mightymouse dot net

November 2008 October 2008 September 2008 August 2008 July 2008 June 2008 May 2008 April 2008 March 2008 February 2008 January 2008 December 2007 November 2007 October 2007 September 2007 August 2007 July 2007 June 2007 May 2007 April 2007 March 2007 February 2007 January 2007 December 2006 November 2006 October 2006 September 2006 August 2006 July 2006 June 2006 May 2006 April 2006 March 2006 February 2006 January 2006 December 2005 November 2005 October 2005 September 2005 August 2005 July 2005 June 2005 May 2005 April 2005 March 2005 February 2005 January 2005 December 2004 November 2004 October 2004 September 2004 August 2004 July 2004 June 2004 May 2004 April 2004 March 2004 February 2004 January 2004 December 2003 November 2003 October 2003 September 2003 August 2003 July 2003 June 2003 May 2003 April 2003 March 2003 February 2003 January 2003 December 2002 November 2002 October 2002 September 2002 August 2002 July 2002 June 2002 May 2002 April 2002 March 2002 February 2002 January 2002 December 2001 November 2001 October 2001 September 2001 August 2001 July 2001 June 2001 May 2001 April 2001 March 2001

APRIL 2004

---

18th April 2004

---

Dammit, why did it have to rain last night. I was all set for an epic bike ride from Harrow to Central London to chill out in a park with an Icecream. The sky is looking pretty menacing and its quite windy too. Ah well. So I've stayed in, listening to The Cardigans and retro-fitting these log entries to fill up the past month. The log has been on quite a backburner for a while, and there is a bunch of other things I want to do as well. I have set a target for myself to finish my Xbox hacking howto before XBMC v1.0 comes out (3 weeks or so). There is a lot to cover as I want it to be explained in a easy way for anyone to understand how to do it. After that I've got to get around to upgrading my router machines, reconfiguring my local DNS in a more sensible way and a whole load of other technology bits and pieces. Also have to get this PHP Statistics thing fixed up some more, there is more work to be done on it (an exclusion list to ignore robots, more dynamic OS and Browser detection...) then hopefully it will be in a state to release (I think I've been saying that for almost a year - yes, I know). Lots of things to do... maybe I'll update my projects page and keep a list in there? Gotta think about it, tired now time for bed.
Track of the day: Lux - Northern Lights

---

16th April 2004

---

The sun is shining, the birds are singing, I'm wearing a T-shirt, there are cute girls driving about in their Mazda MX5s with the roof down, the Tottenham 'posse' are out in force in their ricemobiles testing out their stereos... Summer is here.
Track of the day: Jazzy Jeff & The Fresh Prince - Summertime

---

11th April 2004

---

Got up fairly early and thought about what I'm going to do today. Talked to Raveen for a while before Sid got up and then the plan was to head out in search of Breakfast. I borrowed Raveens bike and we both left on our mission. Cycled around, but couldn't get hold of jok on the phone so I guess we'd meet him for late lunch maybe. Had a nice baguette and tea at the Old Trip outside. The sun was shining and it was a lovely day. Had talked to Jo and he said he would meet us later, so we had some time to kill. What to do? The best we could come up with is ride about, along the canal is nice, go up to campus and take silly videos of us on our bikes with Sids camera. We bombed it up the hill coming up to the union shop on main campus, and then down past the new extension on the Trent building, across the road, up and over the grass and down to the gallery. Sid took a video of this ride with his camera round his neck... it was quite mental (warning: 13MB). Here is another video of Sid testing out his suspension while almost falling off into a water feature... check out the backwards cap, who da man! (2.5MB) Its all good fun :D

Anyway, I had to leave to get a train back before it got too late, so it didn't look like I'd get to spend any time with Jo. Ah well, theres always next time. As we changed trains on the way back we had to cross over to another platform. I spotted a lady with a suitcase and a wheeled bag so I gave her a hand else she'd have to make 2 trips or something. Why do women always have so much stuff?
Track of the day: Lamb - Angelica

---

10th April 2004

---

Its Saturday, and after quite a long day at work I figure I've got to make something of this Bank Holiday weekend. Had dinner, checked train times, chucked my pyjamas, some clothes and a toothbrush into my rucksack and left for Nottingham. Hopefully I can spend some time with Sid and Jo, its going to be fun :)

Made it into Kings Cross just on time and was served by a nice old gentleman at the counter. St Pancras is closed this weekend as it is moving to a new temporary platform to make way for the new Channel Tunnel rail link, which means that Midland Mainline trains are only coming as south as Luton. Right, so the guy who sold me my ticket said I need to go out of the station, turn left and go 200 yards down the road to get to Kings Cross Thameslink station on the right, and change at Luton to go up to Nottingham. Hmm, well, I don't actually know how long a yard is, and he said I should hurry because the last train is about to leave. So off I went, assuming I'd just keep going until I saw the station, which worked ok. I got in just in time to leg it to the platform and jump on the train before it left. It was a relaxing ride up and it was on time to get me into Nottingham at about 00:30.
Track of the day: Stereophonics - Traffic

