In my own words...
This is page 3 of the discussion. Read the opening post
After Hours
This is page 3 of the discussion. Read the opening post
Previously I took you through
the importance of keeping systems 'patched' and up to date; we then
looked at the apt update mechanism used by Debian--how it resolves
dependencies of packages and allows for patches to be quickly and easily
applied. This week, I want to take a look at one of the major commercial
Linux distributions, SUSE Linux Enterprise Server, and see how they
deal with the same issues.
As most of you will know, Novell are behind SUSE Linux Enterprise Server
(SLES). Many enterprises choose to go with this commercial version due
to the peace of mind offered by full support, backed by Novell. Support
means far more than simply being able to ring a call centre in India
and be driven mad while they pass you from person to person; Novell
support provides not only installation support and hardware certification/testing,
they also protect you from possible intellectual property issues which
could arise in relation to Linux (Microsoft push this risk as a big
negative factor when dismissing Linux). The last, and for us, most important
part of the support package is the seven-year lifecycle of the product;
that means that from the launch of a SUSE Enterprise Linux release (SLES9,
for example), operating system patches and security updates will be
available via the SUSE Linux Portal for the length of your support subscription?for
seven years. This means you don?t need to worry about upgrading to
the latest release, just so that you can maintain a secure system free
from vulnerability.
Although I have recently started to favour Debian-based distributions
such as Ubuntu, I still use SUSE Linux Enterprise Server for core service-bearing
machines.
YaST (Yet another Setup Tool)
Control Centre is the main hub for all administration work with SLES--from
here, you can add/remove packages, perform 'Online Updates', change
hardware configuration such as graphics mode, change system preferences
(partition editing, network settings, hostname, firewall setup, time/date
etc), control system services, and even manage users. Of course we?re
interested in the online update functionality, YaST Online Update (YOU
for short), which you can see in Figure A [http://i.i.com.com/cnwk.1d/i/tr/NL_images/Fielding0419_A.jpg] (click to view).
YaST Online Update can be configured
to pick updates from the SUSE Portal, a different HTTP of FTP source,
CD, DVD, or local Windows / NFS shares. Most people will use the direct
SUSE Portal, however if you need to update multiple servers there would
be considerable advantage to mirroring the SUSE repository and performing
the updates via a local share; this would save considerable bandwidth.
Not all patches need to be
applied. Kernel updates, for example, will display a warning before
installation and give you the chance to skip. This is pretty useful
if the reason for the patch does not affect you, or you don?t want
to update due to module dependencies (for example HP CCISS module).
There is one issue that I?ve come across, which is that at some point,
other patches/updates may not apply if the Kernel is not up to date.
If a specific patch requires a service to be restarted, stopped, or
it requires a configuration modification, then a prompt will be displayed
with any relevant information and instructions, such as in Figure
B [http://i.i.com.com/cnwk.1d/i/tr/NL_images/Fielding0419_B.jpg] (click to view).
All things considered, YaST
Online Update is relatively trouble free, solving all dependencies just
as Debian?s apt or RedHat?s yum. As I said, the only
issue I have had with YOU is in applying new updates without updating
to the latest Kernel patch.
I approached Novell and asked
them how long they aim to take in order to patch a vulnerability once
it?s in the public domain. It was stressed that Novell work hand-in-hand
with other members of the Open Source community to fix security holes
as quickly as possible?the time scale varies from a few hours to a
few days, depending on the severity and complexity. Upgrades to packages
(e.g., version updates) are not usually provided unless there is a security
fix in the newer version; however, service packs will sometimes contain
version updates to add new functionality or compatibility. When I raised
the issue of Kernel updates/module dependencies (such as CCISS for HP
Servers), it was mentioned that in the upcoming SUSE Enterprise Server
10, there will be a new way of dealing with Kernel updates. A dedicated
Kernel update tool will be added that will check loaded Kernel modules
and assess their compatibility with the new Kernel?this sounds like
an interesting development which I?m quite eager to see in action.
As a general package management
tool, YOU is not as good as apt due to the lack of available
program updates; however, as a tool for delivering security patches,
it is every bit as good--and of course, fully commercially supported.
Have you been using SLES and
YOU? How do you compare it to apt or yum? Do you prefer
to run updates from the SUSE Portal or a local repository?
A 'proof of concept' (e.g. harmless) cross-platform virus has been
reported by security firm Kaspersky Lab. The virus, named
Virus.Linux.Bi.a/ Virus.Win32.Bi.a, does not seem to have any malicious
function; it does however cause concern as it could mark the beginning
of a new age where Linux users are no longer safe from the infections
spread via Windows.
Files infected with the virus will contain the following text:
[CAPZLOQ TEKNIQ 1.0] (c) 2006 JPanic:
This is Sepultura signing off...
This is The Soul Manager saying goodbye...
Greetz to: Immortal Riot, #RuxCon!
If you don't already have your systems protected by anti-virus software,
it's probably about time to start, the following solutions could be
considered:
BitDefender for Samba is an interesting product if you are running a
samba fileserver
BitDefender for Linux is an on-demand solution for Linux, this is freeware
Clam AntiVirus is a well respected scanning engine, it can be used to
perform on-demand scans, on-the-fly mail scanning and even on-access
scanning (although the last time I checked this was unstable)
by cruel_symbol
I found an interesting piece on reallylinux
about the very real
barrier being created by 'Linux Snobs'. It makes an interesting read
and raises some very good points.
I remember when I first started 'playing' around with Linux, It took me
at least a week to download the ISO's for mandrake (56k)--I was utterly
confused by the abundance of different distributions, all claiming to be
'the best'; hardware support was a major headache, It was such an ordeal
(for a zero experience user) to get online via a modem that I ended up
connecting via a hub and my Windows 98 PC (using NAT). Ok, so once I
was online I needed to learn about using this new OS. Great, there was
an abundance of documentation--however I didn't really want to spend 4
hours reading a very comprehensive how-to when my only goal was to get
my USB mouse working! Not a problem, I went to one of the helpful IRC
rooms and asked if there was an easier way to get things running
smoothly; hmm, 2 seconds after asking a question I was kicked and banned
from the room with some useless comment like 'Read the manual n00b' or
'$tick with Windoez'. There were some IRC rooms with more helpful
members, rooms like #linuxnewbies and #linuxquestions. These did tend
to get quite crowded and unless the question was relatively
straightforward then one was often overlooked.
After this initial experience, lots of wasted time and quite a bit of
frustration I gave up on Linux for a few years. Later I came back to
Linux mainly due to necessity--managing our web/db server which was
running Fedora Core 1; I had no choice but to dig my teeth in and learn
the slow/hard way. I also installed Fedora on my laptop and worked on
it as often as possible (even managing to have Dreamweaver running with
WINE at one point), increasing my exposure to the OS. I found forums to
be a very good source of information and direct help; the responses were
also more detailed and quite mild mannered compared to those found on IRC.
I have found that as I become more experienced, working with Linux
becomes easier, not because I know the ins and outs of a particular
program but rather because I have a better understanding of the
underlying system and how/why thiongs work in the way that they do;
working as a full time Linux Sys Admin I now also find the advice and
knowledge shared by more experienced colleagues to be invaluable. I
find the mailing lists for the specific program or distribution to be
very helpful and finally there are still many forums such as
www.linuxquestions.org which will normally return a quick answer to any
question.
What was your introduction to Linux like? How did you pick up your
initial survival skills and how did you deal with the sometimes less
than polite characters in the realm of Linux?
by justin james
This is reason #7,219 that I prefer working with BSD over Linux. The community is simply much more professional. They recognize that every user that gets welcomed and helped with getting their system set up is another person who may be giving back further down the road. I have never once received a "n00b" or "RTFM" from BSD users. More importantly, the documentation is so much better, I so rarely need to actually put a question to the community. I suggest that if you are having problems with a Linux system, and the documentation stinks, go to a BSD site and look there, anything in userland should be identical or nearly identical.
J.Ja
My initial brush with Linux was almost three years ago. After several weeks, I abandoned it for the same reasons you initially did - no motivating reason to pursue it. I haven't had a reason to go back since it's not required for those technologies we have or will implement. The only skills I've acquired are installing SUSE and RH9, and I couldn't get Windows back on either of the hard drives I used. One day I may stick my toe back in the water, or maybe even put my head under, but it probably won't be this year.
I too was (and still am) confused by the variety of distributions. If the only flavor you've ever even heard of is vanilla, Baskin-Robbins can be a bit overwhelming. Unlike ice cream, I dislike the suggestion "Keep trying, you'll find a distro you like." Continuously reloading different operating systems is not my idea of a good time.
I've found rude people in Windows forums too. I respond to rude Linux respondents the same way I respond to rude Windows people. I ignore their answers and patiently wait for someone else to post the answer. The second or third poster often takes the rude one to task for not responding in a useful fashion. ("RTFB" is not a useful answer if the distro didn't come with a FB. MAN pages are written assuming a familiarity with *nix OSs that newbies rarely have.) Most of the Linux experts hanging out at TR are quite helpful,
although you can expect to have your distro selection second-guessed on
a regular basis. While I was a TR member at the time of my Linux
experiments, I wasn't active in the Q&A or forums. I would have
benefited from the caliber of advice available here.
Nothing says "I'm a Linux Snob" better than intentionally misspelling "Microsoft" or "Windows". Hey Penguinistas, that was funny in 1996 or so, but it got old in a hurry. The second sign of a "Linux Snob" is to reply "Switch to Linux!" in response to every Windows question. There are several links around TR to "Linux is not Windows". While it wonderfully explains what Windows users should expect, "Linux Snobs" should read it to learn why "Switch to Linux!" is not the answer to a Windows problem.
