After finding and acquiring a small bridgeboard, one that's specifically designed to connect to Ipod hard drives, I was able to get the old hard drive to scan long enough to clear out its Pending Sector Reallocation flag. That's important to get that out of the way. Once that got cleared, I was then able to do a quick reformat of the drive and try to scan for missing files. Initially, that didn't go as well as I hoped.
-TestDisk works great on FAT32 and NTFS drives but falls short on HFS+. But by doing a full force-scan on the whole drive, and while it was effectively naked, that caused the Pending flag to go back down. Progress!
-DiscRescue 3 (OSX) found files but oops, I was running the demo version so I couldn't save anything. There went 22 hours on the PowerBook.
-DiscRescue X (OSX) also found files but it's an older version than DR3 so what it did find was block-shifted all to hell and I have no idea where to set the offsets. At this point I've had the PowerBook running nonstop for 2 solid days, which is Not Good.
-DiscRescue 5 (W7) is more intelligent, and figured out the offsets on its own. It also runs on a newer computer so I don't have to leave the PowerBook to roast itself.
Each scan takes the better part of a day non-stop regardless of the host computer, and much of that has to do with the hard drive itself being a 4200RPM IDE drive that's designed for efficiency instead of speed. Good thing I have multiple computers, so that I can leave one to crunch away at that sort of thing and still be able to do other things.
The reason why I'd like to avoid leaning in too hard on the PowerBook is because it's prone to overheating. The internal cooling system isn't what it used to be, and I learned the hard way that it can get hot enough to give me 2nd-degree burns if I have it sitting on my lap while I'm wearing shorts. If it's running too hot for too long, it will grind to a complete halt, because it doesn't quite have the sense to reduce its clock speed. I had it sitting on a metal sheet as something of a makeshift heat sink, and that got warm. Were I to try this in cooler weather, I'd be inclined to have it running in the garage. Secondly, the PB does not have native USB2. It does only by virtue of an add-in card, and that can barely support powering an ipod. So if the card decides that it can't keep up, then the recovery could fail.
After going through with DiscRescue 5, it did find everything and then some. I say "and then some" because it found about 3GB of stuff that I intentionally deleted a couple years ago. Oh well, it's just hard drive space, I have plenty of that on the server. (What's 150GB to 17TB?) The real prize was the itunesDB file, which is where the playlists reside.
Here's where it gets weird. The files I was looking for were not in the folder they should have been in. The Ipod_control folder, which has the OS of the thing, was present but empty. The itunesDB and itunesControl files, which are supposed to be in that folder, were in a much different folder, and one that's not normally present in a functional ipod. This may or may not be related to the sector that needed reallocated.
But! Would everything transfer to the new hard drive that's already formatted in FAT32 when the original was HFS+, or is the Ipod software persnickety enough to throw a fit? I tried something similar back in 2014, when I went to move stuff from this one to a significantly newer one, and I ended up locking up the device hard enough that I had to restore it. As it turns out, the Gen6 and Gen7 versions have software that's just different enough.
Figuring that I might as well try, I copied the newly-extracted files to the "new" hard drive while it was on the adapter. Once that was done, I reconnected it to the ipod, turned it on, and it came up. Everything came up. It was in the same state as it was back in June when I last used it.
What's next? I've gotten everything I wanted off of it, I have the new drive and battery installed, and I've largely given up on trying to increase the capacity. When I'm done with further (non-destructive) experiments, I'll put it all back together. Except I'll modify the back case in such a way that it doesn't have all those tabs holding it on. That's a common maintenance thing; where if you remove 10 fasteners to remove an access panel, you only put 8 or 9 back on when you're done. That's because if you had to remove that panel once then you'll have to remove it again later, or those fasteners somehow got lost or damaged. Or both. Anyway.
-TestDisk works great on FAT32 and NTFS drives but falls short on HFS+. But by doing a full force-scan on the whole drive, and while it was effectively naked, that caused the Pending flag to go back down. Progress!
-DiscRescue 3 (OSX) found files but oops, I was running the demo version so I couldn't save anything. There went 22 hours on the PowerBook.
-DiscRescue X (OSX) also found files but it's an older version than DR3 so what it did find was block-shifted all to hell and I have no idea where to set the offsets. At this point I've had the PowerBook running nonstop for 2 solid days, which is Not Good.
-DiscRescue 5 (W7) is more intelligent, and figured out the offsets on its own. It also runs on a newer computer so I don't have to leave the PowerBook to roast itself.
Each scan takes the better part of a day non-stop regardless of the host computer, and much of that has to do with the hard drive itself being a 4200RPM IDE drive that's designed for efficiency instead of speed. Good thing I have multiple computers, so that I can leave one to crunch away at that sort of thing and still be able to do other things.
The reason why I'd like to avoid leaning in too hard on the PowerBook is because it's prone to overheating. The internal cooling system isn't what it used to be, and I learned the hard way that it can get hot enough to give me 2nd-degree burns if I have it sitting on my lap while I'm wearing shorts. If it's running too hot for too long, it will grind to a complete halt, because it doesn't quite have the sense to reduce its clock speed. I had it sitting on a metal sheet as something of a makeshift heat sink, and that got warm. Were I to try this in cooler weather, I'd be inclined to have it running in the garage. Secondly, the PB does not have native USB2. It does only by virtue of an add-in card, and that can barely support powering an ipod. So if the card decides that it can't keep up, then the recovery could fail.
After going through with DiscRescue 5, it did find everything and then some. I say "and then some" because it found about 3GB of stuff that I intentionally deleted a couple years ago. Oh well, it's just hard drive space, I have plenty of that on the server. (What's 150GB to 17TB?) The real prize was the itunesDB file, which is where the playlists reside.
Here's where it gets weird. The files I was looking for were not in the folder they should have been in. The Ipod_control folder, which has the OS of the thing, was present but empty. The itunesDB and itunesControl files, which are supposed to be in that folder, were in a much different folder, and one that's not normally present in a functional ipod. This may or may not be related to the sector that needed reallocated.
But! Would everything transfer to the new hard drive that's already formatted in FAT32 when the original was HFS+, or is the Ipod software persnickety enough to throw a fit? I tried something similar back in 2014, when I went to move stuff from this one to a significantly newer one, and I ended up locking up the device hard enough that I had to restore it. As it turns out, the Gen6 and Gen7 versions have software that's just different enough.
Figuring that I might as well try, I copied the newly-extracted files to the "new" hard drive while it was on the adapter. Once that was done, I reconnected it to the ipod, turned it on, and it came up. Everything came up. It was in the same state as it was back in June when I last used it.
What's next? I've gotten everything I wanted off of it, I have the new drive and battery installed, and I've largely given up on trying to increase the capacity. When I'm done with further (non-destructive) experiments, I'll put it all back together. Except I'll modify the back case in such a way that it doesn't have all those tabs holding it on. That's a common maintenance thing; where if you remove 10 fasteners to remove an access panel, you only put 8 or 9 back on when you're done. That's because if you had to remove that panel once then you'll have to remove it again later, or those fasteners somehow got lost or damaged. Or both. Anyway.