Now with KDE 4.4.3/PolicyKit sauce …
|
|||||||||||||||
|
A recent posting on the Blog of Helios prompted me to write a short and simple definition of Linux that might be useful for current non-users of this operating system. It is however a difficult definition in the context of what people already know. And the fact of the matter is that what the general computer-using population knows about operating system platforms, is limited. I often get calls from Windows desktop users about something not working. What is not working? “Well it’s something in Microsoft” they might say. Microsoft? Is that Microsoft Office, Windows, something else? Well they’re not sure, but it’s when they are trying to type a document. Ok, so that’s Microsoft Office Word then. What browser are you using? It’s the one with the blue e on the icon. Internet Explorer. What email client do you make use of? Microsoft. Is that Outlook or Outlook Express? Note I’m just using the Windows platform as an example, however this problem is not limited to that platform. If general Windows users have difficulties on that platform, what chance do they have with Linux? And why the confusion in the first place? Let’s try to answer these 2 questions … What is Linux Linux is a software platform that includes a kernel which controls and manages the computer itself, utilities which allow you to perform general tasks like file management and application launching, and applications themselves which allow you to get actual work done ( eg. word processor, email client, web browser ). The original operating system ( to be exact the kernel ) itself was started by Linux Torvalds, a Finish student in 1990 who was frustrated with the licensing of another OS called Minix. Together with the GNU toolset ( a bunch of OS-independent user tools ) and development tools, GNU/Linux as a complete operating system platform was born. Unlike Microsoft Windows, which comes in only 2 forms ( desktop and server ), Linux is packaged in the form of distributions, which put the Linux kernel, GNU utilities and other useful applications together. There are many distributions, some which cater for general use, some orientated towards audio-video use and others for supercomputing purposes. There are about 10 to 20 distributions which are used in mainstream desktop and server environments, the most popular of these being Ubuntu, Mandriva and Fedora ( for desktop use ), and Red Hat Enterprise Server/RHEL, Centos and Suse Linux Enterprise Server/SLES ( for server use ). An important difference vs commercial OS platforms is that Linux distributions typically provide all the day to day applications that you would use, therefore it’s fundamentally different to something like Microsoft Windows, where you only get the operating system and some utilities. The Linux kernel itself and the GNU toolset are FOSS – free and open source software. This means that although they have a license and are copyrighted, the style of the license means anyone is fee to copy, use and alter this software, as long as one keeps to the terms of the license. Typically this includes something as simple as making sure the license is transferred with each copy, and that original and subsequent authors are acknowledged. But how can you give something away for free if it’s copyrighted? I’ve been paying for my Windows and Office software all along … Copyright fundamentally means that someone can assert the right to be acknowledged as the author of a particular creation. It does not infer that something can’t be given away for free, as much of the bumpf from music, movie and publishing concerns would have us believe. So yes you can have copyrighted software that is free. There’s also the misconception that Linux is difficult to use. From a server perspective, this may have some validity ( although not much ), however, from a desktop point of view, Linux is as easy to use as competing platforms like Microsoft Windows and MacOS X. It’s just different – and it’s this difference that many confuse with difficult. There’s also the matter of change – human beings are comfortable with what they know; change is never easy because of this. One important point to remember though is that because Linux is a different platform to Microsoft Windows, it will not run Windows applications natively. Most Windows applications have an equivalent in Linux so this is not a big problem. There is also the possibility of running Windows applications under emulation. Some examples of FOSS application equivalents:
Linux has some distinct advantages over other platforms:
