corrupted inodes (again) and some SSD calibration
Well, Mac OS corrupted the Linux drive again. I'd uninstalled the program to read ext2fs, so that means either darwin/Aqua does the corruption, or shit installed with .dmg is hard to totally remove. Neither would surprise me.
Even though I didn't do an image backup recently, I also didn't do much computer work over my break, and recent work has all been code in my home directory (file structures are all backed-up by rsync regularly, but I'm not especially keen on doing an entire system restore from it, particularly when I need to wipe the drive anyway to hit all the corrupt inodes). But I'm getting pretty decent, so it was hardly any of my time lost (maybe an hour or two) and a bit of data transfer and re-compiling code.
Since the last image backup was after another crash, it had basically everything critical, but there were some funny points like my present kernel is installed and the configuration files are there, but the kernel source code is missing and the modules aren't installed. Otherwise it was basically just copying out my home directory, restoring from disk image, bringing back my home directory from yesterday, and installing some shit through portage (and some backups of make.conf, /etc/portage and the binary forms of system shit I'd installed so I don't have to recompile it.
So, it's amusing that Mac OS corrupting inodes at this point is like, "Oh, that again...meh."
I also need to test a bunch of detectors for an upcoming experiment. So I came in on Sunday at like midnight and did basically a fresh setup. It took me about 4-5 hours to get my first signal, which was longer than I thought at first, but not bad at all, especially considering I've never done an entire setup from scratch myself before. Very few actual problems, and a lot of time was just spent collecting the different things I needed (which was everything!). Actually documented the specs on some detectors (all the ones in the data room of a certain type, being 6 of them), but I'll need to redo it because for the actual experiment we are using a different pre-amp (which I'd kind of forgot last night, but we usually use the pre-amps I'm using for these detectors, and in any case, it would give me one less point to troubleshoot since I know this combination works well).
Probably that will take most of my week to do that work, but I don't mind a bit of lab work (and makes me not notice if my computer is out of commission for half a day transferring data and shit).
Even though I didn't do an image backup recently, I also didn't do much computer work over my break, and recent work has all been code in my home directory (file structures are all backed-up by rsync regularly, but I'm not especially keen on doing an entire system restore from it, particularly when I need to wipe the drive anyway to hit all the corrupt inodes). But I'm getting pretty decent, so it was hardly any of my time lost (maybe an hour or two) and a bit of data transfer and re-compiling code.
Since the last image backup was after another crash, it had basically everything critical, but there were some funny points like my present kernel is installed and the configuration files are there, but the kernel source code is missing and the modules aren't installed. Otherwise it was basically just copying out my home directory, restoring from disk image, bringing back my home directory from yesterday, and installing some shit through portage (and some backups of make.conf, /etc/portage and the binary forms of system shit I'd installed so I don't have to recompile it.
So, it's amusing that Mac OS corrupting inodes at this point is like, "Oh, that again...meh."
I also need to test a bunch of detectors for an upcoming experiment. So I came in on Sunday at like midnight and did basically a fresh setup. It took me about 4-5 hours to get my first signal, which was longer than I thought at first, but not bad at all, especially considering I've never done an entire setup from scratch myself before. Very few actual problems, and a lot of time was just spent collecting the different things I needed (which was everything!). Actually documented the specs on some detectors (all the ones in the data room of a certain type, being 6 of them), but I'll need to redo it because for the actual experiment we are using a different pre-amp (which I'd kind of forgot last night, but we usually use the pre-amps I'm using for these detectors, and in any case, it would give me one less point to troubleshoot since I know this combination works well).
Probably that will take most of my week to do that work, but I don't mind a bit of lab work (and makes me not notice if my computer is out of commission for half a day transferring data and shit).