I know I might bag out CentOS at the best of times, but Karanbir Singh – our fearless leader of all CentOS developers has announced the immediate availability of CentOS-5.7 for i386 and x86_64 Architectures.
Well, what are you waiting for? yum upgrade.
Yesterday I posted about having an existing ssh-agent load on all new shells. Here are two more handy snippets of code from my .bash_profile
The first code snippet is a follow on from yesterday, were I can type ‘lock’ or ‘unlock’ into my shell and the ssh-agent will follow on accordingly. When your ssh-agent is locked, users that have access to the ssh-agent will be required to type in your SSH agent password.
function lock () { ssh-add -x }
function unlock () { ssh-add -X }
The next snippet of code adds a ‘ns’ command. I have issues trying to remember IPs – especially when they are not used too often. This command lets me easily remember 
alias ns='for x in ns1 ns2 ns3 ns4 ; do host $x.google.com; done'
When run:
[tim@2-xlc-controller ~]$ ns ns1.google.com has address 216.239.32.10 ns2.google.com has address 216.239.34.10 ns3.google.com has address 216.239.36.10 ns4.google.com has address 216.239.38.10
The last dirty one liner that I love is another simple time saver. Many people use the ‘whois‘ command to find out what nameservers are used by a domain name. It’s not too long before you work out that it is not really the best way to be finding out what the domain name’s nameservers are.
function nameservers() { echo $1\'s nameservers are:; dig +trace $1 | grep NS | grep "^$1."; }
This handy one liner allows me to do awesome things like:
[tim@3-xlc-controller ~]$ nameservers google.com google.com's nameservers are: google.com. 172800 IN NS ns2.google.com. google.com. 172800 IN NS ns1.google.com. google.com. 172800 IN NS ns3.google.com. google.com. 172800 IN NS ns4.google.com.
I love RedHat Linux. I think it is a beautiful piece of engineering. Hundreds of thousands of man hours spent ensuring systems that have packages installed keep running. Continuously.
One thing however that I have found is that my favorite RHEL clone for years, CentOS has been slacking. Bad.
For example, RedHat Enterprise Linux 6 was released 10/11/2010, not bad, seeing as it’s previous version, RHEL5 was released 15/3/2007 – Three and a half years earlier. CentOS? Well, Version 6 of CentOS was released July 10. In the time that it took for CentOS to release their new fresh build of CentOS 6, RedHat had already released their first service pack/major update to RedHat, RedHat Enterprise Linux 6.1.
I must point out here, the fact that CentOS does not have 6.1 is not a security issue. Any security patches for RHEL have made their way to CentOS.
Scientific Linux on the other hand have been tracking the updates and new releases very well. They had their remix of RHEL released for public consumption on March 3 – not too much of a short wait. The fact that it is not a very faithful clone has scared me from moving to it, but with the long wait of CentOS 6.1, I am starting to think that it might just be time to change my mind and grow up!
Wondering where all the interesting Sydney IT meetups are? I have listed here a rough calendar for 2011. There may be some updates further on in the year as more events are announced.
Big Meetups
Weekly / Fortnightly Meetups
Monthly Meetups
Sparatic
Please let me know (in the comments section “below”) if there is anything that I have missed out on!
A typical day’s worth of updating my machines usually looks a bit like this:
[tim@myhost tmp]$ sudo pacman -Syu :: Synchronizing package databases... core is up to date extra is up to date community is up to date multilib is up to date :: Starting full system upgrade... warning: gnome-disk-utility: ignoring package upgrade (2.30.1-1 => 2.32.0-2) warning: gvfs: ignoring package upgrade (1.6.3-1 => 1.6.6-1) warning: kernel26: ignoring package upgrade (2.6.34.3-1 => 2.6.36.3-1) warning: kernel26-headers: ignoring package upgrade (2.6.34.3-1 => 2.6.36.3-1) warning: libbonoboui: ignoring package upgrade (2.24.3-1 => 2.24.4-1) warning: libgnome: ignoring package upgrade (2.30.0-1 => 2.32.0-3) warning: nvidia: ignoring package upgrade (256.53-1 => 260.19.29-1) warning: nvidia-utils: ignoring package upgrade (256.53-1 => 260.19.29-1) warning: proftpd: local (1.3.3-5) is newer than extra (1.3.3c-1) warning: rdesktop: local (1.6.0_rc1-1) is newer than extra (1.6.0-5) warning: udev: ignoring package upgrade (151-3 => 165-1) warning: udisks: ignoring package upgrade (1.0.0-0.20100223 => 1.0.2-1) warning: youtube-dl: local (2010.12.09-2) is newer than community (2010.12.09-1) resolving dependencies... looking for inter-conflicts...