How do I get support for something that is free? FOSS support is provided by the same community that develops the software as well as the user community around it, through forums, newsgroups, mailing lists and other methods. If that is not suitable, then many of the larger FOSS projects have commercial support options available. FOSS in general The Helios project is a group of volunteer Linux users in Texas, USA who refurbish old donated computers, install Linux and other FOSS applications on these machines, and deliver them to needy, impoverished and foster kids in that state. The financial cost to these volunteers is low because FOSS allows them to have an almost zero product cost. This is something that’s not possible with commercial software. And there are many other groups around the world that do work similar to the Helios group. FOSS lowers the entry barriers to less fortunate people and communities, removing what is arguably the biggest cost of owning a computer – commercial software. This helps with social development, upliftment and education, by giving less fortunate people access tools they would not have had before, allowing them to create, communicate and distribute. While FOSS and Linux are typically ‘free’, this does not mean that the quality of this software is compromised in any way. In fact it’s well acknowledged that FOSS software is generally of a higher standard than commercial software, due to the nature of the Open Source development process. A study by Coverity ( a commercial software vendor of code analysis tools ) in 2009, found that the Linux kernel and some other notable FOSS projects, had 10x less code errors than competing commercial equivalents. Why the confusion? At the start of this article, I asked why users were confused about what OS or applications they were using. An analogy: to drive a car on a public road, one needs to do a drivers test. This involves theoretical and practical training, after which one has a reasonable grasp of the concepts involved as well as some baseline experience to use in the act of driving itself. Using a computer is an altogether different proposition – one goes to the computer store, buys the computer and starts using it. This does not mean however that one is proficient in the use of that computer, and therefore the lack of general knowledge amongst casual computer users. This issue is platform-independent, yet the stigma remains that FOSS and Linux are more difficult to use. Difficulty is not necessarily determined by what platform you use, but rather the training you receive in the use of that platform. So take the time to learn something new today, about whatever platform you are using … The number of articles lately concerning the overwhelming amount of Linux distributions available is quite interesting; why now? Perhaps because Linux on the desktop is becoming a little more mainstream. Or perhaps because of the miriad number of embedded devices that use Linux ( think home automation, signage boards, NAS storage, netbooks, phones, radios, media players, industrial devices, etc. ) and that are more visible. Whatever the reason, I think it’s always been a positive thing, with the competitive nature of open source driving improvement and maturation. And open source does not need proprietary offerings as competition ( as another journalist has recently inferred ). FOSS in itself is its own competition. Just because there are numerous distros available ( distrowatch.com keeps track of the Top100 ) doesn’t mean Joe Blogs needs to become a guru on all or most of these. Many are specialist distros catering for recovery, security, audio, media or other areas and for the most part, one can keep your eye on the top 20 for something to use within mainstream server or desktop areas. For desktop use, Ubuntu, Mandrake, Mint, Fedora, Puppy and openSuse come to mind. On the server side Centos, RedHat, SLES, Debian and Slackware are the main choices. For Business server use ( ala Windows Server SBS ) you can look at ebox and ClearOS. So overall, that is not an overwhelming set of choices – just enough to give you something you are comfortable with. Read reviews on each, do a test run in a virtual machine ( care of VirtualBox ) and make the switch. Here follows a quick matrix showing the main distros in each area.
I recently came across this application ( thanks Nic ) so only natural that I try it out ( especially as I have an issue with the plasma twitter widget in KDE 4.2.4 ). Couple of issues though:
I did some searching and with a couple of articles, managed to piece together an installation routine for Slackware64 13.0:
No too bad an install but it would be nice for Adobe to support non-rpm/deb users with a .tar.gz package and 64-bit Linux. Maybe to come? It’s been a while since my last ‘Interview’ with Slackware and we’ve just had the new 13.0 release ( 26th August ), so time to brush up on Slackware as one of the premier distributions in the Linux landscape. I wasn’t quite sure where to start with this article as all the goodness of previous releases shine through in this release. So I went back to my previous article to see what I had said and what I didn’t. I’ll keep to the differences with 13.0 and start with a quick rundown on the distro from a practical point of view. Check my 12.2 article for a more general overview of Slackware. Features:
