalid user avi from 65.203.231.41
auth.log.3.bz2:Nov 23 10:08:33 delly sshd[66592]: Invalid user april from 65.203.231.41
auth.log.3.bz2:Nov 23 10:24:43 delly sshd[66657]: Invalid user aquila from 65.203.231.41
auth.log.3.bz2:Nov 23 11:22:12 delly sshd[66933]: Invalid user archie from 65.203.231.41
auth.log.4.bz2:Nov 23 04:22:52 delly sshd[65094]: Invalid user anahid from 65.203.231.41
auth.log.4.bz2:Nov 23 05:32:46 delly sshd[65407]: Invalid user andra from 65.203.231.41
auth.log.5.bz2:Nov 22 23:00:09 delly sshd[63393]: Invalid user alisha from 65.203.231.41
auth.log.6.bz2:Nov 22 12:41:17 delly sshd[60534]: Invalid user abraham from 65.203.231.41
auth.log.6.bz2:Nov 22 16:14:07 delly sshd[61564]: Invalid user africa from 65.203.231.41
auth.log.7.bz2:Nov 22 11:48:33 delly sshd[60289]: Invalid user aaralyn from 65.203.231.41

-su-2.05b# bzgrep '123.14.10.64' auth.log.* | grep 'Invalid user' | wc -l
19
-su-2.05b# bzgrep '200.170.141.134' auth.log.* | grep 'Invalid user' | wc -l
6
-su-2.05b# bzgrep '80.51.31.84' auth.log.* | grep 'Invalid user' | wc -l
3
-su-2.05b# bzgrep '200.157.176.13' auth.log.* | grep 'Invalid user' | wc -l
4

So, someone appears to have a pool of compromised machines and is using each one in a scaled SSH brute force attack, based on the referenced user accounts being bruteforced. I'm seeing more of this than standard, blatant SSH BF attempts. I'll be checking Denyhosts' website to see if they've a resolution on how to track and ban such activity.

Distributed SSH Brute Force Attempts?

I'd read not long ago on the ISC Diary that someone has noticed that a there's a newly discovered way to avoid automated tools such as Denyhosts and Fail2ban. It appears that the attacks are now distributed across an IP pool of compromised machines. Maybe botnet masters are leveraging their botnets to attempt to bruteforce login attempts without risking the attacking hosts.

I think I'm seeing this in my home firewall logs:

Nov 23 15:23:01 delly sshd[67946]: error: PAM: authentication error for illegal user artois from 1-1-4-27a.vhe.sth.bostream.se
Nov 23 15:23:01 delly sshd[67946]: Failed keyboard-interactive/pam for invalid user artois from 82.182.188.187 port 35763 ssh2
Nov 23 15:24:18 delly sshd[67949]: Invalid user arty from 58.26.48.162
Nov 23 15:24:18 delly sshd[67949]: error: PAM: authentication error for illegal user arty from 58.26.48.162
Nov 23 15:24:18 delly sshd[67949]: Failed keyboard-interactive/pam for invalid user arty from 58.26.48.162 port 5785 ssh2
Nov 23 15:25:17 delly sshd[67955]: Invalid user arty from 200.170.141.134
Nov 23 15:25:17 delly sshd[67955]: error: PAM: authentication error for illegal user arty from 200-170-141-134.static.ctbctelecom.com.br
Nov 23 15:25:17 delly sshd[67955]: Failed keyboard-interactive/pam for invalid user arty from 200.170.141.134 port 57360 ssh2
Nov 23 15:26:31 delly sshd[67958]: Invalid user arty from 219.76.222.27
Nov 23 15:26:31 delly sshd[67958]: error: PAM: authentication error for illegal user arty from n219076222027.netvigator.com
Nov 23 15:26:31 delly sshd[67958]: Failed keyboard-interactive/pam for invalid user arty from 219.76.222.27 port 47176 ssh2
Nov 23 15:28:48 delly sshd[67963]: Invalid user arva from 58.196.4.2
Nov 23 15:28:49 delly sshd[67963]: error: PAM: authentication error for illegal user arva from 58.196.4.2
Nov 23 15:28:49 delly sshd[67963]: Failed keyboard-interactive/pam for invalid user arva from 58.196.4.2 port 50637 ssh2
Nov 23 15:33:27 delly sshd[67982]: Invalid user arvid from 125.77.106.246
Nov 23 15:33:27 delly sshd[67982]: error: PAM: authentication error for illegal user arvid from 125.77.106.246
Nov 23 15:33:27 delly sshd[67982]: Failed keyboard-interactive/pam for invalid user arvid from 125.77.106.246 port 51673 ssh2
Nov 23 15:34:40 delly sshd[67985]: Invalid user arvin from 85.39.252.226
Nov 23 15:34:40 delly sshd[67985]: error: PAM: authentication error for illegal user arvin from host226-252-static.39-85-b.business.telecomitalia.it
Nov 23 15:34:40 delly sshd[67985]: Failed keyboard-interactive/pam for invalid user arvin from 85.39.252.226 port 43706 ssh2
Nov 23 15:35:54 delly sshd[67991]: Invalid user arvin from 217.126.90.161
Nov 23 15:35:55 delly sshd[67991]: error: PAM: authentication error for illegal user arvin from 161.red-217-126-90.staticip.rima-tde.net
Nov 23 15:35:55 delly sshd[67991]: Failed keyboard-interactive/pam for invalid user arvin from 217.126.90.161 port 36755 ssh2
Nov 23 15:37:11 delly sshd[67994]: Invalid user arvin from 200.232.181.40
Nov 23 15:37:11 delly sshd[67994]: error: PAM: authentication error for illegal user arvin from 200-232-181-40.dsl.telesp.net.br
Nov 23 15:37:11 delly sshd[67994]: Failed keyboard-interactive/pam for invalid user arvin from 200.232.181.40 port 56318 ssh2
Nov 23 15:39:16 delly sshd[67997]: Invalid user arwan from 200.248.82.130
Nov 23 15:39:17 delly sshd[67997]: error: PAM: authentication error for illegal user arwan from 200.248.82.130
Nov 23 15:39:17 delly sshd[67997]: Failed keyboard-interactive/pam for invalid user arwan from 200.248.82.130 port 53388 ssh2
Nov 23 15:40:24 delly sshd[68003]: Invalid user arwan from 217.126.90.161
Nov 23 15:40:25 delly sshd[68003]: error: PAM: authentication error for illegal user arwan from 161.red-217-126-90.staticip.rima-tde.net
Nov 23 15:40:25 delly sshd[68003]: Failed keyboard-interactive/pam for invalid user arwan from 217.126.90.161 port 43871 ssh2
Nov 23 15:41:33 delly sshd[68006]: Invalid user arwen from 200.209.6.130
Nov 23 15:41:34 delly sshd[68006]: error: PAM: authentication error for illegal user arwen from 200.209.6.130
Nov 23 15:41:34 delly sshd[68006]: Failed keyboard-interactive/pam for invalid user arwen from 200.209.6.130 port 14808 ssh2
Nov 23 15:42:48 delly sshd[68016]: Invalid user arwen from 123.14.10.64
Nov 23 15:42:49 delly sshd[68016]: error: PAM: authentication error for illegal user arwen from 123.14.10.64
Nov 23 15:42:49 delly sshd[68016]: Failed keyboard-interactive/pam for invalid user arwen from 123.14.10.64 port 7600 ssh2
Nov 23 15:43:50 delly sshd[68020]: reverse mapping checking getaddrinfo for techregister.worcesteracademy.org [68.112.227.30] failed - POSSIBLE BREAK-IN ATTEMPT!
Nov 23 15:43:50 delly sshd[68020]: Invalid user arwen from 68.112.227.30
Nov 23 15:43:50 delly sshd[68020]: error: PAM: authentication error for illegal user arwen from 68.112.227.30
Nov 23 15:43:50 delly sshd[68020]: Failed keyboard-interactive/pam for invalid user arwen from 68.112.227.30 port 38273 ssh2
Nov 23 15:45:03 delly sshd[68039]: Invalid user arya from 196.28.50.162
Nov 23 15:45:03 delly sshd[68039]: error: PAM: authentication error for illegal user arya from www.cfse.gov.pr
Nov 23 15:45:03 delly sshd[68039]: Failed keyboard-interactive/pam for invalid user arya from 196.28.50.162 port 55647 ssh2
Nov 23 15:46:10 delly sshd[68042]: Invalid user arya from 81.12.221.74
Nov 23 15:46:10 delly sshd[68042]: error: PAM: authentication error for illegal user arya from em.asiban.ro
Nov 23 15:46:10 delly sshd[68042]: Failed keyboard-interactive/pam for invalid user arya from 81.12.221.74 port 16653 ssh2
Nov 23 15:47:23 delly sshd[68050]: Invalid user arya from 190.34.148.178
Nov 23 15:47:23 delly sshd[68050]: error: PAM: authentication error for illegal user arya from 190.34.148.178
Nov 23 15:47:23 delly sshd[68050]: Failed keyboard-interactive/pam for invalid user arya from 190.34.148.178 port 58738 ssh2
Nov 23 15:48:29 delly sshd[68053]: reverse mapping checking getaddrinfo for britannic-iss-medidean-working.e1-4-0-0-57.0.ar2.lon3.gblx.net [64.213.54.106] failed - POSSIBLE BREAK-IN ATTEMPT!
Nov 23 15:48:29 delly sshd[68053]: Invalid user asa from 64.213.54.106
Nov 23 15:48:29 delly sshd[68053]: error: PAM: authentication error for illegal user asa from 64.213.54.106
Nov 23 15:48:29 delly sshd[68053]: Failed keyboard-interactive/pam for invalid user asa from 64.213.54.106 port 42991 ssh2
Nov 23 15:49:46 delly sshd[68056]: Invalid user asa from 91.135.200.86
Nov 23 15:49:47 delly sshd[68056]: error: PAM: authentication error for illegal user asa from 91.135.200.86
Nov 23 15:49:47 delly sshd[68056]: Failed keyboard-interactive/pam for invalid user asa from 91.135.200.86 port 10262 ssh2
Nov 23 15:50:52 delly sshd[68062]: Invalid user asa from 200.20.187.222
Nov 23 15:50:53 delly sshd[68062]: error: PAM: authentication error for illegal user asa from 200.20.187.222
Nov 23 15:50:53 delly sshd[68062]: Failed keyboard-interactive/pam for invalid user asa from 200.20.187.222 port 52959 ssh2

Don't focus on the attacking IPs, but look at the referenced users. There are now tools that look like they're scaling attacks on a listing of common logins (or maybe even dictionary attacks) so that there's less risk of detection. There are current tools that look for attacks in a thresholded manner (example: 4 attacks in 5 sec warrants a block of that attacking IP). This new method of attack will not trigger the thresholding blocks.

More than ever, SSH key-based authentication should be used. This will prevent a successful login when under attack via brute forcing methods.

I can already see attack detection tools being adjusted to focus on tracking user accounts being bruteforced and banning all IPs that try to access user accounts based on time (example: 4 attacks on account asa in 5 sec will warrant a ban of all subsequent IPs for the next day or so...and not block if the IP is listed within a whitelist).

Your thoughts?

Asus ships software cracker on recovery DVD

Asus ships software cracker on recovery DVD:

Asus is accidentally shipping software crackers and confidential documents on the recovery DVDs that come with its laptops.

The startling discovery was made by a PC Pro reader whose antivirus software was triggered by a key cracker for the WinRAR compression software, which was located on the recovery DVD for his Asus laptop.

Script that parses FW logs?

What I'm trying to do now is create a script that will parse FW logs daily and break down how many entries each IP generated.

I want to do this first as a quick command, then leverage BASH scripting to automate this.

What I used in http://slackfiles.blogspot.com/2007/12/modsecurity-again.html apparently isn't working when trying to parse the firewall logs. I don't get it, but then again, I'm tired. I'll try again tomorrow.

