Hi Clarity et all. Apologies for butting in. I don't think this will help, but sometimes an idiot (EDIT- i.e. me, just to be really clear) asks something that triggers things for those (such as rufwoof) that do have knowledge and understanding. Multi-session is one of the ways I run with Fatdog64. However, it is on a USB drive using Grub4Dos to boot. So it is not the same as your situation.
First off, I presume (since you have multi-session working) that you would have read the Fatdog64 Help -> boot options -> Advanced Parameters -> savefile stuff. I know you have multi-session running, but have to ask.
1.6 Gig. Wow. That seems more than just simple configuration stuff. Would there be a chance that there is some data in one save that has subsequently been deleted in a following one? I am not trying to pry, and I know it doesn't answer the question of the base "set" not being generated, but have to ask.
As you have the files, you have obviously backed them up. Would it be worthwhile maybe trying again with just the original "base" sfs files and the first of your real saves (just as a test to minimise the variables)?
Maybe try:
Copy the original two (2) "base set" sfs files back to the location that Fatdog will look for the multi-session saves.
Copy the sfs file of your first non-base save to the same location.
After confirming that your backups are complete, move them out of harm's way (I know they probably already are, just have to mention it).
Delete all other stuff from the location that Fatdog will look for multi-session files.
Run a fresh boot of Fatdog, so that it only "picks up" the stuff you want. Maybe have a quick look to check that something you would expect from your first save is there (just to make sure things are still working). EDIT- Maybe the time.
Mount the partition with the multi-session files (I presume that, like the USB partitions, the hard drive multi-session partition won't be mounted, but will appear on the desktop). Delete all the files in that location (i.e. the ones you had copied earlier).
Unmount that and all other partitions.
Click on the "save.session" icon. Once Fatdog has processed the data it should, I think, mount the multi-session partition, write the new files and then unmount the partition again. Note that I am just going off the top of my head and could have it wrong.
As usual I am too verbose and, again, I don't know if it will help. But it might be worth a try.
Thanks.
EDIT 2: Sorry about adding more but some others things came to mind. If I am asking about info that you have already given above then I apologise- I'm just going on the fly.
a) I wonder if the process is simply not getting a chance to complete.
i) maybe try invoking the process using the shutdown path rather than using the save.session icon. Theoretically Fatdog should be able to complete the process, and any writes, before shutting down.
ii) Could Fatdog be running out of usable RAM for the process? I don't know how you would tell, but there is some mention of RAM usage and considerations in the "How to split... the Fatdog.... initrd..." (I forgot the name of the thread, of course), although it is in the Fatdog section, so I think you should find it. I think Rufwoof also details an alternative swap mechanism in that thread.
b) Are you using an ext4/3/2 filesystem for the multi-session files? Or is it a non-native filesystem (NTFS, Fat32). Again, I know things are working apart from this aspect, but just have to ask.
c) If you might have been experimenting with stuff, could there be some active code running that might interfere with the process?
Just thoughts that I don't know will be of any help. I'll try to run away and leave everyone in peace now
Thanks.