I do often wonder how newbies get advice without a computer with working Internet access. When I was actively trolling the newbie forums I would read questions about how to get a particular distro / hardware combination on line. I can only assume they were using a Windows computer to ask their question.
While I found my local Linux Users Group to be of great assistance via e-mail, I found their meetings to be much less than useful. Based in the CSCI department of the local university, the two meetings I attended revolved around topics highly unsuited to a newbie (a program to debug programs, an upcoming developers' conference in Europe). I don't recall anyone of the 15 or so in attendance who was not with the university or a software developer. No one from the corporate world, no hobbists, no SOHOs, and no other newbies.
Would you mind posting a link to the reallylinux article you referenced above?
by dawgit
Interesting. Some very good info here. Thanks for that. I'll have to give it a wack as soon as I get some time. Again, Thanks. -d
by apche2004
The New York county of Westchester has passed a new law in an attempt to
curb the current growth in identity theft. Businesses which may store
personal information in an unencrypted manor are now required to install
a firewall or change the default SSID of wireless access
points--penalties for non-compliance range from an initial warning to a
$500 fine.
While it is accepted that simple measures like these will not stop
identity theft, this is seen as a move in the right direction.
According to the county's chief information office, Norman Jacknis,
while the law was being considered officials picked up over 100
unsecured access points during a 20 minute drive.
Public hotspots such as cafes will be required to post a warning sign
stating "For your own protection and privacy, you are advised to install
a firewall or other computer security measure when accessing the Internet."
This new legislation has drawn attention from other areas of the US as
well as the UK and Europe.
The Everdream corporation (http://www.everdream.com) has released a
service which will encrypt or delete files on a laptop or PC if stolen.
The software runs as an agent on the computer--this connects to the
control centre of Everdream which extracts network location details from
the computer and if the computer is stolen will either encrypt or delete
files specified by the customer. A policy is created by the customer to
decide whether encryption or deletion would take place, and which files
are targeted.
The service costs around $6 per PC.
The usefulness of such a system is debatable, it assumes that the thief
is stupid enough to connect to the internet. Most stolen computers will
be reformatted as soon as they are switched on; windows logon passwords
will prompt this. If the computer was stolen for the information
contained on the hard disk, it's unlikely that it would be connected to
the internet before all of the desired information has been pulled off.
USB flash drives, pen drives, portable hard disks?these tiny,
high-capacity storage devices have a thousand and one uses. They have rendered
most other types of portable storage?floppy disks, ZIP drives, and even
rewritable CDs?pretty much obsolete. Their high capacity and high reliability
(with no moving/mechanical parts to fail), combined with extreme ease of use
makes them the ultimate in portable storage. You rarely need to install drivers
for them, making them true plug and play devices. At the time of this writing,
a 2GB
USB2 flash drive will set you back ?37?that?s pretty affordable!
There is no question that USB portable storage has changed
the way people view portable media and the way in which people work.
Previously, 3.5" floppy disks were used to cart work back and forth; any
IT support staff can tell you that these were less than reliable, and trying to
explain to users why their disks had become corrupted was no easy task! Any
files over the 1.4-MB limit of a floppy disk would require either a ZIP drive,
or later, a writable CD to transfer. The ZIP drive was awkward to carry around,
requiring drivers to be installed on most PCs, which was often not allowed by the
Windows policies in place. CDR/RW was a great improvement but the media was
delicate and still gave some compatibility issues with older CDROM drives not
being able to read the new disks. USB Flash drives saved us all; however, they
do now present administrators with a new set of problems, and re-present older
issues.
An interesting report in the
Los Angeles Times a few weeks back highlighted one major security issue
which has been created with the introduction of USB storage. I?m sure we would
all imagine military security to be of a high standard when compared to small
businesses or even large corporations?how shocking it is, then, to see that
reporters bought USB flash devices from a bazaar 200 yards outside of the Bagram
military base in Afghanistan. These devices (stolen from the base by cleaners
and other local workers) contained documents marked "Secret," which
named suspected militants, documented U.S. efforts to remove Afghan government
officials, and a classified briefing on "man portable counter-mortar
radar" now being used in Iraq. One device also listed over 700 service
members with their social security details, opening them up to identity theft.
This highlights the greatest danger posed by plug and play
portable storage?data loss. While the spread of viruses and the theft of data
also pose an issue, there are clear and simple methods to deal with these
problems. Virus outbreaks are handled by ensuring that on-access scanning is
enabled on corporate machines and virus definitions are kept up to date. Data
theft can be hampered by making sure that all machines are password-protected
and locked while not in use (including once the screensaver has been
activated). Data loss, however (e.g., lost or stolen USB drives which contain
sensitive information) is much harder to deal with. One solution is to make
sure that users are equipped with secure
storage devices. These devices have encryption/conditional access programs
included, which require a password to access the contained data. If lost or
stolen, these will be pretty much useless to the new ?owner?. The real problem
is that unless you equip every user with once of these devices, someone will
still use their own unsecured device (even if you do equip everyone, they may
still use their own devices). There are only two ways to stop this?both pretty
drastic.
All things considered, USB storage has been a godsend for
both administrators and end users?however, it is important to be aware of the
risks that they pose and to educate our users.
What is your company's policy on the usage of USB
media? How do you control its use? It would be interesting to hear your
experiences and opinions.
by justin james
Turn off the Internet while you're at it! I don't carry around USB devices to transfer data, I stick stuff on my FTP server at home. Slower, yes, but I do not have to worry about physical devices (I am rather abusive to portable equipment). What is really needed is file system encryption that relies upon the user being actively authenticated against the network, in a transparent to the user (and applications) fashion. That doesn't help with people emailing/FTPing/whatever data arounf, but it does help with the portable storage problem.
J.Ja
by flecavalier
The Ultimate in Secure Digital Identity and
Encrypted Storage
Stealth
Stealth MXP? is a Portable Security Device. It is a USB powered, secure, multi-functional
product with on-board processing for seamless, hardware based encryption, secure
storage and strong authentication. Stealth MXP provides unprecedented security,
functionality and flexibility in the management of sensitive corporate information on a
portable device as well as protection and assertion of personal and corporate digital
identities.
Stealth MXP is ideally suited for enterprise security applications such as Single Sign On,
PKI, encryption, and remote access. It has management interfaces that enable
organizations to fully control security policies, deployment and usage of the devices.
As a digital identity device, Stealth MXP interoperates with virtually any security
infrastructure providing unprecedented flexibility and mobility of digital credentials. It
protects these credentials with strong user authentication (biometric, password or both)
which allows systems to enforce 2 or 3 factor authentication. Stealth MXP is also ready for
the emerging digital identity meta-system as a WS-Trust Portable Security Token Service
(PSTS) capable of issuing SAML tokens for secure claims based identity transactions to
target services. Stealth MXP offers a host of general purpose, industry standard, cryptographic
services, including random number generation, key generation with internal or external entropy,
encryption and decryption, signature generation and verification, one time password, and secure hash.
Stealth MXP also offers transparent encryption (AES 256) of portable mass storage in a ?stick? format. The
Stealth MXP uses a patent pending communication protocol that provides true portability on any environment that
supports USB mass storage.
Stealth MXP technology is available in other hardware versions for those who do not need the portable mass
storage or biometric capability. These device alternatives are Stealth MXP? Bio Token, which has no user mass
storage, and Stealth MXP? Token, which does not have mass storage or biometric authentication.
As mentioned while looking at spam filtering in previous blogs, botnets
(networks of 'bots') are being noted as an increasing source of
cyber-criminal activity. A securityfocus article
(http://www.securityfocus.com/brief/195?ref=rss) quotes shocking figures
from The Messaging Anti-Abuse Working Group; as many as 7% of PC's
worldwide are infected and working as active bots, that's around 47
million. They also estimate that bots now send 70% of all spam--this is
an increasing problem for System Administrators as it means many more
residential broadband IP addresses are being blacklisted by projects
like SpamHaus. I have had several instances where users trying to send
mail from outside of the corporate network have been unable to due to
being in a dns blacklist (while not infected themselves, a previous user
of the dynamic IP address must have been).
Previously I took readers over the merits of Nagios, the Open Source
system monitoring tool. For some people Nagios is too heavy and
requires far too much time and effort to configure; what's needed is a
lightweight monitoring tool which can keep an eye on daemons and let you
know when things aren't as they should be. 'monit'
(http://www.tildeslash.com/monit/) seems to fit the bill, it could also
make a useful companion to Nagios as it will detect a dead process and
restart it for you! A handy HTTP interface is built in so you can check
on the system status from the comfort of a web browser. Services can be
put in to groups for easy management, dependencies can be configured
between services and files/disks can be monitored.
All over this looks
like a useful little program (compiles to 200KB), I haven't tried it yet
but intend to in the near future--maybe you can give it a try and let us
know how you got on?
As previously discussed, portable storage (e.g., USB
PenDrives) has changed the way that we move our data, but it brings the issue
of potential data loss firmly into the forefront of our security strategy. These
devices are very small, and they are all too easy to misplace (or have stolen),
so how do we go about making sure that the data stored on these devices is
secure and can only be accessed by those who are authorised to do so?
Some device manufacturers like SanDisk offer encryption /
protection software, which is CruzerLock in its
case. The basic features offered by these packages are pretty much equal across
the board. The free version of CruzerLock includes both file encryption and
compression:
Upgraded versions of the software are available; some of the
additional features offered by these include the ability to assign permission
rights (view, copy, delete etc.), use integrated filesharing, and lock content
to specific machines. To be honest, I don?t think it?s worth paying up to $100
for the upgraded features?I simply want to lock my files away so they are
protected if I lose the device. I currently use a small USB key to store a
backup of my financial records. This is protected by the manufacturers'
included software, which is easy to use and portable, as the unlock program is
run straight from a small unencrypted partition on the key. There is, however,
one reason why I have not used encryption on my main 2GB USB key: very simply,
it?s that the manufacturers' programs never support Linux. I keep copies of
letters, e-mail and bookmark backups, photos, etc. on my key (not exactly top-secret
military documents, but I would rather not have people looking through them if
I lost it). I am in Linux often, so if I want to update my bookmarks to put
them in sync with my Windows bookmarks?my key is encrypted by the Windows
software, and I have no chance of getting to the bookmarks file while in Linux.