Installation: very little has changed here and once you get over the initial task of manually partitioning your disks, you can start the excellent ncurses-based installer and go from there. While the installer may appear somewhat complex to newbies, it’s more a case of giving choice, especially in terms of package selection. In addition, Slackware remains one of the few distro’s supporting mirrored boot disks ( yes that’s including the /boot partition/directory and swap if you want ). This is a manual process but at least you can be sure your whole root volume is protected ( I have a howto if anyone is interested ). Unlike 12.2 ( which [ may have ] had a slight issue with the Gutenprint package during install ), a number of installs of 13.0 to different hardware has yet to show any problems. Just make sure your ISO images are valid when you write them to disk ( use the checksums after downloading ) and write at a low speed to ensure compatible disks. Core: we’re now at 2.6.29.6 for the kernel and besides the usual array of driver updates and additions, filesystems are a big area of development. Specifically both ext4 and btrfs are included. I can’t recall doing fresh install for a particular machine but the root partition appeared to be automatically upgraded ( I’ve been following current in a single stream since 12.2 was released ) with the following parameters added – barrier=1,data=ordered. There have been a number of reports, especially from other distros adopting ext4 at an early stage, concerning file-system corruption and data loss following crashes. This mainly is a result of the change of default data journaling modes from ext3 to 4 and some apps that do not fsync data correctly. Suffice it to say that ext3 and 4 essentially have the same journaling modes and it’s only the default that has changed in ext4. Slackware 13.0 uses data=ordered mode to keep the default ext3 behaviour ( in keeping with its stability orientated vision ). You can always change this – this is a performance vs data consistency discussion and one better served in other articles – more here and here. Note that while ext4 has been getting all the attention, ext3 has been receiving some latency performance improvements in the kernel so don’t write it off just yet. Kernel 2.6.30.5 is available in /testing. udev pretty much works as advertised with automated detection and mounting of all manner of devices. The addition of the xz compression protocol is due to its high compression factors ( like bzip2 ) with the performance of gzip. This allows a corresponding reduction in size of the distribution disks which is always handy. Documentation for this release has been updated and those upgrading from a prior release need to take a good look at the UPGRADE file. For those new to Slackware, there are a number of documents in the root of the distribution disk relating to installation, which are important to make your first installation as smooth as possible. Read that manual. The slackbook project appears to be back on track with a revised edition at www.slackbook.org, a good starting point for all those new to Slackware. Networking: everything and the kitchen sink – need I say more? Yes maybe: The LAMP stack is now completely up-to-date with Apache 2.2.13, php 5.2.10 and MySQL 5.0.84 ( yes, 5.1 is the main stream now but 5.0 is still regarded as the ‘stable’ release ). I used to compile my LAMP stack due to older packages in the distro but this is no longer a necessity seeing as all these packages are bang up to date. The only requirement you may have is with certain support in PHP – I haven’t checked what support is built in the provided package yet – update to come. In addition there is updated wifi and bluetooth support ( in tools and kernel ); about the only thing I’m missing is netcat ( netwrite is included but I’ve yet to try this one out ) and libnetfilter/libnfnetlink for connection tracking on iptables. Applications: a number of new apps have made their appearance with updates to mostly everything else. Of interest is the mplayer plugin for browsers, Glew ( for MESA ), KDE 4.2.4 of course ( can’t miss that one out ), dc3dd ( a disk forensics tool – aka dd on ‘roids ), urwid for wicd, some new aspell internationalisation packages, tightvnc ( hmm interesting one ), Intel GPU tools, the taglib extras package ( for mp3 tagging apps ), thumbnailing support in KDE for videos using mplayerthumbs ( Dolphin and Konqueror ) and snownews ( a console rss reader ). There are also a number of rebuilt packages with support for xz compression, like tar and all the packaging utilities. There has been some discussion in the forums as to why KDE 4.3.0 was not included but there appears to be some issues surrounding PolicyKit integration in the core OS – Robby Workman mentioned compiling both with and without PK, so I’ll just leave that storyline until -current starts up again. The KDE4 version of K3B is still a bit flaky so the KDE3 version is included along with the necessary KDE3 compatibility libraries. Development tools are all upgraded with Perl 5.10.0, Python 2.6.2 and version control tools like git and subversion. Firefox is upgraded to the new 3.5 stream while Thunderbird remains at v2 seeing as v3 is still in beta ( I’ve been testing nightlys since