The new logs look like the below (showing in /var/log/syslog instead of /var/log/messages of my old setup):

Jul 8 22:55:19 starchild kernel: BLOCKED: IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:fc:8b:59:08:00 SRC=64.47.32.59 DST=64.62.231.220 LEN=48 TOS=0x00 PREC=0x00 TTL=249 ID=39747 DF PROTO=TCP SPT=3405 DPT=445 WINDOW=16384 RES=0x00 SYN URGP=0

I may need to leverage Awk (I was close to doing this before I upgraded the server from v9.0 to v12.0).

Stay tuned!

Linode has been upgraded to Slackware v12.0

I decided to take the plunge and upgrade my Linode server from v9.0 to v12.0.

Things are semi-normal, yet I still have some things to do, such as reinstall denyhosts, a new iptables script, mnwclient, and dshieldclient.

So far, I've reinstalled the eggdrop bot (slackboy), using the latest version of the software. I've also reinstalled Apache, using v2.0.63. Snort is running on the system (installed using my v12.0 slackpack), and also Webmin (which eases my pain somewhat, especially when administrating from work...they frown upon using SSH outbound).

This verion of Slackware is barebones and I've had to manually install the following packages so far:

-rw-r--r-- 1 root root 122953 2006-11-01 02:50 checkinstall-1.6.1-i386-1.tgz
-rw-r--r-- 1 root root 551952 2006-05-27 16:24 cyrus-sasl-2.1.22-i486-1.tgz
-rw-r--r-- 1 root root 744871 2007-05-09 14:46 fetchmail-6.3.8-i486-2.tgz
-rw-r--r-- 1 root root 2268306 2006-03-26 19:04 groff-1.19.2-i486-1.tgz
-rw-r--r-- 1 root root 2088735 2007-07-01 19:12 httpd-2.2.4-i486-6.tgz
-rw-r--r-- 1 root root 1898552 2008-02-14 17:37 httpd-2.2.8-i486-1_slack12.0.tgz
-rw-r--r-- 1 root root 313616 2007-06-02 18:52 lsof-4.78-i486-1.tgz
-rw-r--r-- 1 root root 222760 2006-02-10 20:03 man-1.6c-i486-2.tgz
-rw-r--r-- 1 root root 1949092 2007-06-13 13:35 man-pages-2.55-noarch-1.tgz
-rw-r--r-- 1 root root 135893 2004-02-18 05:59 metamail-2.7-i486-2.tgz
-rw-r--r-- 1 root root 16756453 2007-03-24 00:52 mysql-5.0.37-i486-1.tgz
-rw-r--r-- 1 root root 16942549 2007-12-14 17:01 mysql-5.0.51-i486-1_slack12.0.tgz
-rw-r--r-- 1 root root 1483847 2007-05-07 01:15 ntp-4.2.4p0-i486-1.tgz
-rw-r--r-- 1 root root 3440059 2007-06-13 13:40 openssl-0.9.8e-i486-3.tgz
-rw-r--r-- 1 root root 836941 2007-06-13 13:40 openssl-solibs-0.9.8e-i486-3.tgz
-rw-r--r-- 1 root root 143713 2006-09-19 00:11 procmail-3.22-i486-2.tgz
-rw-r--r-- 1 root root 2450931 2003-10-29 01:08 rpm-4.2.1-i486-3.tgz
-rw-r--r-- 1 root root 4603 2007-01-04 22:28 rpm2tgz-1.0-i486-1.tgz
-rw-r--r-- 1 root root 212764 2007-02-10 14:52 rsync-2.6.9-i486-1.tgz
-rw-r--r-- 1 root root 1391679 2007-06-10 01:16 sendmail-8.14.1-i486-1.tgz
-rw-r--r-- 1 root root 276485 2007-06-10 01:16 sendmail-cf-8.14.1-noarch-1.tgz
-rw-r--r-- 1 root root 30270 2006-04-19 00:56 slocate-3.1-i486-1.tgz
-rw-r--r-- 1 root root 763394 2008-06-15 15:30 snort-2.6.1.5-i386-1.tgz
-rw-r--r-- 1 root root 137062 2006-02-06 14:00 sudo-1.6.8p12-i486-1.tgz
-rw-r--r-- 1 root root 1793130 2007-06-02 19:32 tcl-8.4.15-i486-1.tgz
-rw-r--r-- 1 root root 66773 2007-04-30 00:35 telnet-0.17-i486-1.tgz
-rw-r--r-- 1 root root 8614263 2007-06-20 16:34 vim-7.1.012-i486-1.tgz
-rw-r--r-- 1 root root 1882278 2007-06-20 16:50 vim-gvim-7.1.012-i486-1.tgz
-rw-r--r-- 1 root root 13573880 2008-05-26 00:19 webmin-1.420.tar.gz

Stability-wise, the server appears to be running somewhat more efficiently than before (for instance, the CPU and memory utilizations, along with IO, appear to be less 'spiky' when looking at chart readouts. When I shut down the server to upgrade (which is the way its done with Linode.com), I'd had 440+ days of uptime, so I can't argue about stability...it doesn't get much better than that!

Sooner or later, I'm going to try backing up my latest install, then trying to upgrade to v12.1 (just because I can).

wigglit.ath.cx being bombarded with scans of port 1028/UDP

The culprit?

24.64.0.0/13, or 24.64.0.0 - 24.71.255.255, which resolves to SHAWCABLE.NET

There are at least 311 hosts within that range that have tried to connect to UDP port 1028 in the last few days. This isn't really a broad scan but the pepperings of hosts every day for the last few days, each one being unique hosts that have never been logged makes it hard to establish a pattern so that I can block remote hosts that continue to scan for this port, so I've opted to initiate a broad block and keep the block in place for maybe 30 days.

I'll monitor this activity and maybe alert the ISC diary if the scans continue.

If you're caught up in this ban, let me know and I'll see about allowing traffic to specific hosts.

Metaspolit hijacked?

Monday morning, Metasploit.com was temporarily hijacked using an attack on the local area network of Metasploit's hosting provider.

