I'm running v2 final software and haven't actually uploaded any albums for months and months. I'm running preinit, telnetd, ttsclock at startup (strongbad voices, of course), and charcoalgray99's custom UI for khttpd including the VNC daemon.

Got some new albums and was going to upload them to the empeg using jEmplode. The sync performed a full fsck and I minimized jEmplode while waiting. To my surprise, after the sync completed, the changed I had made were still marked red. Must be a bug in jEmplode I figured. I closed jEmplode and thought nothing more of it. Later I was going to play some of my new albums and noticed they were nowhere to be found on my empeg.

Later in the week I try uploading them again. This time no full fsck is performed, but the sync happens WAY too fast. Notice that right after jEmplode reports it is checking filesystems, it reboots the player immediately. I try performing a forced fsck (Tools menu). JEmplode reports there is a serious error on my disk and to check the log for more details (couldn't actually find the log file, but didn't try too hard). Log into the empeg via telnet and run a forced fsck on all partitions. They all check out OK... HRRRM.... oooook. Run forced fsck from jEmplode again, still get "serious error"... Argh!

Boot to windows... download native emplode, run native emplode. Try making minor change (move a playlist) and sync. Native emplode returns quickly and plays the "hooray!" (done syncing) audio clip. Check the playlist and it hasn't moved! Turn debugging output on for native emplode. Notice "CheckMedia failed, status = 0xc0041010". Look up error code here: http://www.trejan.com/empeg/index.html?errcode=0xC0041010#ERRORCODE ant notice it claims the device is busy. Hrm, ok. Log back into player via telnet. View /etc/fstab /etc/mtab /proc/mounts. Ponder. View processes (ps auxfw). Notice that both vncserver and telnetd are running from /drive0. Hmmm. Kill vncserver (can't kill telnetd, I'm using it!). Move telnetd from /drive0 to /var/telnetd to test. Change telnetd's preinit.d script to load from /var/telnetd instead. Reboot player. Login via telnet, kill vncserver.

Try uploading again: Successful Sync!

Hope this helps somebody 6 months from now when they do the same stupid thing as I did

Not sure exactly what was going on from your post, but I think you're describing the long-standing and well-known "can't synch while a process is running from drive 0" problem. It's described in the FAQ here.

That is indeed what was going on.

The funny thing is, I read that FAQ entry but somehow glanced over the explicit warning about processes running from drive0. However, I inferred the same thing from the suggested rememdy of:

ro
umount /dev/hda4
umount /dev/hdc4
swapon /swapfile
fsck -fay /
fsck -fay /dev/hda4
fsck -fay /dev/hdc4
swapoff /swapfile
sync

Thanks for bringing this up. I've had some syncs fail when I'm running CharcoalGrey99's interface w. VNC. I used the default install location, but maybe I should check that out.

I run Charcoalgray's interface and don't have any problems because it's not a "running process" per se. It's just some files that Hijack reads off the hard disk and forwards to you when you request them via HTTP. Having that interface in place doesn't cause the synch trouble.

I don't know about VNC as I don't run that and don't know how it works player-side.

This is funny... I had been having the same problems, and just tonight figured out what it was (via the FAQ of course!). I am sitting wating for a fsck to complete, and popped in here to find this... I need to keep up on Technical more often!