– @page { size: 8.5in 11in; margin: 0.79in } P { margin-bottom: 0.08in } –>

Ok, been a busy busy couple of months, two conferences down, two scheduled.

In February, I helped with the new LinuxWorld/OpenSolutions World Summit in NY. The show did pretty well, but the weather really messed it up (warning to others, don’t schedule a conference during a massive snow/ice storm, really messes things up LOL). I missed most of the first day when my 4 hour drive turned into around 8 hours, and I didn’t get to leave until 4 hours later than planned. The show’s presentation are are being pod casted here:

http://www.linuxworld.com/events/

Next on the list was the FOSE 2007 show where I helped organize the Tux.org non-profit booth. We handed out over 1,000 issues of Linux Journal with an intro to Linux flier, over 1000 CD/DVD Linux Distros, a couple dozen live FreeBSD CDs, and numerous other giveaways. We had some generally interested people talk to us and it’s nice to see more and more people actually know what Linux is and what OpenSource really means. Here’s some information on what we were doing:

http://www.cyberigor.com/fose/

Well, that takes care of what’s been happening, Now coming up this week is the ShmooCon conference:

www.shmoocon.org

I love this conference, has some great speakers, Bruce Potter and the rest of the Shmoo Group do an amazing job with it. Starts in about 3 hours, going to be a fun weekend!

Finally LinuxWorld/OpenSolutions World San Francisco 2007 is starting up. Hoping to get some real good speakers lined up for the Security track (which I’m track chair for). Anyways, not much to see, but here’s the site for that:

http://www.linuxworldexpo.com/live/12/events/12SFO07A

I recently worked on two grid management articles for IBM’s developers center that were just published:

Managing a Grid, Part 3: Monitoring and Scheduling

Managing a Grid, Part 4: Day-to-day tasks for a grid admin

I’m pretty happy with how both of them turned out. They were on a real quick schedule so wasn’t sure how well they would come out. After going back and reading them, I feel they are a decent job. Hope people find them useful.

Oh, and I ran across this:

Care and Feeding of your Grid

So at least one person liked it :-)

Well, once again, I seem to be sparse on the posts. Maybe it’s because I’ve been busy with LinuxWorld Summit NY, or trying to get a new software project at work finished, or maybe cause I’m devoting my free time to my new Nintendo Wii. Well truthfully it’s a little of all the above (and a lot of getting my new software project working), but I did run across this little interesting tidbit:

http://www.nintendoworldreport.com/newsArt.cfm?artid=12687

I hadn’t seen this published broadly, but it appears that the version of opera currently under beta testing to Wii users (any Wii owner can download and play with Opera 9 on their Wii for free) has a vulnerability that can at minimum cause the Wii to hard lock. Currently there is no reported exploit that can run code on the Wii, which would normally be a threat from this exploit, and not clear if it ever will given that Game Consoles are usually pretty strict in what code is or isn’t allowed to run.

Anyways, interesting to see how the age of the desktop vulnerability has come to the game consoles world, even Nintendo. Wonder how long it takes them to issue a fix?

Wow, it’s been a very long time since I last posted something. Well, I do have somewhat of an excuse, I’ve been busy and traveling.

On the busy side, I’ve been trying to get several projects at work moving or finished including: Updates to www.gridswatch.com, Developming some web-based database applications, Working on content and sessions for the upcoming LinuxWorld Summit in New York, and most recently co-authoring some articles on working with Grid computing networks (I’ll post something when they are published).

On the travel/vacation side, I took a long overdue and just plain long vacation. We left for Europe (Barcelona, Spain) late October, spent a day in Barcelona, then a couple of days in France, then a few days in Italy, then back to Barcelona where we spent a few more days before heading to Madrid and Toledo, Spain. Had some great food, met some nice people, and came back with a sword which I purchased at the blacksmith’s shop after spending an hour talking about swords with the man who made my sword. Needless to say I had a great time, and it had been 7 years since the wife and I took a real vacation together. We didn’t get back till mid Novemeber just to give you an idea of how long a vacation this was (we had just a few vacation hours saved up…).

Of course about a week after our return we then had an extended Thanksgiving day weekend (5 days for me), so Novemember has been a light month on work. This translates into me being behind in what I want to have done before the end of the year. I’m going to be busy.

I’ll have a few more updates in December (hopefully articles going live, and some info on the LinuxWorld Summit that is coming). Not to mention gripes about stupid security I’m sure I’ll see.

Ok, been meaning to write a little about this, just couldn’t find the time.

To ALL those in charge of taking private information via secure webforms (credit cards, SSN, etc..) PLEASE READ THIS.

Yes, you must use an SSL encrypted webpage, yes you must only give that information collected to those who are directly responsible for billing the transaction. But DO NOT EMAIL all the information to anyone, and certainly don’t include it in the confirmation email!

