Saturday evening… ‘Geek Speak’

Back again, and thank you all for your kind words in the comments. I’m sorry that some didn’t appear immediately (those from John’s family) but the first post from a new source is always held for moderation as part of ‘SPAM’ prevention! But thank you – I think they have all been liberated now!.

And you may have noticed that the system has been down again today. This was a controlled and deliberate shutdown as part of an ongoing upgrade process to the server.

To recap, the blog and web sites are all hosted on a Linux machine (sitting at my feet) connected to an ADSL line. The machine was originally built about 2 1/2 years ago as a self teaching project for Linux – an operating system I had dabbled with in the past, but wanted to learn more about as it seemed to be a credible alternative to Windows.

I have learned a lot about Linux (and for everyday tasks, it is my machine of choice) by using it, and when I started hosting my own website (after taking over from friend Mark) it was the logical machine to run it on – after all, probably 80% of the internet runs on some form or *nix (generic unix like operating systems) and even more run the Apache web server, as do I.

However, running a website as a service carries some overhead, like maintaining resiliance and availability. The system is protected against power surges with a UPS, but the single point of failure is the hard drive, which is only relatively small at 80Gb, and is now about 65% full.

So I have been experimenting with RAID arrays (using two identical disks with one ‘mirroring’ the other so that if one fails, the other can continue). This has not been a great suceess, for reasons I am still trying to solve, but I think is an incompatability between the controller card and my system. So in the interim I have been trying to replace the original drive with a larger one, and use the old one for backups.

I am now at the stage where all the data has been copied across, but the new drive needs to be configured as the boot or main drive, and this is where the problem lies – I haven’t been able to get the new drive to ‘boot’ the system cleanly. (This was the problem on Monday night – I had got all the changes in place, bar one or two, but when the system was shutdown because of the prolonged power failure, it wouldn’t restart at all – and if it won’t restart, you can’t make changes… at least not without a rescue mode! ‘Chicken and egg’)

This afternoon I have been copying all the files across, and (I thought) getting the boot process sorted – but clearly not as the machine still wouldn’t boot from the new disk – so at the moment I am back in the old configuration. Just to compound matters, one of the small standby servers that puts up the “System down for maintenance” pages has expired, so I had to reconfigure another one. (They are little network file servers that have been hacked to turn them into low power Linux machines – normally they act as music servers to MP3 players round the house). So there may a be a few more outages next week when I have pondered the problem a bit more.

Enough of the ‘Geek Speak’! Kt came home today to see us, and to collect her new laptop that arrived yesterday. She has been using Linux on her desktop machine, but the new laptop is Windows based, so we had a mutual learning experience as we configured it for her. I was ‘home alone’ today, so Kt and I had some time together this evening over a stir-fry!

Health wise, I’m not sure if there aren’t some hints of GvHd – or perhaps that is wishful thinking, as it is only just a week since the infusion, and generally it takes at least 6 weeks for the incomers to multiply and get to work. However, maybe room for a cautious ‘onwards and upwards’