It seems right to post this here, where it all started. Farewell, ~jawtheshark.

It seems right to post this here, where it all started. Farewell, ~talinom.

Very clever. I approve.

I appear to be in the market for a new KVM as my current one has all but stopped working. Any recommendations? Minimum 4 outputs. *Must* do VGA. My current one has mini-DIN for keyboard and mouse, which has sort of worked some of the time with USB-PS2 converters. I could probably go for a full USB only one, but I'd need to get new cables.

leto:/stuff% mkdir mp3
mkdir: cannot create directory `mp3': No space left on device
leto:/stuff% mkdir qwe
leto:/stuff% mv qwe mp3
leto:/stuff% ls -ld mp3
drwxrwxr-x. 2 tet tet 4096 Dec 22 20:53 mp3
leto:/stuff% df -h .
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/leto-stuff 30G 9.9G 19G 35% /stuff
leto:/stuff% df -hi .
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/mapper/leto-stuff 1.9M 322 1.9M 1% /stuff

This appears to be 100% repeatable behaviour.

Crazy war stories, indeed.

http://mina.naguib.ca/blog/2012/10/22/the-little-ssh-that-sometimes-couldnt.html

My decision to rip all of my audio losslessly to FLAC has been vindicated. I rip to Vorbis as well, but I always thought it was worth having the lossless originals around too. That way, if a new codec arrived on the scene at a later date, I could rerip to that without any further loss of quality. That codec is now here. Hello, Opus!

I'd forgotten how much I hated Javascript. I very much appear better suited to server programming than to web programming...

Some people think I'm paranoid. I wouldn't say so. It's just that I pay more attention to the potential worst case outcome that some. So when it comes to storage, I have a mirrored RAID array in my home server. The contents are backed up to a separate disk in the same machine. I also have an offsite backup in a datacentre.

My offsite backup machine died, and is now sat at home waiting for me to rebuild it. So it was somewhat alarming when my backup drive also died. Uncomfortable about running with less redundancy than normal, I immediately went out and bought a replacement drive. When checking the drives in the machine to see which one I needed to pull out, I noticed that one of the mirrored drives had also failed and the array was running in degraded state. Eeeek! Of my four levels of redundancy, three had failed. If I'd had fewer, I'd be screwed right now!

I'm not sure why I wasn't notified about the RAID failure. Normally I automatically get an email when the array enters a degraded state. That's something I need to look into. For now, the array is rebuilding. I'll fit the new backup drive when I get home this evening.

Yes. I've been saying this for a long time. Email is pretty much the definition of a perfect store and forward messaging system. Plus, I completely agree with him about the omission from the spec.

Time for more storage. Where's the current sweet spot? It looks like drives are considerably more expensive than I was expecting. I'm guessing that's still a hangover from the flooding in the Far East. I'm considering a Samsung HD204UI or a Seagate ST2000DL003. Both are 2TB, which is about right for what I need. Both are slow (5400 and 5900 rpm respectively). That's OK. I/O performance is not the bottleneck here. That said, I'd rather have a bit quicker, but going up to 7200 rpm adds significant extra expenditure.

I'm way out of touch with PC hardware standards. The Seagate is nominally SATA III (a misnomer, IIRC). Will that be backwardly compatible with my SATA I/II controllers, or will I need to upgrade those as well? Any other suggestions on models I should be looking at, or things I should be considering?

Dear Mozilla developers. I know you're a bunch of incompetent morons, but would it really be so hard to change that and release a decent product? Please?

Ever increasing storage capacity is making me lazy. In years gone by, when resources were more limited, a full filesystem would be cause for investigation, to find out why it had become full, and what could be done about it. These days, the easy option is to just extend the volume a bit more and grow the filesystem. I'm running a bit low on free PV space, so last week I didn't grow my home filesystem by as much as I normally would when it filled up. But even so, I was a little surprised to find out that within a couple of days it was already full again. That's not normal. It turned out that a single log file was continually growing and had taken up nearly a third of the total space on that filesystem. But because the default assumption is that I have more disk space than I'll be able to use, I hadn't been keeping an eye on what was taking up the space. D'oh! So a single rm has taken me back to having loads of free space.

Yes, yes and more yes.