b2 and it looks good although I’ve a few Bugzilla reports open ). Other areas of 13.0 haven’t changed much in terms of usage and you can check my 12.2 article for more general details. One area I will touch on is Gnome and I’m possibly being a bit ambitious here – running GSB 2.26.3 ( compiled for 12.2 ) on 13.0. For the most part, everything works fine but I’ve got a nasty performance degradation when running any Gnome apps. The X process clobbers one of my DualCores, bringing KDE/plasma specifically to a halt and continues to do so for about 2 minutes after I exit those apps. There was a previous issue with the Xorg 1.4.x server in 12.2 with the event handler code that caused similar issues but this should not be a problem with Xorg 1.6.3 in 13.0. I’ve spoken to both the GSB guys and Robby Workman but a resolution is still hidden to me ( thanks for the help Robby! ). Package management and installation: There is now a whole host of tools, internally and available from 3rd parties supporting package installation/tracking. My favourite at the moment is src2pkg but there’s also slacktrack ( in /extra ), build scripts at slackbuilds.org and checkinstall, amongst others. For package installs and updates, slapt-get, pkgsrc ( is anyone still using this? ), slackpkg and swaret provide the goods. I still don’t see Slackware’s internal lack of dependency checking an issue, but to each their own. Out to Pasture: the old raidtools package is now gone – I think this functionality is now all handled by mdadm ( now available from the boot disk to assist with pre-OS installation RAID setup ). Others include fvwm95, the older libxml and bison 1 packages, the Berkeley db1 and db3 libs and some printing utils like lprng ( handled by CUPS ) and ifhp. Conclusion: Slackware remains my favourite server distro and personally stays on my desktop too ( it takes a bit more work here than some other distros perhaps but no matter ). 13.0 comes with almost everything you would expect in terms of applications, all at the latest versions and solid as usual. As a development workstation or LAMP/application server, I think it’s untouched. While it may not fill everyone’s expectations, you owe it to yourself to take a look if you haven’t before. For existing users, this is just another solid release. All that’s left to say is thanks to Pat and the rest of the gang – your effort is well appreciated. In conclusion, a parting quote from fortune: “It is easier for a camel to pass through the eye of a needle if it is Updates: 1. GSB 2.26.3 ( -current ) is now available in 64-bit form for Slackware 13.0 – announcement here 2. Techiemoe’s review 3. Why you should try Slackware – LinuxMag interview with Eric Hameleers 4. 32-bit support in Slackware64: AlienBob ( Eric Hameleers ) has both done the work of providing 32-bit support as well as written a great instruction manual for installing 32-bit support in Slackcware64. A quick note that after installing all the provided packages, it’s a good idea to restart your machine so that the appropriate Pango modules are built ( I ran into a problem with Thunderbird 32-bit which was sorted after a reboot ). 5. install your multilib support before your binary video drivers 6. K3B 1.66.0-alpha appears to be more solid than the supplied KDE4 K3B This is a surprisingly large ecosystem for an area of OS’s that would on the face of it seem very specialised. I think there are a few facts that are bringing these features to the mainstream: 1. we’re looking at new IT innovations like cloud computing 2. management interfaces to provide these features are being developed and simplified 3. features have stabilised Distributed filesystems include GlusterFS, Lustre, GFS2 and Hadoop while clustered filesystems include OCFS2, GFS2 and DRBD. Red Hat Enterprise includes GFS2 allowing clustered storage solutions through simple gui tools and DRBD is now available in the canned Openfiler storage solution ( not simple to setup but there is a good tut at howtoforge ). GlusterFS uses a very interesting technique of aggregating storage bricks ( using items like iSCSI, DAS, AoE and SAN-supplied LUNs ) in mirrored or other RAID configurations providing redundancy for data across your storage environment. Setup does not involve kernel modules or changes so it’s fairly straightforward. ZFS is also included in the latest FreeNAS rc snapshot meaning you can slice and dice your storage exactly how you want it with writable snapshots, storage pooling and simple administration. The upside of all these storage applications means you can provide non-stop storage across an enterprise at a very low cost and simultaneously aggregate performance for those high-bandwidth apps. Virtualisation of storage, previously the domain of high-end commercial companies like Netapp, StorAge and Falconstor, is now available for little to no cost through a whole host of FOSS and commercial OSS software. This probably sounds like quite an anti-climax, but considering its 60% year on year growth, this is quite an achievement. And Windows has dropped 4% in the last