Of course, if I set up an encrypted filesystem on the key from Linux, I can?t
access my data in Windows.
Well, now I have found a solution?an Open Source project
called TrueCrypt.
Not only is TrueCrypt available for Windows, but I was
delighted to see that Linux packages are available in many flavours: Fedora
rpm, Debian/Ubuntu deb, plus SuSE rpm. Add to that the source code for both the
Windows and Linux applications, and the only major OS lacking is Apples OS X. A
future release is apparently planned. So what are the main features?
The documentation is very complete. It seems there is also a
traveller mode for Windows use?this installs an application which can automatically
launch the unlock program and then, once supplied with the password, it will
mount your encrypted drive. It would be nice to have this functionality for
Linux too, but I can?t see it happening (due to differing Kernel versions etc.).
Next week we?ll take a look at setting up TrueCrypt in Windows. How well will
it work?
by dawgit
Ok, I'll byte. As some-one who is swaping things back & forth between Windows & Linux, doing so a little more secure sounds good to me. Thanks for the heads-up on this. -d
by dlandry9
by lizseal
You can call this a blog, but it is really excellent Security Awareness information that I am glad to have. And, I will be passing this on to a couple of Security individuals, that I have worked with. Thinking of these devices as extended infrastructure puts a new twist into Security Policy and Desktop Support. Thanks
The long awaited Playstation 3 launch has been set for this November,
one year (and 10-million units) behind the launch of Microsofts
Xbox360. The Playstation 3 is claimed to offer twice the performance of
the X360 with a 3-core 3.2Ghz PowerPC CPU. The wireless controllers
will use bluetooth based connectivity, similar to those of the X360 and
there will be two versions of the console on sale priced at $499 and
$599--the major difference being cited is a 20GB hard disk in the $499
model and 60GB hard disk in the $599 model.
What Sony have not been so forthcoming in advertising is that the
cheaper model will lack support for removable media (SD, CF, Sony Duo),
the WiFi will be removed and so will the HDMI outputs which are
necessary for high definition--this could create a lot of unhappy owners
when they are brought the base model for Christmas!
The new controllers will be tilt sensitive; general opinion seems to be
that although a novel feature, this could end up being more irritating
than absorbing.
by frylock
"...HDMI outputs which are
necessary for high definition..."
The $499 model won't even have component video outputs? HDMI is not
necessary for high-def, you can get it from component anolog
connections as well. HDMI is digital but it's not necessarily better,
and it is by no means the only way to get high definition video (by
which I expect you mean >480p).
by gkurcon
HDMI WILL be required if you want to watch Blu Ray movies in full HD
resolution-assuming the studios set the disc to disable HD over
component...a main feature of the new HDCP rules being applied to
Blu-Ray and HD-DVD. Most of the studios say they will not do
this, for now. But come on, this is the same industry that sues
11 year old girls for millions of dollars because she downloaded some
mp3s off of Kazaa. My advice is bank on the industry crippling
your PS3's component outs in the next year or two when you go to watch
a Blu-Ray movie.
by frylock
Ah, HDCP with Blu-Ray. There is no technical reason you couldn't send
HD content over component, so they'll create a politcal one. Of course,
I should have expected this! Still, I'm a little surprised they're
considering copy protection for analog, in the past at least most of
their paranioa was centered around digital copies.
by smorty71
by gkurcon
Evil bot-master Jeanson James Ancheta (aged 20) was sentenced to 57
months (4 1/2+ years) in prison for spreading a bot program which
compromised over 400,000 computers including those used in national
defence and healthcare. The 20-year-old made over $100,000 in affiliate
schemes involving the installation of adware the victims machines--he
also admitted to selling access to his botnets for unknown activities
which could have included DOS attacks and sending of SPAM.
After serving his prison term, Ancheta will be under 3 years of
supervised release in which time his access to computers will be limited.
A full account of the whole investigation, trial and conviction can be
found on teh Department of Justice' cybercrime.org
(http://www.cybercrime.gov/anchetaSent.htm)
by dawgit
1 down, X? to go. I'd rather they let him have access to computers though, to explain just how he did it. It should be a requirement for release from prison for any one convicted of such computer crimes, write a paper on how & why they did it. (and no, they don't get to keep the money) -d
by bfilmfan
His new name is "That Lifer's Girlfriend."
I hope he REALLY enjoys learning that spam is very, very painful.
So, last week I talked about TrueCrypt, the disk/file encryption package
for both Windows and Linux. With the rising threat posed by identity theft, we
should all be careful of what information we store on portable media?the devices
are easily lost, so we need to assume that all of the data stored can be
accessed by any random individual (be they good or bad). Encrypting our stored
data negates this risk. At the very worst, you simply lose your data, but you
don?t reveal anything. The most anyone who finds the media can do is format it
and keep it. Losing your data may be a minor inconvenience, but it would be far
worse to lose your data and also find out that ?50k in debt has been created
under your identity!
Luckily, we now have a way to consistently encrypt and decrypt
in both a Windows and Linux environment. Let's start by taking a look at
TrueCrypt in Windows. The package can be downloaded here, and installation is
pretty self-explanatory. Download the archive, unpack the files in a temporary
directory, and then run the TrueCrypt_Setup.exe application. During the
installation, there is an option to create a system restore point?I used this,
but I don?t think it?s a necessary step; it's more for peace of mind.
Once installed, a TrueCrypt shortcut will be accessible from
either the Desktop or StartMenu. When the application is opened, it will remain
minimised in the tray when closed?in the preferences there is an option to have
TrueCrypt run on startup.
So let's take a look at the main application:

It?s all pretty standard with the main functions well
placed. Creating a new encrypted volume is simple: Tools > Volume Creation Wizard. The wizard guides us through the
creation of a new volume very smoothly, offering support content along the way
so as to explain everything. First we need to select whether we want to create
a hidden or standard volume; since I just want to keep my data secure in case
of loss, I don?t see the need for hiding the volume, therefore I?m creating a
standard one.
Next up, we need to select a file or device to encrypt. The
nice thing about TrueCrypt is that it gives this choice?if encrypting an entire
hard disk, we may well want to select the entire device or a partition on that
device; however, for the USB key, I have taken a different approach and used an
encrypted file. You?ll see why later.
I created a file called ?123.iso? placed on my freshly
formatted (FAT32) 2GB pen drive. Click next and you get a choice of encryption.
Many algorithms are available for use, and I quite like the fact that it lets
me choose. I have gone for Twofish, which uses a 256-bit key and 128-bit block?good
enough, I think. The next screen asks you how large the encrypted filesystem
should be. Underneath the box, it shows you the amount of free space on the
device where the filesystem will be created. I choose to use all but 10-MB of
the space for my volume. The final steps before creation are setting a password
and selecting a filesystem. For the password, over 20 characters are
recommended and up to 64 are allowed; ten characters seem a bit more likely to
me. Yes, it?s not as secure as 20, but you try remembering a 20-character
string of jumble and then typing it every time you switch computers! I opted
for a FAT filesystem with the default cluster size. Hitting format will create
the volume, which can then be mounted in the main application.
Now you?re probably wondering why I left 10MB of ?wasted?
space on the disk? TrueCrypt allows us to create a ?traveller disk?. The
traveller disk basically contains the application and driver which need to be
used to access the encrypted volume in Windows; this can be run on any Windows
machine (if you have rights to run executables) and can even automatically
launch when the disk is inserted (if using Windows XP SP2). The ?traveller
disk? option in the Tools menu
simply asks for the root directory of the disk and auto-mount options. It then
creates the necessary files on the disk. It?s a shame there isn?t a feature
like this for roaming on Linux machines, however I doubt it would be possible
due to the way in which the Linux program runs (loadable Kernel module
required). A few MB of leftover space will provide fast access to some
unimportant small files which you want to quickly move from one place to
another. How much you leave (if any at all) is down to personal preference.
Next week I?ll take a final look at TrueCrypt; I plan
to install the Linux variant on my Ubuntu workhorse and attempt to mount the
volume I have created in Windows.
A number of reports by market research giant Frost & Sullivan have
prompted many calls for business to wake up and prepare for the
inevitable move to IPv6. A recent report by Australian iTnews quotes Frost &
Sullivan's principal analyst for the area in saying "2012, that's when
we estimate the world will be out of IPv4 addresses,". So less than 6
years, this is also backed up by a report from Cisco
who predict that if growth in IP usage continues to grow at the current
rate, we'll run out within 7 years. The huge demand for smartphones and
similar devices is suggested to be behind the surge of uptake.
There is of course the opposing opinion that with sensible use, IPv4 can
last much longer--the use of NAT being heralded as our saviour.
However, IPv6 and it's inbuilt security features combined with the huge
address pool will restore end to end computing, greatly aiding P2P
applications and mobility/roaming.
IPv6 is coming, it may be 5 years away but given the huge amount of
planning involved it's probably worth starting to read up and getting
ahead of the game.
Unintended filesharing has again come in to the news with the Japan
Times report
on a security leak of the Chubu Electric Power Company. Highly
sensitive information including personnel details were accidentally made
accessible on the internet after a 40 year old employee had started to
use P2P filesharing which was miss-configured. There have been several
high profile cases including a report in 2004 that military data was
being leaked on P2P networks.