---

6th April 2004

---

OK, time for some logical thinking. debugreiserfs -D /dev/hde shows that the files on that disk are actually still there somewhere in the quagmire of the broken filesystem. resiserfsck had problems reconstructing /dev/md0 which had /dev/hde added to the array. Now, /dev/hde was the drive that was kicked out of the array for being bad, so I'll see whether to get some sense out of /dev/hdg. I checked /dev/hdg with fsck and as it was a mirror of /dev/hde it unsuprisingly wanted reiserfsck to --rebuild-tree just as /dev/md0 had wanted to earlier. This time it actually worked. The nodes were reconstructed and I could successfully mount the disk on /data/raid and all my stuff was back, and it wasn't horribly corrupted (like I thought it might be). Oh my goodness. Copied everything back to my 160GB root disk so I can make a good few backups! Wow that was a horrible couple of days, I really was not looking forward to FLACing all my CDs again and losing all the music that I had on MP3/Ogg and all my latest photos that I haven't backed up yet. Well, I have a feeling that hde is dead, although SMART says its ok. I will run the Seagate Seatools diagnostic bootable CD thing at some point. It can't just die for no reason. I blame my Promise Ultra 133 TX2 IDE Controller for killing it, and Jon says that he has had bad experiences with Promise controllers in the past as well. Although things got broken when I was putting my stuff on RAID, it was the fact that it was RAID which meant that I got my data back. If my stuff was only on hde when it died with no RAID I think at best some of it might be horribly corrupted or at worst just completely lost forever. And even though it was some strange hardware problem, the RAID came to the rescue and it was me who made things go pear-shaped by reconstructing the array without a thorough backup when something was obviously wrong.

As they say, avoiding bad decisions comes from experience, and experience comes from making bad decisions.
Track of the day: System of a Down - Highway Song

---

5th April 2004

---

Woke up early, got myself ready for work and looked at /proc/mdstat to see how my array was doing. Everything was sync'd up and it was healthy. Great. Looked in /data/raid and there was nothing there. Shit. How could this be? I rebooted the server, and it couldn't mount /data/raid. 'Not a directory' it says. fdisk /dev/md0 says that the reiserfs filesystem is all bustimacated and may be fixed with --rebuild-tree. OK then. How did this happen? Hmm, the only possible way would be if /dev/hde became bad and this synced over to /dev/hdg. Crap. ide-smart /dev/hde /dev/hdg said both my drives were healthy, so I crossed my fingers, ran reiserfsck --rebuild-tree /dev/md0 inside a screen and went to have my breakfast.

Got to work, SSH'd into the fileserver, reattached the screen and it said that it failed to rebuild the tree and it looks to have been a hardware error. Bollocks. I stopped the raid and now have to think about how the hell I'm going to get all my photos, important documents and 40GB of music back from the dead.
Track of the day: The total silence of a busted fileserver

---

4th April 2004

---

Bought lots of new hardware for my dads shop over the past week (10K SATA disks in a phat fileserver, a couple of Shuttle machines, Gigabit LAN, etc). Along with that stuff I ordered a 160GB Maxtor Diamondmax Plus 9 to go in my fileserver as it was a really good deal and I could do with some more space. Anyway, this 160GB is going to be my new root disk for a new Gentoo install and my other two 120GB Seagates on a Promise Ultra 133 TX2 are going to be a volume in RAID1 as /data/raid. Well, I got all of that installed and configured today, the stage 3 Gentoo install didn't take any time at all, making the RAID device was totally painless and before I knew it everything was up and working great.

I was burning some CDs for my cousin and for some strange reason the machine just stopped responding after doing 4 in a row. Very odd, I have never seen this machine crash before. Hmm, well without thinking anything of it, I powered it off, and it booted right back up again. After burning the rest of the CDs I checked on my raid array. In /data/raid all my music, photos and documents were still there, but /proc/mdstat was saying that only one of my disks was part of the array! /dev/hdg was part of my array and /dev/hde was marked as bad. This is very strange. Its quite late at night and this should really be the time where I think in my mind that "Maybe I should backup my data while it is still there...". For some reason, I don't think the fact that /dev/hde has a Bad Superblock in the dmesg output is a very bad thing. I typed 'raidhotadd /dev/hde /dev/md0' as root and watched the array sync itself up. I figured since /dev/hdg was active it would sync from /dev/hdg to /dev/hde. That would make sense right? Anyway I didn't think anything of it and went to sleep.
Track of the day: Fatboy Slim - Demons