I say this because I recently registered for a workshop I plan on attending. I’m not going to name the institution that is running it, nor am I going to mention the name of the course (though I must admit if I was presenting at the workshop I would be very pissed to learn that this was how they were sending confirmation emails). My company is paying for the workshop so they used the company credit card and the administrative assistant took care of the registration for me. Shortly after they registered me, I received the confirmation email. What did I find in that email that they sent to me (and to one other email address that we didn’t recognize), my contact information, all the contact information for the person holding our company card, the full credit card number, the Expiration date, and the CCV Code!

They emailed out everything you could possibly need to use the credit card at any online vendor in a plain text email over the unencrypted PUBLIC INTERNET!!!!

The fact that they had a nice SSL encrypted website to take this information just made the situation worse. Through their actions they have violated the trust relation they setup by presenting what appeared to be a secure internet transaction. By emailing the information they collected back over the internet, they placed that information at even more risk than if it was not emailed, but didn’t use an SSL cert. Now our credit information is being cached unencrypted on at least 2 email servers (most likely 4 or more) for who knows how long. If those machines are compromised or someone was having fun watching that traffic, they could now be purchasing a couple of big screen HDTV’s maybe a laptop or 4, subscribing to every porn site they want, etc..

People have got to remember that your responsibility for the secure transaction on the web doesn’t end at the SSL encrypted webform. It continues for as long as you hold and maintain that private information. End-to-end, review your policies, before it comes back to bite you.

I’ve been nice and I’m trying to work with these people to make sure they get this corrected. So far they seem to be listening (though action is a little slower). Hopefully they will get it, time will tell. If I had been someone less friendly, this could have been a much bigger headache for them.

Well, I just got back from LinuxWorld and have to say I had a great time. This was a much better show than Boston was. Actually, after attending this show I realized how much the Boston show lacked for attendance and things happening. I still think part of what killed the Boston show was the move out of New York to Boston, but now that that the Boston show has been dropped and the LinuxWorld Summit in New York set as a replacement, hopefully it will get back on track.

Anyways, had a lot of fun at a couple of the booths, notable were the Fedora guys, who willingly let me hose the FC6 test 2 box in the name of experimentation, the USENIX booth for some good information, and the Trolltech booth with their cool developer phone coming out soon. Outside of the actual show, I had a good time hanging out with Joshua Abraham (pbnj developer) and Jon “maddog” Hall. Both of them were very cool, and introduced me to several other very cool people. Unfortunately, I’m terrible with names, and besides Pixel (aka “Bob”), can’t remember who I had the pleasure of having dinner and hanging out with Wednesday evening. Here is a rather bad picture taken with my Treo at dinner of some of the folks:

LinuxWorld group at dinner in SanFrancisco

Ok, here is the actual interview that I had with Network World.  I really like how it came out, I think Phil Hochmuth did an excellent job taking what I had said and presenting it to the reader.  This is also the interview where they had taken the quote from for the previous article.   Well here it is if you are interested:

LinuxWorld experts: Securing Web-based application on Linux

A pretty good article about the upcoming LinuxWorld/OpenSolutions World Conference and expo came out in the online magazine Network World today. I was even quoted in it at the bottom of the second page. It was an interesting interview, and I believe I rambled on for way too long, but they managed to get my main point, that is, Security is about finding the right compromises, and there are good tools now and in the near future to help us get there.

Now there is one small issue, and I’m posting this as a correction. I am not currently working on development of SELinux technology. I am working on some policy stuff, but I am not active in the community development of the technology currently. I have worked on it in the recent past, and plan to continue in the near future (though I am thinking of helping out indirectly via SEDarwin). I just wanted to make that clear. I support SELinux, I’m trying to help promote and improve SELinux, but I’m not a core developer of the technology. There are others such as the NSA, Trusted Computer Solutions, IBM, Tresys Technology and several other groups and companies that are putting in the hard work to make this technology a reality in the production world.

Here is the article for those interested.

Interesting article, and I would love to see this presentation at the BlackHat conference. Jon Ellch and David Myanor will be showing off how they can hijack a MacBook laptop in about 60 seconds using vulnerabilities in the wireless card driver. There are a couple of things that make this interesting:

1. All that has to happen is that your wireless card be turned on. You don’t have to be connected to a network. If you wireless card is on, you are a target, period.

2. In theory, there is nothing to say that BlueTooth is safe from this either. I would imagine that similer vulnerabilities could be found in Bluetooth drivers as well.

3. This is not Mac OS specific! Though they used a Mac for the demo, they have also discovered vulnerabilities in Windows. And I see no reason that it couldn’t affect Linux/*BSD as well.