This should prompt us all to look at our networks and attempt to prevent
the use of P2P applications--it only takes one rouge user to seriously
breach security!
As I?m sure you would have read in the last two weeks, I
have been looking at the security issues caused by the recent explosion in USB
storage; not only are the old issues of virus infection / migration brought up,
but due to the small size of USB flash keys?data loss and theft are brought to
the front line. The only way to ensure
that our data can?t be read by other people is to encrypt it.
Encrypting these devices isn?t as straight forward as it
seems?yes for Windows users some of these devices come with proprietary
software which will allow the device to be locked. Linux users can simply create an encrypted
filesystem on the device much like any other disk. The big issue comes when a user wants to
switch between operating systems. Many
Linux users will also use Windows be that at work or home?Mac OS X is another
operating system to consider as it is becoming rapidly more popular, especially
in the home for the less technically savvy (or those who appreciate its stability
and media related software). What we
need is a strong encryption application which can encrypt/decrypt independent
of platform?this requirement cries out Open Source; the solution is indeed
provided as an Open Source project?enter TrueCrypt.
Last week we went over creating an encrypted USB key with
TrueCrypt in Windows, now let?s look at installing TrueCrypt on a Linux system
and mounting our previously created volume.
I have decided to install TrueCrypt under Ubuntu, this is
simply because Ubuntu has become my most frequently used Linux distribution; I
like the stability and repository system of Debian with the user friendliness
of Fedora. The TrueCrypt download page
offers some precompiled packages; I did try to install one of these, however
because my Kernel had been updated from the original provided on the Ubuntu
5.10 DVD (via apt-get upgrade) the package would not install?therefore it?s
necessary to compile from source.
First the Kernel source needs to be installed if it?s not
already. Let?s check which Kernel
version we?re running and also which compiler was used to build it:
# cat /proc/version
Linux Version 2.6.12-10-686 (buildd@terranova) (gcc version
3.4.5 20050809 (pre-release) (Ubuntu 3.4.4-6ubuntu8.1)) #1 Sat Mar 11 16:22:51
UTC 2006
The Kernel version is 2.6.12-10-686, the actually source
version is therefore 2.6.12, the gcc version used to compile was 3.4.5. Now we
need to download the source with apt-get:
# apt-get install liunx-source-2.6.12
Unpack the source and create a link from /usr/src/linux to
the current Kernel source:
# cd /usr/src
# tar ?xvjf linux-source-2.6.12.tar.bz2
# ln ?s linux-source-2.6.12 linux
Remember that if you aren?t logged in as root you will need
to either switch to root (?su ??) or prepend ?sudo? to the above commands.
If you don?t have gcc installed already, install with the
following:
# apt-get install build-essential gcc-3.4
Now let?s make sure we are using gcc 3.4:
# export CC=gcc-3.4
And build the Kernel modules:
# make ?C /usr/src/linux-source-2.6.12 config modules
You will be presented with a myriad of questions and options
which can be set?assuming you are using the standard build of Kernel you can
just accept the default answer for these.
Once that?s taken car of we can compile TrueCrypt which is
much more straightforward. Unpack the
source and enter the Linux source directory:
# tar -xvfz truecrypt-4.1-source-code.tar.gz ?C ~/build/
# cd ~/build/truecrypt-4.1-source-code/Linux/
Building and installing is very simple, runt the appropriate
scripts and follow any on-screen prompts:
# ./build.sh
# ./install.sh
Mounting the device is now completely painless, I have
created a directory in ?/mnt? called ?tc?, this will be my mount point for the
filesystem. The USB key is automatically
mounted by Ubuntu under ?/media/USB-KEY?.
You will remember that last week we created a file called ?123.iso?
which contains our encrypted filesystem, mounting it happens with one command:
# truecrypt /media/USB-KEY/123.iso /mnt/tc
Easy huh, let?s just check:
# ls /mnt/tc
Encrypted.txt
Great!
All things considered I would highly recommend looking in to
the use of TrueCrypt?I have been using this package since I discovered it just
over three weeks ago and I have no regrets; for the most part I?m working in
Windows and the traveller program allows me to mount my volume even if I?m
roaming on another machine. When I?m
using Linux I can still access my files, therefore I have no excuse to leave my
portable storage unsecured. I notice
that several readers commented on previous posts mentioning TrueCrypt, I would
be interested to hear how you got on (if you have had time to try it yet)?
by jc2it
Berners-Lee, the founder/creator of the world wide web has warned that
the efforts of some companies in the US to have legislation
changed--could curb net neutrality. This is the latest of many concerns
to be aired on the subject of net neutrality; infrastructure providers
who have invested heavily in creating an open and interconnected
infrastructure (which make up the internet as we know it) want to take a
share of revenues generated by companies who use it to deliver their
services--a Verizon executive recently said in a Washington Post report
that Google and other internet based businesses shouldn't expect to
enjoy a "free lunch" on its pipes! Service providers want to be able to
charge companies to receive a higher bandwidth allocation, therefore
meaning that the consumer can enjoy a higher quality of service from
them--they claim that without this source of revenue there is no
incentive for the service providers to perform expensive upgrades and
invest in new infrastructure. The other side of the argument is that
these companies already receive income from their broadband customers to
support this infrastructure--it's also argued that such quality of
service adjustments destroy both the neutrality of the internet and the
potential for new start-up internet firms who would not be able to
compete with the established giants if such a policy cam in to place.
This is a hot topic and one which will effect us all sooner or later!
Here's some more in-depth coverage.
1.
2.
by dawgit
This looks like the scary scene that bothered most of the WORLD, and was expressed last year in Tunis. The question once again, do we or the rest of the world want or need the US Gov, involved in policing the internet?
by larrydou38
Whenever the government gets involved, we lose. Gov. is about politicians protecting their jobs and pandering to who feeds the kitty.
larrydou
by btljooz
China (a COMMUNIST regime) has already split up and sequestered the net within it's own borders WITH the help of such US companies as Microsoft, Google and Yahoo.
Do WE want that for OURSELVES as well as the rest of the world? NO!!!!!!!!!!!!!!!!!!!!
An earlier post asked who's going to stop the government from ruining the net. I say here and now that it MUST be WE THE PEOPLE!!!!!! WE MUST do EVERYTHING in our power from voting to BLASTING e-mails (Boston Tea Party style) until our government GETS the CLUE!!! If one has time to troll on message boards then one has the time to STAND up to our government each in our freely chosen way to tell them that WE THE PEOPLE have had ENOUGH!!!!
United We Stand - Divided We Fall!!!!
by wjbernard
To help stop the companies who would like to change how the internet functions to their economic benefit, start now with constant emails to their corporate headquarters with "we're mad as hell and wont take it anymore" stuff. Send your email to ten friends, ask them to send it to ten, etc., etc. Verison would be a good start. I don't think this is a national security issue so there should be no JD intrusion. Unless the Verison folks have put a bundle in the Bush camp. For you out of the country, no problem. Let's just do it!!
Microsoft have released Windows Vista Beta2 only 6 weeks later than the
original target of April 12th. The first release candidate has also
been put back a month to August 25th (as a target). Manufacturers were
due to receive Windows Vista on August 25th, however, this has been put
back until October 25th! This long line of delays is creating
speculation over the state of the next major Windows release--consumer
release is expected to be in January 2007, missing the pre-Christmas
sales boom. Running past this holiday period will however take some
pressure off of Microsoft; many analysts say that after missing the
holiday sales period means that January 2007 may as well be July!
When we'll really see Windows Vista on the shelves is a mystery--I'm
personally praying Apple will release OS X for general PC installation;
there would be no contest over which to buy!
by jc2it
It?s often difficult to decide on a policy for locking down
your workstations, especially in a small- to medium-size enterprise. It?s nice
to allow users some freedom to personalise their computers--install tickers and
screensavers, etc; however, there comes a point where an inordinate amount of
time is spent repairing and rebuilding machines due to users fiddling with
things that they don?t understand or solving software conflicts caused by
random applications, games, and malware. The only way to stop this regular
rebuild cycle is to lock down what users can do on their machines and create
standard builds including only the software required by users to meet their
business objectives.
Most people working within a large enterprise environment
would be shocked at the suggestion that the average users would have anything
but the most basic rights, but this is pretty normal for smaller companies
which tend to follow more user-friendly standards. Microsoft allows a majority
of its employees full local admin rights, although this may
change with the rollout of Windows Vista, for security reasons.
So, locking down Windows workstations within a Windows
domain isn?t too tricky; group policies can be distributed, software deployed
and user group permissions defined. However, it can seem a lot more daunting to
set restrictive policies on workstations that are not operating in a Windows
domain environment.
So what are our basic tools for controlling what a user can
and can?t do?
Account Permissions:
One way of limiting what users can and can?t do is to set
them up on the workstation with a limited account. This is a very simple
process, when a user is created via control panel; there is a choice of making
the user an Administrator (they can do anything on the machine) or a Limited
user. Limited users are very restricted; they cannot install software or hardware,
make system-wide changes, or access files of other users on the computer. One
issue you could encounter as a limited user is that some legacy (i.e.,
non-WinXP-approved) applications may not run correctly. Some users could be
added to the Power User group--which has more rights--however, these problems
can still occur. You may also find that setting up users with limited accounts
will stop them from installing random applications, but it doesn?t stop them
from messing around with other settings in their own account. This is where our
second tool comes in.
Group Policy Editor:
The Group Policy Editor can easily be accessed by typing ?gpedit.msc?
in your Run dialog on the Start menu. You will see straight away that the
layout is presented in a Windows Explorer style?if you descend through the
menus you?ll find the ?Administrative Templates? under ?User Configuration?. In
this section, you can dictate a huge range of system permissions for the
current user. You can remove various options from the system, including access
to the Control Panel and Run dialog. This is a very powerful tool, even when
not combined with a Windows domain and the associated benefits. It will take
some time to fine-tune policies and find the right balance for your
organisation.
