Archive for the 'Linux' Category

Going Postal…

I swear one of these days I’m going to go postal on whoever had the bright idea of inventing the DMCA. It does absolutely nothing to protect copyright, but does plenty to kill the honest work of open source projects. Like CoreAVC for Linux, for example. This project did not borrow or steal any code - you still required a copy of CoreAVC to run it. Yet we must constantly be in a think-of-the-children mindset when it comes to copyright, and it’s really starting to piss me off something chronic.

If copyright holders are so worried about their rights that they think tying the hands of honest customers is the way to go, then in my opinion, they don’t deserve copyright.

Update: It seems CoreCodec have realised the error of their ways, and I applaud them for vowing to work more closely with the open source community. That said, this highlights the fundamental flaws with the DMCA - it’s a shoot first, ask questions later policy which is so open to abuse it isn’t funny.

First impressions of Ubuntu Server

I'll be in the Angry Dome!So I got around to backing up the boot drive in my server and proceeded to format it so I could play with Ubuntu Server 7.10. I’d been thoroughly impressed by the experience I had under the desktop version, so I expected much the same from the server version too. In the end, it didn’t go quite as smooth as I expected, but to be fair none of the problems were really Ubuntu’s fault. Following is a bit of a long-winded description of what not to do if you plan on running a system with either software RAID, or bridged network cards.
Continue reading ‘First impressions of Ubuntu Server’

Farewell to Gentoo Linux

I’ve had a dedicated fileserver for a while now. It ran Windows Home Server for a brief stint a few months ago, but as I’ve already posted, it was a load of rubbish. For the majority of the time, it’s been running Gentoo Linux - purely because I liked to keep the system as streamlined as possible, installing only what I used. It’s also taught me a lot about the inner workings of a modern Linux system.

Unfortunately, the time has come to say goodbye to Gentoo. It does what I need for now, but troubles are starting to arise with the future of the Gentoo Foundation that suggest switching to another distro would be better sooner rather than later. The foundation’s charter has been revoked for a while now, and even a simple visit to their website shows that it’s not being updated anymore. It’s an unfortunate situation. But anyone can see that the community is beginning to fall apart and the management just don’t have the same ideals that they used to.

So I’m making the obvious choice, and moving to Ubuntu. It’s much easier to run and maintain, support is extremely easy to come by, and the leadership is much more solid. Gentoo’s future remains uncertain, but Ubuntu’s isn’t.

Linux Hacks: Copy a list of files with cpio

So here’s the problem, I listen to all my music streaming from my linux server with MPD and Pitchfork (which is quite an awesome little AJAX interface for MPD). I’ve got a playlist made with Pitchfork that I want to copy to my iPod. These playlists are saved in /mnt/store01/Playlists, while music is stored in /mnt/store01/Music.

How do I copy the files in the playlist to a new directory so I can just copy that whole directory to my iPod?

After a bit of screwing around with trying to feed output from cat into cp with a pipe, and lots of fruitless Googling, I finally found a way with a neat little Linux tool called ‘cpio’:

safari ~ # cd /mnt/store01/Music
safari Music # cpio -Vdp ../musictemp < ../Playlists/1.m3u
............................................................
............................................................
............................................................
............................................................
................
3272929 blocks

And it’s that simple.

Windows Vista: Fucking with Windows Home Server?

Well, I wanted to see what Windows Home Server was like, and also took the opportunity to test whether or not my Vista desktop got faster network speeds reading from a WHS server as opposed to a Linux Samba server. Now as I said in my last post, read speeds were about 5-6Mb/s when copying from my Samba server to a Vista client. Surely I thought Microsoft’s own server OS would absolutely trample Samba in the mud.

Anyone want to take a guess at the speeds I got between my WHS test box and Vista? Same configuration, gigabit network all round, etc…. Continue reading ‘Windows Vista: Fucking with Windows Home Server?’

Windows Vista: Fucking with Samba?

My fileserver of choice for a while now has been a Gentoo Linux box as I’ve posted previously. It’s been working like a charm with Windows XP, I can’t complain at all about the transfer speeds.

However, Microsoft have changed pretty much everything in Windows Vista, and for no apparent reason than for the hell of it. One of the changes is SMB2, a new version of the Windows File Sharing protocol which, although not adding any improvements past changing the integer count from 16 to 32-bits, is apparently a huge improvement. If you believe Microsoft PR. Continue reading ‘Windows Vista: Fucking with Samba?’

Linux as a fileserver….

…is goddamn awesome.

Here’s why. I’ve been running a system the last few months which has consisted of 5×80gb drives I had lying around (one for the OS, the other four in a software RAID 5 setup) plus a few other single drives. It was a P4 3.0GHz Prescott running on an Intel 865 chipset, which, although acceptable, struggled a little under full load. Especially considering the board doesn’t support PCI-E and only has 2 SATA ports and 100mbit ethernet onboard.

So I ponied up this week for an Athlon 64 3000+ processor from eBay this week. $90 later, and I’ve thrown it in a spare Gigabyte K8NF9 board, which had double the SATA ports and onboard gigabit ethernet. This turned out to be just a little zippier than the P4 setup. Since I run Gentoo Linux, everything was set up to compile for a P3 architecture (P4 is bugged, so I didn’t use it). It’d probably work on an Athlon 64, but I’d rather have everything done properly, which either meant recompiling the whole system for a K8 processor, or reinstalling Linux. I decided on the latter, and threw the 64-bit version on anyway.

Now this is the first software RAID setup I’ve used before, and it was working well up until this point. I wasn’t sure how it’d handle a reinstall of Linux though, I was expecting the worst (having to let mdadm rebuild it). I was pleasantly surprised though, after sorting out my bugged GRUB config, to find that after compiling the RAID drivers into the kernel, it automatically detected the array and all the drives in it. It definitely wasn’t that easy with my last hardware RAID controller (admittedly, it was a Promise card).

Then came adding another 80gb drive to the array. A bit of Googling found this excellent howto on adding a spare drive to a software RAID 5 setup, extending it to the spare drive, and extending the filesystem to fit. About 2 hours later, it was all done. Couldn’t be easier.

And although I haven’t got a gigabit network card on the other end to test the speeds, it all seems pretty damn quick. Should give me enough storage to last till I can afford a trio of 500gb drives, which seem to be rapidly dropping in price.

Here’s how it looks:
Filesystem Type Size Used Avail Use% Mounted on
/dev/sda2 ext3 72G 2.3G 66G 4% /
/dev/sda1 ext2 16M 3.4M 12M 24% /boot
/dev/md/1 ext3 294G 182G 101G 65% /mnt/store01
/dev/sdc1 ext3 230G 158G 60G 73% /mnt/store02
/dev/sdd1 ext3 147G 100G 41G 72% /mnt/store03

And here’s the tower of doom!
Doom!

(Trusty old AthlonXP 1700+ IPCop firewall on the bottom, the fileserver on top)