More info here

Slamd64

I've installed Slamd64 on my new AMD system. Initially, I had SATA issues that have mysteriously disappeared (no idea why, other than maybe swapping out that SATA cables helped). I installed using Disk 1 only (I didn't download anything else), which has the core system components. I then wanted to boot X, which required me downloading the X and KDE software. In fact, I went ahead and copied a whole mirror site and will continue to rsync the site against my local copy, using the updated local mirror as a upgrade repository.

I had issues getting my mouse recognized. I've a Logitech MX1000 wireless mouse, which is connected to a KVM. Slamd64 detects it as a PS/2 mouse. Slamd64 has PS/2 mice blacklisted, so I had to unblock the psmouse module from /etc/modules.d/blacklist (by uncommenting the module entry). Then I loaded the module and rebooted. The reboot detected the mouse. It took me a while to find this tidbit of info (although, it was recorded on the Slamd64 forums). I'll be adding this to my local knowledgbase (in fact, I need to add a whole Slamd64 category first).

I've also installed phpsysinfo on this machine, so I can see the hardware and how it is detected by this tool. From what I've seen so far, the second CPU core takes the brunt of the load, with the first CPU core assisting when the second is maxed out...I don't know if this is normal or a software issue (maybe phpsysinfo needs to be optimized for dual core usage?), but I only tested this by refreshing the phpsysinfo browser session (it appears to put a quick load on the system...dunno if that's normal or not).

Anyways, if things go well with this distro, I'll be using this machine as my main Linux machine, eventually.

AMD vs. Intel Comparison

About the new AMD system I recently bought...

The AMD Athlon 64 X2 4400+ rocks when compared to my Intel system, which uses a Pentium D830. The AMD CPU is running on an Abit NF-M2SV board with 1GB of RAM. The Intel system is running on an ECS nForce 570 SLIT-A v5.1 mainboard and 2GB of RAM. While the Intel system outguns the AMD system, spec-wise, the AMD system is quite a bit more responsive...the whole system seems and responds like it is extremely lightweight. Both are running XP Pro. The AMD CPU runs 104F temps on the average, while the Intel CPU is in the 120F range...and this system has the beefier CPU fan/heatsink too! The AMD CPU cooler is here. It is an Arctic Cooling Freezer 64 LP. The Intel CPU cooler is here and is apparently a standard issue HSF (Intel Socket 775 Cooling Fan, although it is badged as an Ultra unit).

I'm very impressed!

No, I don't have any benchmark specs (you can probably find these online via Google), but this is really a seat-of-the-pants comparison. That I noticed such a difference in this manner should speak for itself.

Now, I've split the 250GB drive on the AMD system in half, to test Slamd64, an unofficial 64-bit port of Slackware (which is a 32-bit OS). I've run into issues booting up Slamd64, though...I've been getting SATA-specific errors that hint that the hard drive is going bad (which I seriously doubt). I think I've seen these errors before when I last installed Slackware on a SATA drive...I think I selected the wrong kernel. I need to select a kernel specific to SATA support. I'll work on this during the next 7 days and report my findings here.

I'm seriously thinking on swapping the Intel machine for the AMD one, since the AMD machine appears more robust.

Bake-off: NoScript and Firekeeper

I decided to mention Firekeeper on the security forums at LQ.org. One of the moderators there mentioned that NoScript was better at blocking malcode than Firekeeper. In order to understand what he was talking about (I'm confused about that comment), I decided to install both to see if one can layer and leverage both of these tools. I also wanted to see which was better at blocking and alerting on malcode in general.

It appears that NoScript is specific to javascript, although it looks to detect cross-site scripting, flash, and MS' version of Flash. It also works via whitelists and blacklists and not pattern matching (other than focusing on the word "script" and occasionally focusing on "ath.cx" (I haven't determined why it does this yet).

Both tools work in conjunction with another fine, though (so far).

I'm partial to Snort because an efficient and focused rule will always beat someone adding a site to a whitelist. I've seen trusted sites be hacked before, so if a trusted site is violated and begins serving malware, you're going to be visiting that site and that site will be in your white list...with Firekeeper, it will alert and block any malicious traffic.

The bad thing about Firekeeper is that someone always has to maintain the ruleset (be it the user or the developer or a combination of both).

I'll continue to comment as I learn both tools.

Just ordered another machine

Yeah, yeah, I've ordered yet ANOTHER machine:

Abit NF-M2SV GeForce 6100 Socket AM2 Motherboard
AMD Athlon 64 X2 4400+ Socket AM2 CPU
Crucial 1024MB PC4200 DDR2 533MHz (X2)
Seagate 250GB Serial ATA w/NCQ 7200/8MB/SATA-3G
Power Up Silver 5511 ATX Mid-T Case w/450w

All for $199 after $30 in rebates. Note that there's no CPU heatsink/fan, no OS, no CD/DVD burner, and no vidcard (although there's an integrated one on the motherboard, which may get me through the testing/burn-in phase).

I've just ordered an Arctic Cooling Freezer 64 LP CPU Cooler for $25 from Microcenter. I'll order a CD/DVD burner in another week, and within the next month, I'll install a new vidcard. I don't know what OS I'll utilize yet...maybe Linux, but more than likely Windows (only I don't want to buy Vista [or XP, really]).

At this point, the barebones I ordered last year is still the better computer, but I spent quite a bit more for it. Until the new one is up and running, I'm also using its 2gb of RAM in the older computer, for a total of 3gb of RAM (COD4 flies during loading!).

This is the deal I saw on Tigerdirect.com that made me purchase this machine.

Verizon hit with GPL copyright lawsuit over router software

This article is old but interesting. I've this router and I've Verizon's FIOS service. Not long after purchasing this service, I perused Actiontec's website and had seen that they utilized Linux (this is, specifically, an issue with Verizon not including the source code for BusyBox to its customers, per v2 of the GPL) as the firmware for this router. I also saw that Verizon offered firmware versions for this router on their pages. I didn't think that they'd not release their software as GPL, though. I think it was either forgotten or GPL was taken for granted (because GPL software is usually free).

Anyways, this is a good read.

Port 33435

I'm doing some additional research on this ISC SANS diary entry. It appears that I have a prominent host attempting to connect to port 33435/UDP. The traffic is showing in my FW logs but I wanted to get a sniff going to provide to ISC.sans.org.