I?m still looking for a definitive answer as to how one
would create a master policy on one workstation, limiting a single user in
various ways (but leaving the administrator free to do anything and everything)
and then transferring this policy to other workstations in order to create a
standard build of permissions (without having to define each part of the policy
on every machine). I have also found that if using a Limited User account, one
does not have permissions to run the ?gpedit.msc? tool, therefore Group Policy
would need to be created as the administrator, and applied to the computer
rather than current user, which in turn limits the administrator. So far I
haven?t found a satisfactory way to limit a local user with the Group Policy
editor, while leaving the local administrator user unrestricted. Has anyone had
experience with achieving this task? If so please leave some clues in the
comments for this blog :)
by porzuc
A way to do this is to use a tool that can enhance and help define the corporate policy like ClientSync from a company called DeskSync www.desksync.com with it you can actually update and monitor the Corporate policy and automatically limit the user yet allow administrator rights to the computer either remotely or by maintaining the corporate policy for software. It's inexpensive as well. This bundled tool is scalable to many thousands of computers is not a bandwith hog and can search and destroy non corporate software it finds on users computers! If you are a Systems Admin this toll makes your job a breeze, and allow you to spend your time administering your computers rather than visiting them each time there is a ticket (Break Fix).
George Porzuc, SimpliCTI Software Services, Cary, NC
I previously posted on the comments of Berners-Lee, regarding the threat
posed to the openness of the internet by new US legislation. Today I
came across an interesting blog via Slashdot; Max Fomitchev
does a very good job of breaking down the proposed legislation and what
it really means to the internet. Not only does the proposed two-tier
system mean that systems are effectively being paid three times (1.
Broadband user subscriptions, in the UK these have bandwidth
rates/limits. 2. Hosting providers charge customers for bandwidth
usage. 3. ISP's would charge these companies again for delivering this
content to the end user), start-up companies are forced out of the
market by established (e.g. higher paying) competitors but also it
creates an opportunity for indirect censorship of personal blogs and
websites. How so? Well if you don't pay, the ISP's may not deliver
your content to it's customers!
A worrying prospect.
A proof of concept virus effecting OpenOffice an StarOffice has been
created. The virus is apparently purely proof of concept and has no
harmful intent, it simply loads a picture from the internet. Macro
viruses have become much less common than in previous years, however
there has been a recent case of a trojan attacking via a flaw in
Microsoft Word (http://www.securityfocus.com/brief/213).
This proof of concept virus is the most recent of quite a few which have
targeted Linux and Mac OS X
(http://computerworld.com/securitytopics/security/virus/story/0,10801,110330,00.html)--previously
these platforms had been considered as pretty safe from the evil virus
writers who concentrated their efforts on Windows (an easy target). As
more users move to these other platforms, it's only natural that virus
writers will start to shift their efforts.
As previously discussed
there are many good anti-virus engines available for these platforms so
there's no need to go without protection.
by jaqui
Backups, from both a compliance and a business continuity point-of-view,
are essential, particularly in the financial services sector. For example, backing
up of e-mail communication is important as it may be required even years later for
reference in a legal case--either in prosecution or defence.
Backups are inherently vulnerable, however; they can easily
be damaged and rendered useless by a number of culprits, including fire, water,
sunlight exposure or magnetic radiation. This is the main reason that backups
should be stored off-site. You could have the most advanced backup mechanism in
the world, but if the tapes are stored in the same building as your systems,
there?ll be a small problem when the building burns down. ?Ah but we have ours
in a fireproof safe,? you say; that won?t help much if the building collapses
on top of it!
Backups are also physically vulnerable to theft. If I wanted
to steal data from a company, I would have a few options. (1) Compromise the
premises and physically bug the systems (key logger, data box, etc.): It's difficult
to do, slow, and even if you gain access to internal systems like this?data
will no doubt still be protected and retrieval will need further work. (2)
Compromise the premises and physically steal the data: Not a good idea, very
intrusive. (3) Hire a hacker to break in and steal the data: Probably
expensive, definitely very slow, and he or she will probably just vanish with
your money. (4) Plant an employee within the company to mine data: Slow,
expensive, and difficult. (5) Wait until the IT manager throws the latest
backup in the back of his car, then break in and steal it: Really, not too
hard. Okay, so I don?t actually want to steal anyone?s data, but the point is
that backups not only need to be stored off-site, but they need to be
transported and stored securely?better known off-site vaulting firms will
collect your tapes in an armoured van!
It used to be that there was only one media for the backup
of data?magnetic tape. However, as technology has developed (and gotten
cheaper) alternative forms of storage have become available. Tapes can often be
slow and cumbersome, especially as they lack the ability to randomly access
data. For small (less than 10 GB) amounts of data; writable DVD media have become
quite tempting. The problem is that optical media (CD-R, DVD-R, DVD+R, etc) is
volatile. It may be cheap, fast to create, and even faster to access, but that
won?t matter when you go to retrieve an important document in two years time
and find the media can?t be read!
There is much evidence for so called ?DVD Rot?,
basically, the physical degradation of DVD media. While one may be tempted to put
this down to cheap or brandless media, it is worth considering that ?DVD Rot?
has been reported in pressed media too, which we would be led to believe is
much more durable and long-lasting than its writable cousins. 20th
Century Fox only offers a 90-day warranty on their DVDs!
Unfortunately, for long-term, stable storage, magnetic tape
looks like our only option. LTO tapes can carry 400 GB of data and have a 30
year shelf life. Even if we?re conservative and assume that in real world
conditions the tapes are good for 15 years, that?s a lot better than more
volatile media can offer.
Next week we?ll look at some other solutions for backup and
disaster recovery.
What media does your firm use for backups? How do you assure
the security of your data? Let us know, leave a comment!
by cpr
Even though tapes may have a life expectancy of 15+ years, if you really need to keep data for any length of time (1+ years), you should re-write the data on an annual basis.
I had a situation where we kept data on tapes (off-site with a reputable firm) for 5+ years. When we upgraded our tapes and drives, we experienced 30% tape read failures when converting the off-site tapes. We had to spend additional funds hiring a data recovery firm to rebuild our files. Refresh your off-site files annually and don't forget about the off-site data when/if you upgrade your hardware.
by leonv
This is a good article in terms of encouraging off site backup, but I think it falls short in its analysis of fire proof safes. It mentions that some might rely on fireproof safes, and that buildings could collapse, creating problems. In this scenario, some might still be tempted to plan on digging through the rubble for the fireproof safe to recover their media and their data.
At that point they are likely to discover a much bigger problem with fireproof safes. "Fireproof? safes are generally rated for paper, not for media. The temperature at which paper will spontaneously combust is about 450 degrees F (232 degrees C). Depending on the plastic, deflection and warping can start as low as 158 degrees F (70 degrees C). Full melting can occur as low as 266 degrees F (130 degrees C). Plastic (CDs, DVDs, and tapes) will be destroyed long before paper is any danger.
Also, the paper rating is based on a limit of oxygen within the safe. Should the temperature get high enough for paper to combust, the limited oxygen will limit the amount of paper that will actually burn. Losing the top few pages and having scorched edges on a stack of paper may not be a big deal. Plastic, on the other hand, requires no oxygen to melt. Digging out a safe may reveal paper in tact, but media is likely to be TOTALLY destroyed.
Real data rated storage is substantially more expensive, and is still limited by the amount of time it can protect media. Unless your data has little value, fireproof safes are, in this DR Planners opinion, NO match for (redundant) off site data storage.
by robinheat
I was repeatedly warned by my brother to take a back up of whatever I had done or else i would have to suffer one if my disk crashed. But i payed no heed to his warnings and i was not ready for more space wastage. All was good when suddenly one morning my disk crashed and i was left with nothing and i didnt knew what to do. The disk contained all my business details, Contacts and upcoming projects. Then i was referred to Disk Doctors Labs Inc which helped me recover my data and in Good time before i could face any further loss.
People who are trying to first know about them then here is the link http://www.diskdoctors.com/ . These people also provide online consultation and lot of other stuff
I mentioned last month that following several reports by Frost &
Sullivan, people are being urged to prepare for a move to IPv6. It
seems more analysts are urging companies to start planning as the switch
to IPv6 comes ever closer. An article in this weeks computer weekly
further backs this stance. Arif Mohamed writes that while Operating
Systems such a Windows Vista, OS-X and Linux all have IPv6 installed by
default--network devices such as routers, switches and printers should
be reviewed sooner rather than later, as some will need to be replaced
due to incomparability with the new protocol.
Michael Disablo, vice president of analyst firm Burton Group warned "It
is essential to plan for IPv6. Eventually you will need to upgrade".
Neil Richard, research vice president at Gartner said "Migrating to IPv6
is a huge undertaking, at least as big and complicated as Y2K projects,
if not more so".
I wonder whether we'll see a last minute panic when the world moves to
IPv6, much like the mass hysteria which surrounded Y2K? If you have
experience and a full understanding of IPv6 at this time, there may well
be some great opportunities in consulting!
The latest annual report on server
reliability from the Yankee Group has put Windows 2003 Server above
Linux--Windows 2003 Server was actually reported to boast almost 20%
more annual uptime than Red Hat Enterprise Linux! The most solid server
operating systems were UNIX based, HP-UX for example. Linux systems
were said to be offline more often and for longer than Windows and UNIX
systems; the main reason was cited as scarcity of documentation. I do
agree that this can be an issue--although there is an abundance of
documentation, man pages and mailing lists; finding useful information
or direct answers can often be a nightmare. My experience is that you
will normally find ten other people trying to solve the same issue as
yourself, nine of who have given up and one who's still trying. That's
not always the case of course, but neither is it infrequent.