4. Firewalls and anti-virus programs won’t and can’t protect you from this. This is a much lower level attack and will always bypass this. The only way to protect against it is either through better device driver security or not using wireless. SELinux/SEBSD/SEDarwin may help this somewhat, but again drivers are usually in the OS kernel and once you’re in the kernel it’s hard to stop attacks. I’ll have to look into the SE* solutions and see if they might be used to help mitigate this attack (though I’m doubtful).

Currently, there isn’t much you can do to protect yourself. Just turn off wireless when you don’t need it. Apple’s patches just came out, but there was no mention of a fix for this. The researchers are talking to Apple, Microsoft, and others to get this fixed. Also, they are not showing how they did it, just that they did it, so no current “in the wild” exploits are known of at this point.

Ok, I recently have been given the opportunity to play around with a new MacBook Pro 15″ laptop (Mac OS X 10.4 - Tiger). So far I’m impressed, clean easy to use user interface with a nice Unix/BSD system underneath. In the process of getting it set up, I did go through and take care of some security issues to make sure I was happy. Some of these are obvious, some less so:

I. Click on the Apple Icon on the top left and select System Preferences

1. Click on Security
a. Set a master Password, and don’t forget it, this is used to recover lost accounts and such.
b. Turn on FileVault, this is a great security item, but will slow down your computer and could make crash recovery harder. I haven’t done this one yet.
c. Check require password to wake computer.
d. Check Disable Auto Login, don’t make it easier for someone who steals your laptop, it can happen.
e. Check Require password to unlock secure system preferences, this will help against trojans and such that could attack MacOSx.
f. Check user secure virtual memory, this is mostly for a multiple user system. I haven’t done this yet myself.
g. Check disable remote control infrared receiver, less critical, but if you aren’t using, why enable it?

2. Click on Bluetooth
a. Disable Discoverable, you don’t need to advertise that you are a possible hacking target. Most bluetooth devices you use don’t require your desktop to be discoverable. Only when you are trying to send files and such to the desktop for the first time with a device does this need to be enabled. After a pairing trust is setup you don’t need this enabled again for that device.

3. Click on Network
a. Select Airport and then options, then check Require admin password for Computer-to-Computer networks. There have been attacks in the past where machines (in that case, Windows) were able to create a computer-to-computer network while sitting in the airport without the need for the users’ intervention. It’s best to set this option just to make sure it doesn’t happen without your express consent.

4. Click on Sharing
a. Turn on “Remote Login” this turns on the ssh daemon so you can ssh into your box like you normally do with linux.
b. Choose Firewall and turn it on. By default, it seems MacOSX doesn’t turn on its firewall. I personally prefer to have it up and running. You can then enable different remote services though the firewall below that. I enabled Remote Login - SSH, iChat, and Network time.
c. Under Firewall Advanced, enable Block UDP Traffic and Stealth Mode. So far, neither of these have blocked traffic such as iChat Video/Sound or anything else, so better to block unwanted traffic.

5. Click on Startup Disk
a. Make sure that the lock icon on the bottom is selected. Unless you are reinstalling your base OS, no reason to have this easily changed.

Those are the preference you can change via gui. Here are some to change via command line Terminal:

II. Start Terminal, you can find this by clicking on the search tool (magnifying glass in the top right corner) and using the term terminal.


1. Set a root password. There is a root user on MacOSX, and by default it’s disabled from normal use. But I’m paranoid, so unless I know the root password I don’t like it. You can set it by using the command “sudo passwd root” which will then ask you for the new root password. You may want to set this to the same as the masterpassword. I’m not positive, but they may be linked, I haven’t researched it that far yet. Warning, this will enable the root user account. I still prefer having the password set to something I know vs being blank and disabled. Consider this optional and your preference.

2. If you’re using SSHD for remote login, make it more secure. Using “sudo vi /etc/sshd_config” set “Protocol 2″, “PermitRootLogin no”, and “AllowUsers username” to your “username” for your main account if you only want that account to ever be able to SSH into your Mac. This is very important if you enable the root account like I did in step 1.

3. Double check the sudoers file. By default, it’s set up pretty well, only root and admin users can use sudo (which means do anything as admin/root all powerful user). You might want to double check it to make sure “sudo vi /etc/sudoers”.

4. Change your users directory permission. By default, your new users directory is readable by any user on your computer. Though there may not be another user on your computer, it’s best to change that to only be accessible by you. In the terminal you could type in “cd ..” which will put you in the /Users folder. Type ls -l will give you a list of users, most likely just a Shared and your username. Then issue the command “chmod 750 username” username being your actual username. This will give you full control over your directory, but no other users besides root has access full access and admin users have read access. I would go with chmod 700 to block other admin users, but I don’t know about Mac OSX enough and what other system level problems that might cause with software daemons running.

Well that’s what I found, if you know something I missed, or a mistake I made please let me know. So far I haven’t found anything impaired by these settings for normal day-to-day use, but I’m only starting to play with Mac OS X.

« Previous PageNext Page »