I used the following to capture the traffic:

tcpdump -Xvvnnes -0 -i eth0 -w /tmp/isc-inv/isc-inv1 port 14323 or port 33435

I got seven hits over several days:

root@starchild:~# screen -r 32692
7 packets received by filter
0 packets dropped by kernel

root@starchild:~# tcpdump -Xvvnnes -0 -r /tmp/isc-inv/isc-inv1
reading from file /tmp/isc-inv/isc-inv1, link-type EN10MB (Ethernet)
20:59:13.181494 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 659, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 0293 0000 0111 ae43 d834 6104 E..........C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
20:59:54.435063 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 2451, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 0993 0000 0111 a743 d834 6104 E..........C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
21:00:35.451099 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 4243, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 1093 0000 0111 a043 d834 6104 E..........C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
21:01:17.435358 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 6035, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 1793 0000 0111 9943 d834 6104 E..........C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
21:01:58.435072 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 7827, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 1e93 0000 0111 9243 d834 6104 E..........C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
21:02:40.432363 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 9619, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 2593 0000 0111 8b43 d834 6104 E...%......C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG
21:03:21.431071 00:0c:db:fc:8b:59 > fe:fd:40:3e:e7:dc, ethertype IPv4 (0x0800), length 46: (tos 0x0, ttl 1, id 11411, offset 0, flags [none], proto UDP (17), length 32) 216.52.97.4.11941 > xxx.xxx.xxx.xxx.33435: [udp sum ok] UDP, length 4
0x0000: 4500 0020 2c93 0000 0111 8443 d834 6104 E...,......C.4a.
0x0010: 42a0 8d1e 2ea5 829b 000c 8f00 6956 4d47 B...........iVMG

I've not yet taken the time to delve into the capture (will have some time when I get home today).

BASH script to parse FW logs

I've created a BASH script that parses my FW logs to show me the activity in one screen dump and also show me the total hit count per log file (I have my FW logs show in /var/log/messages).

The script is below:

root@starchild:/tmp# cat fwlogsearch2.sh
#!/bin/bash

# Searches FW logs on Linode, which are contained in /var/log/messages* files
#
# v0.1: couldn't get the script to work but could get the raw grep command to run flawlessly manually. Changed the "grep "$ip" /var/log/messages*" to "grep "$1" /var/log/messages*" and it worked! Same for the wordcount line.

function search {
local ip #ip is local to the function
echo "Searching... "
echo " "
grep "$1" /var/log/messages*
#cat /var/log/messages* | grep $ip
wordcount=`grep -c "$1" /var/log/messages*`
#wordcount=`cat /var/log/messages* | grep $ip | wc -l`
echo " "
echo "The number of instances this IP shows in $wordcount"
}
echo " "
echo " "
echo "Type in a number to search. Output will be dumped to stdout:"
read number
value_returned=$(search $number)
echo "$value_returned"
echo " "
echo " "

The results look like:

root@starchild:/tmp# ./fwlogsearch2.sh


Type in a number to search. Output will be dumped to stdout:
216.218.230.82
Searching...

/var/log/messages:Jun 3 05:23:30 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=18621 DF PROTO=TCP SPT=1121 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0
/var/log/messages:Jun 3 05:23:33 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=19854 DF PROTO=TCP SPT=1121 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0
/var/log/messages:Jun 21 15:57:52 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=3853 DF PROTO=TCP SPT=45085 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0
/var/log/messages:Jun 21 15:57:55 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=5091 DF PROTO=TCP SPT=45085 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0
/var/log/messages.1:May 29 22:08:44 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=28369 DF PROTO=TCP SPT=29144 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0
/var/log/messages.1:May 29 22:08:47 starchild kernel: Connection attempt (UNPRIV): IN=eth0 OUT= MAC=fe:fd:40:3e:e7:dc:00:0c:db:f5:90:00:08:00 SRC=216.218.230.82 DST=xxx.xxx.xxx.xxx LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=29195 DF PROTO=TCP SPT=29144 DPT=5900 WINDOW=65535 RES=0x00 SYN URGP=0

The number of instances this IP shows in /var/log/messages:4
/var/log/messages.1:2
/var/log/messages.2:0
/var/log/messages.3:0
/var/log/messages.4:0


root@starchild:/tmp#

The plan is to add more functionality to this simple script (yeah, I'm enthused because I don't normally script things and rarely get it right without some type of extreme research or problem).

Regarding Snort, I've recently added the following sigs to all three of my IDSs (regarding detecting Kraken activity):

# Kraken sigs (Emerging Threats sigs)
alert tcp $HOME_NET 1024: -> $EXTERNAL_NET 447 (msg:"ET CURRENT_EVENTS Bobax/Kraken/Oderoor TCP 447 CnC? Channel Initial Packet Outbound"; flow:established; dsize:24; threshold:type threshold, track by_src, count 2, seconds 360; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008103; rev:1;)
alert udp $HOME_NET 1024: -> $EXTERNAL_NET 447 (msg:"ET CURRENT_EVENTS Bobax/Kraken/Oderoor UDP 447 CnC? Channel Initial Packet Outbound"; dsize:24; threshold:type threshold, track by_src, count 2, seconds 360; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008104; rev:1;)
alert udp $EXTERNAL_NET 447 -> $HOME_NET 1024: (msg:"ET CURRENT_EVENTS Bobax/Kraken/Oderoor UDP 447 CnC? Channel Initial Packet Inbound"; dsize:24; threshold:type threshold, track by_src, count 2, seconds 360; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008105; rev:1;)
alert tcp $EXTERNAL_NET 447 -> $HOME_NET 1024: (msg:"ET CURRENT_EVENTS Bobax/Kraken/Oderoor TCP 447 CnC? Channel Initial Packet Inbound"; flow:established; dsize:24; threshold:type threshold, track by_src, count 2, seconds 360; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008106; rev:1;)
alert udp $EXTERNAL_NET 447 -> $HOME_NET 1024: (msg:"ET CURRENT_EVENTS Possible Bobax/Kraken/Oderoor UDP 447 CnC? Channel Inbound"; threshold:type threshold, track by_src, count 5, seconds 60; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008107; rev:1;)
alert tcp $EXTERNAL_NET 447 -> $HOME_NET 1024: (msg:"ET CURRENT_EVENTS Possible Bobax/Kraken/Oderoor TCP 447 CnC? Channel Inbound"; flow:established; threshold:type threshold, track by_src, count 5, seconds 60; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008108; rev:1;)
alert udp $HOME_NET 1024: -> $EXTERNAL_NET 447 (msg:"ET CURRENT_EVENTS Possible Bobax/Kraken/Oderoor UDP 447 CnC? Channel Outbound"; threshold:type threshold, track by_src, count 5, seconds 60; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008109; rev:1;)
alert tcp $HOME_NET 1024: -> $EXTERNAL_NET 447 (msg:"ET CURRENT_EVENTS Possible Bobax/Kraken/Oderoor TCP 447 CnC? Channel Outbound"; flow:established; threshold:type threshold, track by_src, count 5, seconds 60; classtype:trojan-activity; reference:url,doc.emergingthreats.net/bin/view/Main/OdeRoor; sid:2008110; rev:1;)