The Yankee Group did make sure it was clear that this review is
independent, their work not being sponsored or commissioned by any of
the companies in question. Most will remember a campaign made by
Microsoft, boasting that Windows 2003 Server beats Linux for
reliability--the source of information used in this campaign was seen by
many as suspicious. The most recent report has caused similar
accusations to pop up on forums all over the internet; one claims that
these findings are based on the firm Sunbelt Software--a Microsoft
consulting firm!
What are your opinions on the subject? It would be interesting to hear
from people who use both platfo
by hkai
by than.yujin
I have been doing with Windows, Linux and UNIX. I learn UNIX environment for 2 years, then only starting to learn Microsoft Windows recently. Below which is my comment.
Microsoft Windows - Pretty easy to catch up if you have the UNIX environment skills. You will be tired to study its security patches, tons of it. The malware and viruses are mostly looking for its vulnerability. There is also a good point of it which you can get the MS Windows resources easilly from the web.
UNIX - The command is too tecnical, but if we compare with its OS stability, it is solid. I believe we will need the support from vendor specialization skills.
Linux - This is the best for doing research and development. The widely develop of "Open Source" tools makes Linux growing rapidly. This is the best OS for Small Medium Enterprise. Linux is pretty Solid too, but also need skills to manage it.
All OS has its pro and cons, it is just the matter wheather you have the skills to maintain it or not.
UNIX is still the most solid OS, if you have the skills to maintain it.
The best design of environment is always a mix of all, then only you will be able to covers the negative point of it.
Last week, I took a look at the merits and pitfalls of
various backup media. The main area of concern was that some people may be
using unstable media like CD-R and DVD to back up important data?these media
can be a tempting resource for backups smaller than 10 GB due to their low cost
and fast access. However, the instability of the media (susceptibility to
?rot?) makes the chance of recovery at a later date uncertain. The traditional
media for long-term backup and archiving is, of course, the magnetic tape?with
a native capacity of up to 400 GB and a shelf life of up to 30 years, there
really is no more reliable safety net.
In recent years, there has been another backup and disaster
recovery solution which is becoming more popular. Online backup involves data
being sent to a remote site via the Internet. So how exactly does online backup
work? DPS,
a leading online backup service provider do a very good job of describing the
process:
That?s all very well for restoring accidentally deleted
files, but what happens in a disaster if your building burns down? Most online
vaulting companies will offer emergency restores in the form of DVD media or a
mobile vault for large amounts of data.
The major advantage to online vaulting as opposed to the
traditional tape/offline vault seems to be that data is immediately available
for download, and small restores become much faster and cheaper because you
don?t need to recall a tape from the vault. With Internet connectivity becoming
faster and cheaper by the day, online backups of some sort will probably become
the norm for many medium to large firms. It may never replace the process of
taking physical backups because people like to feel they have a ?real?,
physical safety net; however, it is probably going to run along side on-site
backups, allowing easy recovery of files, while the physically vaulted media is
kept for long-term archiving.
Does your company use online backup services? What issues
have you come up against? Do you continue to take hard backups along side the
on-line backup?
by btljooz
To attest to the longevity of magnetic tapes as a a backup medium, doesn't one only have to go as far as the Movie or Music Industry vaults to find out for SURE???
And I have back up CDs that I burned over five years ago that read just fine. I simply TAKE __CARE__ of them allowing no extreme change of climate and no scratches.
~~~ Just my half-cent's worth........ :)
The proposal to safeguard net neutrality has been rejected by the House
of Representatives 269-152. The proposed amendment to the
Communications Opportunity, Promotion and Enhancement Act (Cope Act)
added clauses to force service providers to treat all internet traffic
equally--preserving the very pillars which founded the internet--the
free and unbiased flow of information.
Sir Tim Berners-Lee, founder of the internet as we know it, warned that
the internet looks to be entering a "dark period". The act now moves to
the US Senate where the Commerce, Science and Transportation Committee
will vote late in June.
A full story can be found here.
Yell, the business directory provider has rolled
out biometric single sign on to 1500 of their mobile staff. The chosen
implementation - Imprivata OneSign ESSO, provides access to 10
applications via a single, biometric (fingerprint) sign-on. The
biometric scanner used is the inbuilt security device in IBM notebooks.
The move to a single sign-on will not only ease administration but also
help the firm meet compliance targets--the system will keep records of
staffs access to applications and computers; it can also detect
suspicious behaviour such as the sharing of log-ons, activity outside of
business hours etc. Fingerprint recognition for desktops is also being
trailed.
Have you had any experience with biometric interfaces? I briefly used
the Microsoft fingerprint reader which was great, until I found they had
engineered it not to work with FireFox!
Recently, I've been looking at the various technologies for
backing up data. In addition to the previously discussed optical
media solutions and online
backups, one technology undergoing heavy development (pointed out to me by
a colleague) is that of holographic media. Holographic media is a next-generation
optical storage candidate. Forget HD-DVD and Blu-ray--the huge potential of HVD
puts them to shame. A single DVD-sized HVD disc could potentially store 3.9
terabytes; that?s 6000 CDs, 830 DVDs, or 160 Blu-ray discs! This definitely surpasses
the high-volume storage capacity offered by magnetic tapes!
How does HVD
work? Blu-ray and HD-DVD get more data on to a disc than conventional DVD
by using a shorter wavelength of light to cram more data in the same space. HVD
actually uses two lasers (green and red) and two layers on the disc. The bottom
layer is much like the metal film layer of a DVD-R?this contains servo
information picked up by the red laser. The green laser deals with data storage
on the upper layer. Between these two layers, a dichroic mirror layer reflects
the green laser while allowing the red laser through. While current optical
media store 1-bit per pulse, HVD hopes to increase this to 60,000-bits per
pulse! The HVD alliance
web site gives some more detailed information for anyone who?s interested in
exactly how this is achieved.
Although we?re a long way from seeing HVD in daily use?some
rapid progress is being made. Optware
are set to release a 30-GB holographic card before the end of 2006?the media
will cost around $1 USD, reader devices should come in at roughly $1700, with a
reader/writer setting you back over $8000. The drive prices can be compared to those
of DVD-R in the late 90?s; blank media, however, seem to be much cheaper. The
current cutting edge formats, Blu-ray and HD-DVD, have just started to become
economically viable for daily use. Read/write drives are currently around the
$1000 mark with media coming in at around $30. Some, however, are describing
these as transitional formats, claiming
that they will not enjoy as great a lifespan as current DVD formats due to the
rapid development of HVD and other high volume holographic technology. Whether
this is the case or not will remain to be seen. A great deal of money has been
invested in Blu-ray and HD-DVD; manufacturers will likely want to see as much
return on their investment as possible.
Whatever the short-term outcome, holographic storage will probably
become the norm, just as DVDs are at present. Whether this media can be used
for long-term data archiving will still depend on the stability of the media. The
huge capacity does overcome this in some manner?re-writing a few 4TB discs
annually would be much more acceptable than 800-fold, if DVD media was used!
What are your opinions on next generation optical media?
Will we see the lifespan of new formats decrease and a rapid development of new
or will this be artificially slowed to give consumer markets a chance to keep
up, as well as increase profits for technology companies? Have you started to use
Blu-ray or HD-DVD? If not, do you plan to?
Oracle have
released a list of validated Linux configurations. These configurations
have been tested and validated--they include pre-tested hardware,
software and configuration guidelines, as well as documented best
practices. Oracle finds that many support issues are due to systems
being incorrectly configured; by providing recommended configurations it
is hoped that a large number of these issues are avoided, therefore
reducing support costs.
The validated configurations obviously only involve Oracles stragic
partners--on the hardware front this is not too restrictive (Dell, EMC,
HP, IBM, Network Appliance, Sun), however this does mean that when it
comes to choosing an underlying Linux distribution customers are limited
to Red Hats' Enterprise Linux or Novells' Suse Linux Enterprise Server;
this is obviously because they are commercially supported systems.
The current validated configurations can be found here. A full press release is also available.
At the beginning of the year I looked at the installation
of OpenBSD as well as configuration of firewall and VPN
services. OpenBSD is one of the most secure operating systems out there, they
proudly claim on the front page of their website ?Only one remote hole in the default install, in more than
8 years!? which is quite something.
So how do the OpenBSD team make sure that OpenBSD remains secure? As with
most open source projects/groups, full disclosure is a key belief and practice.
Many companies will try to hide potential security flaws and bugs in an effort
to reduce publicity, and therefore exploitation, before their patching schedule
comes around. This approach is, however, fundamentally flawed?information of
new vulnerabilities spreads like lightening within the underground groups who
would exploit them. Without full disclosure, users are not even aware that a
potential issue exists?at least not until they fall victim to it. On the other
side of the fence, full disclosure means that most issues (especially security
vulnerabilities) can be fixed within an hour or so of being discovered.
Since 1996 the OpenBSD team have been proactively auditing source code of
every critical software component. This file by file analysis is looking for
bugs rather than security holes; however, most security holes tend to be a
result of bugs. Code is often re-audited when new types of flaws are
discovered?they are also audited by different people, each holding different
auditing skills. This proactive approach means not only that very few bugs
appear in OpenBSD, but also that vulnerabilities are often discovered and fixed
before they can be exploited by others, unlike many other operating systems.
OpenBSD also claims to ship as ?secure by default?, what they mean by
this is that all non-essential services are disabled by default. The thinking
behind this is twofold?firstly, if an administrator needs a service, he will
enable it. If the service isn?t needed, then why run it? If a vulnerability is
found in the NFS daemon, which you don't even use, why leave it running and
expose yourself to unnecessary threats? The second positive feature of this
practice is that while reading the man pages and learning how to run/configure
the service, the administrator will hopefully become aware of the security
implications and best practices?reducing the chances of misconfiguration that causes
a weakness. How many vulnerabilities apply only when a system is left with the
?default configuration??