year, which is a considerable dent in its market share. This together with the erosion of Internet Explorer’s browser market, indicates a general downturn for Microsoft. The desktop remains a very important piece of the puzzle for MS – without a desktop market, they don’t have anywhere to run their server applications. And those 2 areas remain MS’s biggest markets. This is of course a good thing as it means that Microsoft has less of a say in world computing circles in which it traditionally has driven events in directions only advantageous for itself. Slackware, in my mind, has never been a Linux distribution for novices who aren’t interested in learning about the whole Linux ecosystem. Note that’s not a criticism, just a comment. And in that comment lies the issue with a lot of the Slackware articles that have appeared over the years. Many of them appear to be authored from the point of view of a novice or the person who just needs a straight desktop-orientated distribution; I’m not saying Slackware is difficult, just that it requires a better understanding of Linux as a whole rather than the desktop alone that most are used to. At the same time, Slackware is probably amongst the best as a learning tool for Linux, if you have the interest. It’s been said: To learn Red Hat is to know Red hat; to learn Slackware is to learn Linux Slackware is best suited as a server in my mind, but with some work makes a fine desktop and laptop system too. And now onto 12.2 itself. Installation: Much has been said about the ‘archaic’ Slackware installer but the fact of the matter is that it’s still simple and bulletproof, providing most of what’s needed for the base install. Everything is covered from install location, file system type and swap setup, to network setup, boot loader installation, GUI selection and package selection. Granted it’s not pretty like most graphical installers, but it just works. The only real pre-install requirement is to make sure your disks are partitioned correctly. That slightly advanced requirement for novices, aids though when you need to do complex installs such as mirrored boot disks, something that remains missing from most other mainstream distros. Post-install tasks include creating users, adding users to groups and configuring X amongst other things. I’ve been running current ( development ) on one machine since 12.1 and as such, the difference is nothing much that you’d notice. Just more of the same reliability and stability that you’re used to with Slackware, and updates to most of the popular included packages. A different machine running 12.1 upgraded cleanly and without fuss using Patrick’s recommended method, slacktrack and slapt-get. It has been mentioned by some users that they’ve experienced pauses during the install. The only attributable factor I can find is the compilation of Gutenprint driver files during that package’s install. I’ll look into this closer with subsequent installs. The core: The included kernel 2.6.27.7 brings improvements in filesystem ( XFS, JFS, EXT and Reiser ) and hardware support, and other areas like virtualisation ( KVM is now nicely embedded with a good feature set ) and wireless. Udev and HAL provide a neat solution to hardware detection and work well in the Slackware environment providing you follow the rules for adding users to the appropriate groups. Recent versions of Udev have differing places for configuration files so make sure you read the provided documents. Detection in the provided GUI’s work flawlessly providing users with fool-proof access to pluggable and other devices. Sound devices are well supported too with the kernel modules being automatically loaded, immediately after installation. The BSD-style initialisation mechanism is easy to understand and master, besting some of the more complex systems used in even popular distros. This extends to the venerable inetd system used for starting services on demand. pm-tools are also included for power management on desktop, server and laptop systems. Networking: Slackware continues to excel in this area with a wide variety of network applications ( including my favourites IPtraf, netwatch and nmap ) and easy to setup configuration. Also included are a number of Wifi tools, drivers and firmware, as well as Openvpn, ProFTPD/VSFTPD, RP PPP, the netkit package and Net-SNMP. Certainly it’s not something that other distribution users will be used to, but editing the network configuration file is easy to do for both static and DHCP setups. DNS and DHCP is covered by dnsmasq, although I tend to use the supplied ISC DHCP server and pdnsd, a 3rd party caching dns server with some nifty advanced features. Other 3rd party applications such as snort, nessus, nagios and wireshark are available and easily installed using standard compile methods. Server applications such as MySQL, Apache and PHP ( LAMP stack ) are all included in their latest incarnations, Perl is at 5.10 and Tk/TCL at 8.5 is also still around. Software/Packaging: The bugbear of many an article author, Slackware has what I believe to be the best