I doubt I'll see anything, but I'm a bit concerned, as this malware affects Windows systems and is supposed to alert on non-internet activity...I do have Windows machines on my LAN.

I also conducted some research on this ISC SANS diary entry. It appears that I have a prominent host attempting to connect to port 33435/UDP. I counted 50 FW log hits from maybe 4 different IPs, with one IP being more active than the rest.

root@starchild:/tmp# cat /var/log/messages* | grep "PT=33435" | wc -l
50

root@starchild:/tmp# whois 216.52.97.4
Internap Network Services PNAP-8-98 (NET-216-52-0-0-1)
216.52.0.0 - 216.52.255.255
InterNAP Network Services, PNAP-OCY PNAP-OCY-INAP-BB-1 (NET-216-52-96-0-1)
216.52.96.0 - 216.52.97.255

Looking at my logs, I also see 33436/UDP, 33437/UDP, 33438/UDP, and 33439/UDP being hit by hosts from PNAP hosts...strange...I'm thinking about blocking that whole huge range.

Anyways, I thought some of this would be cool to share.

Until next time!

Firekeeper, an IDPS system (plugin) for Firefox

http://isc.sans.org/diary.html?storyid=2403 explains Firekeeper, an IDS/IPS Firefox browser plugin.

I'm running it on two machines that run Slackware (versions 11.0 and 12.0). I may throw it on my work machine (which runs Windows XP), but that may be a bit daring.

Firekeeper's homepage is at http://firekeeper.mozdev.org/installation.html

Please share your experiences with this plugin...this is a great idea and may be a Holy Grail for malware that infects via browsers.

Also, I've found what may be a good security site, http://www.megasecurity.org/Main.html. It may take me awhile to read, as it has tons of data, it seems.

Kernel Upgrade

I've done the following (copy/paste):

root@slackbox:~/kernel-patches# ls
kernel-generic-2.6.21.5-i486-2_slack12.0.tgz
kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
kernel-huge-2.6.21.5-i486-2_slack12.0.tgz
kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
root@slackbox:~/kernel-patches# md5sum kernel-*
ebf025aa30af925ac6817fe58811e921 kernel-generic-2.6.21.5-i486-2_slack12.0.tgz
e35c66f2d765a221b509f1b7b463c9fe kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
3f9e3783dd7d799a277ec3e79e8bb82d kernel-huge-2.6.21.5-i486-2_slack12.0.tgz
0503193191731bba693ed6ce35b8c26d kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
root@slackbox:~/kernel-patches#
root@slackbox:~/kernel-patches#
root@slackbox:~/kernel-patches#
root@slackbox:~/kernel-patches# upgradepkg kernel-generic-2.6.21.5-i486-2_slack12.0.tgz

+==============================================================================
| Upgrading kernel-generic-2.6.21.5-i486-2 package using ./kernel-generic-2.6.21.5-i486-2_slack12.0.tgz
+==============================================================================

Pre-installing package kernel-generic-2.6.21.5-i486-2_slack12.0...

Removing package /var/log/packages/kernel-generic-2.6.21.5-i486-2-upgraded-2008-02-21,19:59:56...

Installing package kernel-generic-2.6.21.5-i486-2_slack12.0...
PACKAGE DESCRIPTION:
kernel-generic: kernel-generic (a general purpose single processor Linux kernel)
kernel-generic:
kernel-generic: This is a Linux kernel with built-in support for most IDE controllers.
kernel-generic: For filesystem support, or if you need to load support for a SCSI or
kernel-generic: other controller, then you'll need to load one or more kernel modules
kernel-generic: using an initial ramdisk, or initrd. For more information about
kernel-generic: creating an initrd, see the README.initrd file in the /boot directory.
kernel-generic:
Executing install script for kernel-generic-2.6.21.5-i486-2_slack12.0...

Package kernel-generic-2.6.21.5-i486-2 upgraded with new package ./kernel-generic-2.6.21.5-i486-2_slack12.0.tgz.

root@slackbox:~/kernel-patches# upgradepkg kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0.tgz

+==============================================================================
| Upgrading kernel-generic-smp-2.6.21.5_smp-i686-2 package using ./kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
+==============================================================================

Pre-installing package kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0...

Removing package /var/log/packages/kernel-generic-smp-2.6.21.5_smp-i686-2-upgraded-2008-02-21,20:01:00...

Installing package kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0...
PACKAGE DESCRIPTION:
kernel-generic-smp: kernel-generic-smp (a general purpose SMP Linux kernel)
kernel-generic-smp:
kernel-generic-smp: This is a Linux kernel with built-in support for most disk
kernel-generic-smp: controllers. To use filesystems, or to load support for a SCSI or
kernel-generic-smp: other controller, then you'll need to load one or more kernel
kernel-generic-smp: modules using an initial ramdisk, or initrd. For more information
kernel-generic-smp: about creating an initrd, see the README.initrd file in the /boot
kernel-generic-smp: directory.
kernel-generic-smp:
kernel-generic-smp: SMP is "Symmetric multiprocessing", or multiple CPU/core support.
kernel-generic-smp:
Executing install script for kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0...