You know it’s bad when you have so many custom packages, that you may as well have your own distribution .
Quick and simple MAC address generator:
dd if=/dev/urandom bs=1 count=3 2>/dev/null | od -tx1 | head -1 \
| cut -d' ' -f2- | awk '{ print "00:50:19:"$1":"$2":"$3 }'
I normally have this set up as an alias in my profile under ‘mac’.
When I faced with a decision of what web server I am going to roll out with, I usually end up picking between two web servers: Apache and Cherokee.
Apache is like your grandfather. He is 80 years old, and He has been around for a while. He is not always the guy you go to when you want something done quickly, but when you want it done reliably, you would not even think of going to anyone else.
Cherokee is like a kid out of Uni. Sure, he is still only a little baby, but he is packed full of the latest knowledge and has been taught how to do the job. Quick. He is also much better then his Granddaddy at doing the easier things, such as giving you the same document over and over and over again.
The way that Apache is written, it does use up a lot of memory, however, there is also the benefit with Apache that if you want to do it, you can. There is literally any type of module you would ever want for Apache. There are more then 400 modules to download, compile, install and try.
Cherokee on the other hand has a much smaller selection of modules you can choose to run with, but don’t let this scare you! If you run pretty much stock standard Apache setups (such as I do for timg.ws and sharesource.org), then Cherokee will be able to come to the table with everything you need and more.
The really cool benefit that you get from using Cherokee is the fact that, it does more out of the box with all its modules.
At the moment, I really like the development that is going into Cherokee. They do look very hard at security, such as this the new spawning mechanism introduced into Cherokee earlier this year.
But really, there is a lot of hype over the whole “lightweight” httpds in some sense. Sure, lighttpd and Cherokee are really fast to deliver static files, but are they really faster then for anything else? In all honestly, not that much faster.
I hear about people doing a lot of really interesting things when it comes to web servers, like Apache+nginx+fastcgi, and then I wonder, couldn’t you just pick one product and stick with it? It’s not like the extra milliseconds are going to save you $200,000 a year.
I usually choose Cherokee now for new installations, simply because it does everything and I don’t need to actually do that much hunting around to make it ‘just work’. Not only that, but it has a nice web interface for administrating it that would even make my Dad happy (yes, literally).
I was just booting up my Mac to take some nice pretty screenshots of my cluster install process, and I realised that NoMachine’s client for OS X does not even properly support Intel CPU’s -.-”
Anyways, unlike Adobe porting Creative Suite to Linux, at least NoMachine are actually working on it.
What does a computer nerd do when he is at work and trying to use one keyboard and mouse for two machines and it just does not work? Well, he fixes it of course!
For the last few weeks, since switching my KDE desktop at work to Gnome, I have had this seriously pressing issue that every now and then when I moved my mouse to the Linux machine, synergy would crash on that screen, with a nice assertion error.
INFO: CScreen.cpp,99: entering screen synergyc: ../../src/xcb_io.c:243: process_responses: Assertion `(((long) (dpy->last_request_read) - (long) (dpy->request)) <= 0)' failed.
Well,I looked through the code for a few hours and I finally tracked down the issue. The synergyc application was actually trying to do calls to the X server using multiple threads, when X did not know that it was being used from a multi-threaded application.
Basically, after hunting down where the issue was, fixing the problem was as simple as adding a call to the XInitThreads() function on line 100 of lib/platform/CXWindowsScreen.cpp.
Rebuilt my package and boom, a perfect synergy.
Well, I finally got all (but one…) of the parts of my new computer. My new case, the Cooler Master COSMOS “S” is in one word: absolutely awesome. I love the fact that all of the drives can be pushed in, and with one press of a button, the drive will lock in.
This comes with the terrible actuality that my hot swap device, the Chenbro five-in-three hot-swap device fails to just slide and lock in, because of the grooves in the case that allow the normal 5.25″ drives (like a CD drive) to just plug in with out any screws. So, it looks like I am going to need to get a file, and file away these damn things.
So yes, the one thing that is stopping me from putting this computer together, is (one again, mind you) the fact that I want so much storage. If I had only known that putting so much storage in a decent case would be so much hassle.
Other then that, everything else plugs in perfectly fine, which is pretty cool. I still really don’t know why there is a SATA connector on the top of the case, especially seeing as I have nothing that will plug into it, yet…
Can’t wait to file the things that are stopping me from putting the hot swap drive bay in.