package management system available. Packages are available from a number of popular sources on the Internet ( slacky.it and linuxpackages.net to name two ) and wrappers such as slapt-get and swaret provide for automatic download/installation and even some dependency tracking. Keeping a Slackware install updated is easy in the extreme. You will find though that from time to time, you may need to compile an application that has no Slackware package available. This is easily done with the provided development tools and there are some additional tools available such as checkinstall, slacktrack and src2pkg that allow even novices to compile and install applications. Development: Slackware remains one of the best development platforms available for Linux with everything covered from c, c++, java, fortran to debuggers, compilers and assemblers. The supplied KDE GUI includes the Kdevelop tools for developing KDE applications. GUIs: Slackware has long forgone maintaining 2 primary GUIs in KDE and GNOME, and Patrick has selected KDE as his choice of GUI. Given KDE’s more advanced configurability, this suits the nature and ethos of Slackware to the tee. As an aside, I’ve compiled a number of KDE 3.5.x releases and can attest to the fact that it’s far easier than what Gnome has provided. There are no special alterations to the stock KDE installation but I find that a good thing as it means I can configure KDE the way I like it without interruption. XFCE is also included and as a lightweight solution for older systems, it’s unbeatable. No binary video drivers are included but the supplied Xorg server has a good selection of built-in support for the majority of video cards available, and both the 3rd party NVidia and ATI drivers install cleanly and without fuss. Applications: Slackware ships with a good range of popular applications including encryption tools, editors ( including my favourite Joe ), browsers ( Firefox and Seamonkey ), email clients ( Thunderbird and KMail ) and office apps ( KOffice ); but crucially missing from that list is OpenOffice – not sure but perhaps something to do with the license? There are internationalisation files for KDE and aspell which provides for an easy to use system for alternative language users. Full sources are included for everything so you’re welcome to tinker around if you’d like. The supplied libraries are featureful and cover everything from AV, fonts and hardware to databases, UI interfaces and programming. Application issues: Recent changes in the Xorg server have broken VMware 1.0.x compatibility – there is apparently a way to get around this by installing older X libraries but I’ve yet to get this to work. Note this is not a Slackware-specific issue. VMware 2′s switch to a poorly conceived web gui for admin means that I’ve not bothered with it, but a search on Google will bring up some details on how to get VMware 2 installed on Slackware. This has led me to VirtualBox which in many respects, is better than VMware: it installs cleanly on Slackware, can use vmdk files from VMware and is much quicker to boot. The final nail in the VMware coffin is VB’s RDP-based headless mode which beats VMware’s non-gui management by a mile. VMware’s networking is still the best available but VirtualBox’s setup is covered in detail in the installation manual and simply requires some configuration of the system with the Slackware-supplied bridge utilities. VB’s wrapper scripts make this a straightforward task. KVM tools are not included but are easily installed although getting the RedHat-supported libvirt packages and it’s GUI components installed, remains a challenge. Openoffice has been an easy install in the past by downloading it, converting it to Slackware package format using rpm2tgz and running the install. But I had a strange issue in OO 3 ( including the betas ) where backspace and cursor keys refused to work in Writer ( only ). A simple workaround was to use Robby Workman’s OO3 packages and I’ve had no further issues relating to OO. Slackware still ships without a number of AV packages, specifically codecs; but this is for obvious legal reasons. The free AV codecs such as Flac and OGG Vorbis are included as well as a number of applications; the rest ( eg. MP3, Windows codecs ) can be obtained from 3rd party sources as Slackware packages. The MPlayer project continues to host all necessary Windows codecs. Gnome: This probably requires a section on it’s own as not only was the removal of Gnome from Slackware a slightly controversial decision, but it left a hole for certain applications that has been filled by a number of 3rd party projects including GWare, GnomeSlackBuild ( GSB ) and Dropline. GWare seems to lag in development a bit but the others are as current as you can get. GSB specifically, also includes a good number of AV applications including mainstream codecs, media rippers and converters. Support, Security and Updates: There are many avenues for support in the Slackware ecosystem. Two alternate methods are irc through the ##slackware channel and the vibrant web-based support page