Package kernel-generic-smp-2.6.21.5_smp-i686-2 upgraded with new package ./kernel-generic-smp-2.6.21.5_smp-i686-2_slack12.0.tgz.

root@slackbox:~/kernel-patches# upgradepkg kernel-huge-2.6.21.5-i486-2_slack12.0.tgz

+==============================================================================
| Upgrading kernel-huge-2.6.21.5-i486-2 package using ./kernel-huge-2.6.21.5-i486-2_slack12.0.tgz
+==============================================================================

Pre-installing package kernel-huge-2.6.21.5-i486-2_slack12.0...

Removing package /var/log/packages/kernel-huge-2.6.21.5-i486-2-upgraded-2008-02-21,20:01:34...

Installing package kernel-huge-2.6.21.5-i486-2_slack12.0...
PACKAGE DESCRIPTION:
kernel-huge: kernel-huge (a fully-loaded single processor Linux kernel)
kernel-huge:
kernel-huge: This is a Linux kernel with built-in support for most disk controllers
kernel-huge: and filesystems. If you're looking for a more stripped down kernel
kernel-huge: (this one contains everything but the kitchen sink ;-), then install
kernel-huge: the kernel-generic from the /boot directory along with an initrd to
kernel-huge: load support for your boot device and filesystem. For instructions
kernel-huge: on the initrd, see README.initrd in the /boot directory.
kernel-huge:
Executing install script for kernel-huge-2.6.21.5-i486-2_slack12.0...

Package kernel-huge-2.6.21.5-i486-2 upgraded with new package ./kernel-huge-2.6.21.5-i486-2_slack12.0.tgz.

root@slackbox:~/kernel-patches# upgradepkg kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0.tgz

+==============================================================================
| Upgrading kernel-huge-smp-2.6.21.5_smp-i686-2 package using ./kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0.tgz
+==============================================================================

Pre-installing package kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0...

Removing package /var/log/packages/kernel-huge-smp-2.6.21.5_smp-i686-2-upgraded-2008-02-21,20:02:13...

Installing package kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0...
PACKAGE DESCRIPTION:
kernel-huge-smp: kernel-huge-smp (a fully-loaded SMP Linux kernel)
kernel-huge-smp:
kernel-huge-smp: This is a Linux kernel with built-in support for most disk
kernel-huge-smp: controllers. If you're looking for a more stripped down kernel
kernel-huge-smp: (this one contains everything but the kitchen sink ;-), then install
kernel-huge-smp: the kernel-generic-smp in the /boot directory along with an initrd to
kernel-huge-smp: load support for your boot device and filesystem. For instructions
kernel-huge-smp: on the initrd, see README.initrd in the /boot directory.
kernel-huge-smp:
kernel-huge-smp: SMP is "Symmetric multiprocessing", or multiple CPU/core support.
kernel-huge-smp:
Executing install script for kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0...

Package kernel-huge-smp-2.6.21.5_smp-i686-2 upgraded with new package ./kernel-huge-smp-2.6.21.5_smp-i686-2_slack12.0.tgz.

root@slackbox:~/kernel-patches#

root@slackbox:~/kernel-patches# lilo
Fatal: VolumeID read error: sector 0 of /dev/sda not readable

OUCH!

I read this: http://unixadmintalk.com/f11/lilo-fails-dev-sda-not-readable-65533/

It appears to help:

root@slackbox:~/kernel-patches# lilo
Warning: bypassing VolumeID scan of drive flagged INACCESSIBLE: /dev/sda
Warning: The boot sector and map file are on different disks.
Added Windows *
Added Linux
2 warnings were issued.

Will reboot then test to see if this upgraded kernel is still vulnerable...

Kernel vulnerabilities affecting Linux machines

Whenever there's some kernel-level vulnerability, it seems that the whole community goes ape-crap over something that should be a no-brainer.

The recent vulnerability is documented here:

http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-0010
http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-0163
http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2008-0600

So, what's the big deal? It is a locally exploitable vulnerability. Everyone is acting like its the end of the world. Why? Are people actually giving people access to their systems that they don't trust? Why am I not worried? Because I want to learn things about security. Think about this for a second: in an enterprise environment, you're not going to be able to always apply kernel patches to production machines. You're not always going to be able to test by standing up a development environment. There is not always going to be one distribution used and not every platform will share the same hardware. What's readily apparent is that security should always be applied in layers. This means that no one should be accessing machines on your local network that you can't trust. If someone is not trustworthy, you should always be worrying about what they're doing on the network, instead of only when kernel-level vulnerabilities are discovered.

Does that lessen the responsibility of the system admins? No, but if everyone thought less of patching applications and more as a security administrator, the workload of the system administrator would probably be less. What I'm seeing in chatrooms and forums is this: "Oh shit...this exploit gives local root access...I have to apply this patch NOW!!" Someone said something similar in an IRC channel that I frequent:

SiegeX - I dunno, having a local root exploit (which ive tested with existing code) on a box that runs any sort of service would worry the hell out of me
W|GGL|T - SiegeX: in all actuality, you could have root exploits locally all over the place and you'd not know about it
SiegeX - and I probably do, but its no excuse for not patching the ones I do
W|GGL|T - security is more than just patching....in a corporate scenario, you have to balance out if you can even apply the patch....you bet your ass we're not going to take down a production system that has a localized vulnerability if it is indeed only local
SiegeX - heh, step 1) su root 2) cat /etc/shadow 3) ??? 4) profit
W|GGL|T - its called mitigation
W|GGL|T - if security is applied in layers, certain risks are lessened
SiegeX - W|GGL|T: why wouldnt you apply the patch on a production clone for testing purposes and do regression testing on that to make sure everything is a-ok before moving it over ?
W|GGL|T - SiegeX: if the corporate network has 10 different security layers, the need for immediate patching is small. sure, we'd patch but we'd do it in a sane manner
SiegeX - W|GGL|T: since you're into the corp security let me ask you if there was a solid way for a corp to not allow outbound tunnels while still allowing https?
SiegeX - s/was/is
W|GGL|T - SiegeX: nope, but then again, those who don't follow corporate policy need to be fired
SiegeX - afaik, if you tunnel over https, not even a L7 filter will look at it funny since the connection setup looks legit. Only thing i can think of is traffic analysis
W|GGL|T - there are always checks and balances
W|GGL|T - SiegeX: hrmm....there is IDS
W|GGL|T - and there is also a concept called behavioral analysis