OpenBSD may ship as secure by
default and have a great track record for bug fixes, but how do we make sure
our systems are kept up to date when a new bug or vulnerability is found? Next
week, I will take a look at the OpenBSD security advisories, the channels by
which these advisories are released, and finally downloading and applying
patches.
by jaqui
There are only a few issues that most people would have with OpenBSD.
The last two items do have more of an impact than it may appear, since the percieved lack of end user applications for linux is one of the drawbacks of linux, the minimally smaller number for OpenBSD will seem to be much smaller than it is. OpenBSD is slower than linux at getting support for new hardware, because of the OpenBSD stand on security. This cuts both ways, you do have a secure os, but you probably won't have drivers for your 6 month old hardware.
I myself was not impressed that OpenBSD couldn't make use of an ext2fs partition, as an ext2fs partition, for installing it on. OpenBSD required a special bsd formatted partition to install on. I think that this needs to be re-examined by the OpenBSD team to bring OpenBSD more into the mainstream open source community by improving the compatability between the operating systems.
by apotheon
As Jaqui pointed out, there are some issues with using OpenBSD rather than a Linux distro. I'll address his three points, and add some of my own:
If you have some purpose for OpenBSD that is handled well by OpenBSD, though, it is definitely worth getting acquainted with the OS. Firewalls come immediately to mind as an excellent reason to get familiar with OpenBSD. Its security characteristics are nothing to sneeze at.
by jaqui
"The installer is harder to come by than Linux installers.
You either have to be the sort of expert that can build an installer
ISO yourself using the official OpenBSD sources, or you have to pay
money to Theo DeRaadt for his "official" installation CD."
Actually Apotheon, I was just looking at OpenBSD, and they have network install from mirrors, as does Linux. There is no longer a need to build an install iso for it, or buy the cdroms from them for a cd install.
As discussed last week, OpenBSD is an operating system with
its focus firmly set on security. Unused services are disabled by default,
continuous auditing of source code takes place, and any bugs discovered are
given full disclosure.
Most exploited vulnerabilities have actually been fixed; the
problem is that not all systems are kept up to date by administrators, either
because they are not aware that a bug and fix exist, or because they simply don?t
know how to apply the fix. Hackers actively search for systems like this. Unpatched
networks are an easy target and can often be compromised without detection, because
a system which is not kept up to date is also likely to be a system which isn?t
closely monitored. As OpenBSD systems are usually chosen to provide Internet
facing services, it?s vital to keep them updated.
So how do you know that a new patch is available and whether
or not it should be applied? The most obvious way to find out about released
patches is to check the errata and patch list on the official OpenBSD website. The release
versions are given their own sections, each containing both Security and
Reliability fixes. A much better way to stay on top of patch releases is to
subscribe to the official security-announce
mailing list. This will mean you are proactively notified when a new
security issue arises, rather than relying on yourself to revisit the errata
page at regular intervals.
You may notice that there are two types of fixes addressed
on the errata pages?Security and Reliability. This is very simple--Security
fixes patch against a bug which has a direct impact on security; it?s vital
that these are applied. Reliability fixes are just that: they address issues that
may affect the reliability of services, but do not pose a direct threat to
security. It isn?t necessary to apply all patches; for example, you may not
want to bother patching against a bug in sendmail if you don?t actually use
sendmail on the system. Needless to say, this also applies on an architectural
level. Why apply a patch which only addresses an issue in the x86_64 source, while
you?re actually running x86?
Patches can be downloaded either individually or as a
package. If you want to download individual patches, they can be found in the /patches/{version}
subdirectory of the main OpenBSD FTP root. A tar.gz package of all current
patches can be found directly in the patches
folder. I personally prefer to download the archive when first installing a
system and then add individual patches as they are released. Don?t forget that
both common and platform-specific patches need to be applied (e.g., both those
in the common and i386 directories).
Before any patches can be applied, the kernel and system
source must be installed on the system (I tend to install this during the
initial system setup). If you don?t have the source installed it?s a very
simple process. Download the files src.tar.gz
and sys.tar.gz from the OpenBSD ftp
mirror of choice. The files are found in the base directory of each
release. If you are using any ports or the X windows system, XF4.tar.gz and
ports.tar.gz should be downloaded from the same location.
We now have all of the information required to patch our
system and make sure it?s as secure as possible. Next week we?ll pick up right where we?ve left
off and start to get our hands dirty.
In the past few months I have discussed the various threats posed by
portable storage, particularly USB keys. I recently ran across an
interesting account of a social engineering project designed to test the
information security of an anonymous firm. The plan was very simple and
frighteningly effective. Auditors had a small trojan written which
would collect passwords and other sensitive information--then email this
data back to them for analysis. The trojan was designed to be hidden
amongst image files and planted on 20 USB keys. These keys were then
scattered around regularly frequented areas of the firm (car park,
smoking areas etc). Out of the 20 keys, 15 were found by employees and
all of those 15 were used, activating the planted trojan. This surely
shows what a threat USB devices can pose to the security of corporate
networks and the data they carry. A full report can be found here.
In one of the follow up comments to this article there has been some
discussion on how to stop this type of threat. One suggestion has been
a piece of software called gatekeeper--this looks like a
pretty useful application with some interesting features.
After a spate of embarrassing and potentially dangerous incidents within
various civilian government agencies, the Bush administration has
released an official memo stipulating
that all mobile devices (notebooks, smartphones and I would imagine USB
storage) which carry sensitive data must be encrypted. Remote access
should only be given with two-factor authentication--one of the factors
must be provided by a device separate from the computer gaining access.
Remote access and mobile devices should time-out after 30 minutes of
inactivity. The document also dictates that all data extracts from
databases holding sensitive information should be erased within 90 days
unless its use is still required.
It seems hard to believe the statement inside this memo which reads
"Most departments and agencies have these measures already in place"
given the recent incidents which have no doubt prompted the memos issue.
SecurityFocus has a deeper analysis available and the official memo
is found here.
by nathank9
Wow...I must applaud the government for FINALLY implementing this.
These new rules and regulations are a start and hopefully can only get
stricter over time. The next step we need to take is to move this
beyond a "recommendation".
I have been up to date on nearly every security breach over the past few months and according to:
http://www.privacyrights.org/ar/ChronDataBreaches.htm#CP
the total number of people affected has been 88,795,619 since February of 2005.
While
the government has done a good job in finally establishing this as a
major National problem, it took them far too long. Why did they wait
until nearly 90 MILLION people were affected before this. It's not too
late, but it coulda/shoulda been done quite a while back.
Sadly,
a majority of these data breaches were caused by the government's lack
of emphasis on security so establishing these laws is the least they
can do. While the government continues to ammend their policies on
security and encryption they should read up on some recent facts of
email encryption:
http://www.essentialsecurity.com/Documents/article15.htm
The Washington
Post reports that on June 29th officials announced that the
Laptop stolen from an employee of the Department of Veterans Affairs had been
recovered. The Laptop hosted a database
containing the personal details (names, dob, social security numbers etc) of
millions of current and former personnel.
The recovered laptop and external hard drive is now in possession
of law enforcement agencies who will be conducting a forensic analysis, they
claim that ?there is reason to be optimistic.?
The stolen equipment was returned after the advertisement of
a $50,000 reward. FBI agent William D
Chase said that it?s too early to say whether the person will be given the
reward or face criminal charges?however an FBI spokeswoman said that the
tipster is not a suspect.
by nathank9
So in complete honesty, this is the best thing that could have
happened, saved quite a few people their jobs and quite a few people
their identities. However, this discovery is a complete false
sense of relief.
On June 28th, the government learned that
they had 45 days to implement a new policy on their laptop security.
The government has been so lax with their laptop security these days
that leads me to believe that this is not the last incident we will
hear about with the government. We need to improve our data security,
specifically laptop security or else our identities will be in even
worse shape.
After forensic analysis of the previously stolen VA Laptop,
the FBI
claimed it has ?determined that the data base remains intact and has not
been accessed since it was stolen?. This
is a very interesting statement; questions of how the FBI would be able to
determine this have started to be asked.
While reading an article on Slashdot I found this interesting
discussion of the possible steps which the FBI would have taken to analyse
the Laptop and data.
The problem is, neither of these method can be called
reliable. If the Laptop was stolen by
professional fraudsters, specifically for the purpose of identity theft, they
would be well prepared. Physical
examination can easily be cheated, latex gloves and plastic screwdrivers being
the tools of choice. However, there is a
method of stealing the data without opening up the computer, or even booting
from the hard disk (therefore meaning a-times will not be altered at all)?by
simply booting from a Linux live CD like Knoppix, the internal hard disk can be
mounted as read only and then an exact copy made to either a removable disk or
network share. The duplicate disk could
then be used to access information and the Laptop returned to the
authorities.
The first thing the FBI will do is make a 1:1 duplicate of
the disk so that investigative work does not have any effect on the
original. I really don?t know why they
think a professional data thief would do any different.
by dr dij
right on! I think this was bluster by the FBI to help white-wash the theft.
(oh, it's OK, since we got it back! ) yeah, right.
by mroonie
Last week I looked out how we should keep ourselves informed
of any new patches for OpenBSD, and how to find the correct patch files once
they are released. This week, we?ll get straight into the application of these
patches.