at Linuxquestions.org. There is a thriving Slackware community on the Internet so don’t be scared to search or ask. The packages in Slackware have always been of such high quality that the amount of updates for a particular release are a fraction of what you might see in other distributions. Patrick Volkerding’s attention to detail here and minimal patching means that Slackware packages are as close to the original as possible. There’s less to go wrong as a result of the reworking of packages, and packages and package versions are strictly chosen for their security history hence the fewer updates required. Updates are also provided for versions of Slackware going back many years which bodes well for those not inclined to upgrade ( especially servers ) – I recently saw an update for 8.0 ( isn’t that from the 90′s? ). Conclusion: Covering a Linux distribution is hard work as there are so many areas to look at but hopefully I’ve given you a taste of what Slackware can do. The strength of Slackware lies in it’s reliable and stable base, it’s configurable method of setup, and it’s adherence to the principle of KISS ( keep it simple, stupid ). Patrick’s ongoing commitment to the distro in the face of personal difficulties shows the quality of people behind this oft-overlooked distro. Its continued use as a starting point for many other distro’s/LiveCDs and the fact that it’s the oldest usable Linux on the market shows its inherent importance in the grander scheme of things. If you are a medium to advanced user, you won’t be disappointed. If you’re a novice and are looking to do more than just use Linux, Slackware offers the best base to start from. Give it a try – you may be surprised. The Debian SSH key fiasco from earlier this year is starting to bear bad fruit. The original issue ( listed earlier in this blog ) is that the Debian project took out some code from the SSH source as part of a code cleansing exercise – this code unfortunately was responsible for inserting randomness into the generated keys. Now there are only 65k keys being generated and that is a fairly small amount to break. Someone could essentially spend a day generating all these keys and then trying them against random servers. In fact someone has, as CERT has indicated activity relating to these compromised keys. And this is being used in conjunction with a local kernel exploit to gain root access. And that is being used in conjunction with dropping a root kit onto the box ( phalanx2 in this case ). Not one to choose sides in the Linux distro war, I’ve not been a fan of Debian and related distros ( including Ubuntu ). A number of issues have occurred over the years which has reduced my confidence in these distros and I think others should be listening when I say: stay away unless you’re sure. Open Source and its community-based development model is starting to give some in the commercial world a bit of a headache. You’ll remember that I’ve blogged about the fact that I think all software development will go the OS-type way in future and it seems this is happening already. We’ve already seen behemoths like Microsoft change their documentation and community practices based on this ( Microsoft now has quite a few projects on software forges like Sourceforge ). Note that I’m not indicating that products will all be free in terms of cost some day. There’s a big difference in the development model of software and whether it has a cost attached or not. Ultimately though, the OS nature of things will permeate the way software is sold. Red hat, for example, is doing very good commercial business ( against industry heavyweights like BEA, Orcale and others ) as a result of its purchase of JBoss, all of which is based on an Open Source dev model. Yes, the basic software is ‘free’, and by the way, if you need professional support, we’ve got this contract that will put you in touch with support, systems design and implementation. Around the clock. Anywhere in the world. Many have complained specifically about how OS spans the divide between free and commercial software. I believe the model works by providing ‘free’ ( as in cost ) products to all with professional services waiting in the wings if you need it. Both OpenSuse and Fedora are reasonable examples of this methodology. They may not be perfect but for most of us, they provide an OS, Application and development platform, while providing enterprise level software in the form of products like RHEL, SLES and JBoss. Everyone wins. Hans Reiser has been found guilty on the charge of 1st degree murder ( of his wife ) yesterday. From the little information I’ve read in the press, I understand that there’s mostly circumstantial evidence in this case, and the body of Hans’ wife hasn’t even been found. Based on this, I’m quite surprised with this verdict. In any case, I don’t think ReiserFS 4 was very popular so the rest of us will jsut keep on using the venerable ext3 or some of the alternatives like JFS and XFS. |
|||||||||||||||
|
Copyright © 2010 Robby Pedrica’s Tech Blog - All Rights Reserved |
|||||||||||||||