The conversation dies shortly thereafter. I do think SiegeX was thinking in a sane manner. What he's worried about is someone either breaking into the machine or someone from inside tunneling and somehow letting an unauthorized user into the network. Layered security addresses both of those concerns. You lock down your firewall to only allow certain traffic in/out of the network. You set up either an IDS or an IPS to either log suspicious traffic or actively log and block unusual traffic. Yes, IDS/IPS can detect layer-7 traffic anomalies (but only if there are rules patterned after the unwanted traffic). Those people that tunnel out of the corporate network can be either reprimanded or handed their walking papers...that problem can be solved rather quickly.

I take it that SiegeX didn't want to deal with traffic analysis. That's the only way ANYONE is going to see stuff. Think about it. When you look at firewall logs, you're looking at logged traffic. If you're looking at your system logs (for instance, /var/log/secure, /var/log/faillog, or /var/log/messages (which may contain snort log and/or firewall log entries)), you're pretty much conducting traffic analysis. This should be within the realm of every system admin.

The easier way would be to address the kernel vulnerability, but I've also seen places that will NOT update a kernel unless absolutely necessary. The train-of-thought is that they wanted absolute stability and that stability overruled patch updating. What type of organization would think in this manner? Think of organizations that deal in national flight systems.

So, when am I going to apply a patched kernel? I don't know...my LAN is so layered with security that its not a hot priority for me to apply this patch.

Lastly, here's a Secunia link of the vulnerabilities in question: http://secunia.com/advisories/28835/

Mystery infestation strikes Linux/Apache Web sites

http://www.linux.com/feature/125548

"According to a press release issued earlier this month by Finjan, a security research firm, compromised Web servers are infecting thousands of visitors daily with malware that turns their Windows machines into unwitting bots to do the bidding of an as yet unidentified criminal organization. Security firms ScanSafe and SecureWorks have since added their own takes on the situation, though with varying estimates on the number of sites affected. All reports thus far say the compromised servers are running Linux and Apache."

What's New?

What's new for 2008?

I've quit smoking. The last time I smoked was on the 31st of Dec 2007. I've also enrolled in my company's benefits as a non-smoker (as an incentive and as punishment, as a smoker who has claimed non-smoker status can be disciplined or fired). I've been using smoking cessation aids (ie, Nicoderm and other aids).

Other than that, nothing is new, other than I'm burned out at work. Shiftwork and looking at packets all day (along with customer firewall requests and the semi-management stuff I do) has taken its toll, so my resume is out there and I've gotten some interesting hits. Sadly, most of
it is contract work (which sucks) or requires a clearance (my clearance status is still in some black hole somewhere). Soooo...I'm applying within the company for other positions of interest. I'd like to stay in my field and have completed one assessment 'test'...it blew my mind, along with it being like 60 questions long, essay format. The things I do to get a freekin' job... :)

Anyways, I've a tidbit for you. If anyone has ever perused their web server logs and saw the below:

193.205.4.38 - - [19/Jan/2008:16:31:56 -0500] "HEAD / HTTP/1.0" 200 0
193.205.4.38 - - [19/Jan/2008:16:31:56 -0500] "HEAD / HTTP/1.0" 200 0 "-" "-"
193.205.4.38 - - [19/Jan/2008:16:31:56 -0500] "POST /_vti_bin/_vti_aut/author.dll HTTP/1.1" 500 544
193.205.4.38 - - [19/Jan/2008:16:31:56 -0500] "POST /_vti_bin/_vti_aut/author.dll HTTP/1.1" 500 544 "-" "core-project/1.0"

It looks harmless, eh? Seen this tons of times before? I know I have. Well, take a look at how my Snort setup detected it:

WEB-MISC cross site scripting attempt 1 1 2008-01-19 16:31:56 2008-01-19 16:31:56

Digging deeper:

[ GAAAHHH...the code renders like pure dung when I post! ]

Note that I've disabled the harmful HTML flags and Snort removed the garbage (noted as non-ASCII characters).

And, no, I don't allow any inputting of text on my site, and I also don't allow any scripts to be run. My site is a static site, so I'm safe enough, along with using modsecurity and Snort for blocking of HTTP traffic and detection of badness. I refuse to be a statistic, although my stubbornness limits dynamic content serving.

The script looks like it checked for a live webserver then began the attack, quick-fast. Most people will associate the Frontpage attack as an old attack. The payload of the Frontpage attacks show:

method=put+document%3a4%2e0%2e2%2e4715&service%5fname=&document=%5bdocument%5fname%3dindex.htm%3bmeta%5finfo%3d%5b%5d%5d&put%5foption=overwrite&comment=&keep%5fchecked%5fout=false

I will not pretend I know what all it does. It is attempting to inject data into my server, though. The red flag for me is the 'method=put+document'. Also, there were two of these, happening 24 hours apart (but only one cross-site scripting event). I'll not block the site, as I may actually learn something from recording its attacks (and I can't block the whole internet, either).

Another host to block

I've just blocked 202.75.33.249. I haven't been paying heed to my Dshield reports and when I compared two reports today, I saw the same IP generating many hits. I checked the firewall logs and processed how many alerts this IP has generated. I found that the attacks began Nov 18th and the total number of alerts are 863.

This IP was a prime candidate for blocking.

Why don't I use Snort-inline? Because I don't have that much control over the network that my host is on (its a colo box running on a virtual server). So, I have to do things manually...it's not a problem, as it keeps me on my toes.

EDIT - I actually blocked 3 other IPs also. What's funny is that I saw one that was trying to connect on port 3389 (MS Term Svcs)...to a Linux machine...