The full-source archives should be downloaded to /usr/src
directory and the contents extracted; having the source installed is, of course,
a prerequisite of any patch application:
# wget ftp://ftp.openbsd.org/pub/OpenBSD/3.9/src.tar.gz
# wget ftp://ftp.openbsd.org/pub/OpenBSD/3.9/sys.tar.gz
# tar -xzf src.tar.gz
# tar -xzf sys.tar.gz
And let?s get the patches:
# cd /root
# wget ftp://ftp.mirrorservice.org/pub/OpenBSD/patches/3.6.tar.gz
# tar ?xzf 3.6.tar.gz
Applying the patches may seem daunting, but it is actually
very simple. Each patch file contains concise instructions of what must be done
to apply it?an example:
ftp://ftp.mirrorservice.org/pub/OpenBSD/patches/3.6/common/005_isakmpd.patch
This patch is for OpenBSD 3.6 and addresses a reliability
issue where ISAKMPD may have problems communicating with other key exchange
implementations. If we look into the file:
# cd 3.6/common
# head 005_isakmpd.patch
Apply by doing:
cd /usr/src
patch -p0 < 005_isakmpd.patch
Then rebuild and install isakmpd:
cd sbin/isakmpd
make obj
make depend
make
make install
Let?s follow those instructions and make the patch:
# patch ?p0 < /root/3.6/common/005_isakmpd.patch
|Index: sbin/isakmpd/nat_traversal.c
|===================================================================
|RCS file: /cvs/openbsd/src/sbin/isakmpd/nat_traversal.c,v
|retrieving revision 1.7
|diff -u -p -r1.7 nat_traversal.c
|--- sbin/isakmpd/nat_traversal.c 8 Aug 2004 19:11:06 -0000 1.7
|+++ sbin/isakmpd/nat_traversal.c 18 Nov 2004 18:25:35 -0000
--------------------------
Patching file sbin/isakmpd/nat_traversal.c using Plan A...
Hunk #1 succeeded at 58.
Hunk #2 succeeded at 232.
Hunk #3 succeeded at 241.
Hunk #4 succeeded at 254.
Hunk #5 succeeded at 304.
Done
# cd sbin/isakmpd
# make obj && make depend && make
Check that everything looks okay, and then install with the
final command:
# make install
You can see that, in fact, the patching process couldn?t be
more self-explanatory. The only thing to notice here is that if you don?t keep
patches directly in the /usr/src directory, then the patch reference will have
to be changed to reflect its location. After the ?make install? command the
processes will need to be restarted; this can either be done with a reboot, or
by killing/restarting the running instance.
Best practice, of course, suggests that patches are not
applied directly to a production machine?in fact, a great number of
administrators would not even leave source on a production box. Instead,
compile the updated binaries (as described above) on a development machine and
then once compiled and tested, they can be copied over to the production
system. The output from ?make install? will show you which binaries and files
have been replaced, thus telling you which files to copy across. I personally
use a VMWare-hosted virtual machine to compile patches/updates.
After the initial few patches this becomes a fairly
trivial process and there really is no excuse for not keeping up to date with
?security? tagged releases. I hope this has been useful for anyone who wasn?t
quite sure how to apply patches to their system. If there are any experienced
users who have tips or advice on keeping things up to date, please leave a
comment.
Sophos
security has recommended that users concerned about security should
consider moving to Mac.. The top 10 malicious software applications are all
targeted at Windows machines, none of which are capable of infecting a machine
running Mac OS X.
It is generally accepted that the reason for this is that
the majority of computers around the globe run Windows; therefore targeting
Windows machines is more profitable for creators of Malware.
Sophos say ?It seems likely that Macs will continue to be
the safer place for computer users for some time to come?, however Brian
Gammage of Gartner does not agree, saying ?If you have smaller walls, you
attract less graffiti... There is nothing architecturally safer about Macs. If
everyone moved to them then the situation would change overnight".
Both opinions are well founded, indeed if everybody switched
to Macs overnight then we may well see malicious software developers shift
their focus, but the reality is most users in the foreseeable future will stick
with Windows, therefore Macs will still offer a safer alternative for the few.
by yobtaf
by it-slave
by billtahoe
Brian Gammage of Gartner says "...There is nothing architecturally safer about Macs..." I disagree. I think that there is significantly greater architectural safety in the Mac OS. It would be interesting to see some impartial expert analysis of the relative architectural safety of these two OSes. Most of what I have seen is far from impartial, but from the references I have been able to follow (MCSE for NT and MCSA for W2k with 30 years making a living working with computers, Macintosh at home) the Mac OS seems to have a large advantage.
Over the last few years and months, there have been many
reported incidents in which confidential information relating to customers or
employees has been leaked, stolen or lost. Various legislation ties to set standards
for how data should be collected, stored, and accessed?including the Data
Protection Act, Human Rights Act (in the UK), and other industry-specific
regulations. These standards define which activities are allowed and who is
authorised to carry them out. It?s important to define a clear information
security policy for your organization, particularly if you are in the financial
services sector.
The security of information can be breached in any number of
ways be it via hostile attack (hacking or physical theft), sloppy handling
(lost tapes or memory sticks), or leaking via an insider. Ensuring the security
of information not only needs to consider the risk of data theft but also loss
of integrity?information held by a company is a valuable asset which needs to
be protected from human error, hardware failures, and other potential
disasters. The loss or theft of sensitive information (such as customer
records, financial information, and employee data) can be a major embarrassment
to a firm?a few examples of data lost recently along with the number of people
potentially opened up to fraud/identity theft (thanks to privacyrights.org):
So what are the aims of defining a policy and what are the
consequences of its creation? There are two core principles behind the
definition of a security policy?the first is to define the relationship between
privilege and responsibility. Staff should clearly understand what is and what
is not allowed; the responsibilities which come with privilege must be defined
in clear guidelines. The second function of this policy will ensure that a firm
can react in an appropriate manner should any incident occur.
Next week I?ll take a look at what should be detailed
within a security policy and the various aspects of information security. If
you have any comments or suggestions?maybe experiences from drafting your own
security policies?then why not leave a comment and share your knowledge?
by nathank9
SUSE Linux Enterprise 10 has been officially released?So
what?s featured in this new release?
To read about the entire host of new features and
functionality visit the official website.
RFID tags are being used more and more frequently to track
products, shipments and stock levels. As
prices drop their use is becoming more widespread, it is expected that
eventually RFID tags will replace conventional barcodes (this does however
raise questions of consumer privacy rights)?pet owners can even have a tag
implanted in their pet for identification purposes.
Last week, researchers in Vrije
University (Amsterdam) wrote a proof of concept virus and
demonstrated how it can be used to corrupt related databases. The virus was planted in the RFID tag, when
the tag is read the virus is transferred and trouble begins. This would have huge implications, activists
could target stores by infecting tagged items therefore potentially destroying
or modifying stock data?luggage control in airports would also be a vulnerable
area.
The BBC has some
useful information on RFID.
by jdw242
by janta27
This is not a blog, this is history, published by the Vrije Universiteit (VU) in Amsterdam on March 15th (fortunately of this year, yes). What researchers at the VU did publish however this week is that they are about to launch the RFID Guardian, a device warning you when you are being sniffed at by an RFID reader.
JanT
Last week I looked at some of the high-profile data leaks
over the last year?it?s clear that enormous numbers of people have had their
personal details put at risk because firms fail to properly protect them. Today,
identity theft is rampant and all measures should be taken to protect people
from criminals, both by having the proper measures in place to eliminate the
risk of unauthorised access and by having processes in place to minimise
exposure should the worst happen.
The Joint Information
Systems Committee (JISC) defines the three key aspects of information
security as:
It seems to be widely accepted that a high-level document is
the best approach when developing a policy?this means the document should be
short (but also comprehensive), easy to understand, and only contain the key
points to be understood by those required to comply. Inclusions would be
definitions of responsibilities, limitations, emergency procedures, and the
consequences of failing to comply with these requirements. Ideally a policy
should avoid being tied in to any particular system or technology; rather it
should give a well rounded overview which will not need to change as technology
evolves. If required, separate guidelines and procedures can be referenced,
dealing with more specific areas of concern. Above all, the most important
consideration should be that the policy needs to be enforceable.
To summarise, the JISC
provide a very useful set of bullet points to show what must be included to
ensure the construction of a comprehensive policy:
The guidelines for day-to-day security practice can include
many areas of security, from user access and password guidelines to the backup
of confidential data. Other areas include the use of encryption for wireless
access, mobile storage, and site-to-site tunnels; firewall and anti-virus
requirements for individual machines; and the types of attachments that are allowed
to pass through an email system. It's also a good idea to define procedures for
keeping systems up to date with the latest security patches, therefore ensuring
that security won?t be breached by someone exploiting a known weakness (this
could be costly, not to mention very embarrassing!).
There is a great deal of reference material available to
help you construct an appropriate information security policy for your
organisation??Writing
Information Security Policies? by Scott Barman contains a wealth of
information covering all areas of an information security policy--physical
security, Internet and Email, viruses, and encryption. In this book, Scott also
covers maintenance of policies, acceptable use and enforcement. As I said
before, the JISC provides some very useful guidelines on Information Strategies,
and even though it's geared to give guidance to higher education institutions,
the information is directly applicable to all other sectors.
I hope you have been inspired to craft your own information
security policy, or will be polishing up one that you have already. If you have
any good resources to share on this subject, why not leave a comment and share
them with us?
by mroonie
It's great to see some blogging going on about this issue. Internet security has definitely become one of the areas that many companies have let slide through their fingers when it should be a matter of immediate attention. Fortunately, there are a good number of companies who have taken the matter seriously and have found great solutions for some of the problems mentioned....
"user access and password guidelines to the backup
of confidential data. Other areas include the use of encryption for wireless
access, mobile storage, and site-to-site tunnels; firewall and anti-virus
requirements for individual machines; and the types of attachments that are allowed
to pass through an email system."
Companies should definitely take the time to reevaulate their current security measures and reiterate a new and improved security policy that give some attention to this much deserved issue.
Sort by Trending