I don’t know if the bug is reproducible. I’m going to try restore a backup.
Found in
Version: 1.8.0
Build: 840
Attachments
When I ran disk check on my Mac I saw
% sudo fsck_msdos /dev/rdisk2s1
** /dev/rdisk2s1 (NO WRITE)
** Phase 1 - Preparing FAT
** Phase 2 - Checking Directories
/MyProjects/(???1i.n?? starts with cluster out of range (496977500)
Remove? no
/MyProjects/3JI?Ȑ?f.=P? starts with cluster out of range (600824261)
Truncate? no
When I run a diff -r against a backup…
Pattern files in projects that I didn’t open are different
Binary files /Volumes/TRKR231221/MyProjects/cas test midi 60 patt/patterns/pattern_92.mtp and /Users/ger/Documents/Tracker/Backups/20240318/MyProjects/cas test midi 60 patt/patterns/pattern_92.mtp differ
There’s garbage file names on the disk:
Only in /Volumes/TRKR231221/MyProjects: òkâ[wá(.ü+Ñ
Only in /Volumes/TRKR231221/MyProjects: o„¦)w.â}é
Only in /Volumes/TRKR231221/MyProjects: q;q3ek.âb~
Only in /Volumes/TRKR231221/MyProjects: rhŒs’)Éô.58f
Only in /Volumes/TRKR231221/MyProjects: r¥"Ð3CV.
That’s crazy!
I had a similar experience this weekend…
…with the difference that my SD-Card (the original which came with the tracker) was unusable altogether.
Neither the tracker nor my Mac could read it anymore. It would not show up in the finder anymore.
Disk utilities app couldn’t find it.
All my projects (samples and recordings) so far are gone.
The card has not been moved inbetween. It was in the tracker, worked the day before, next day it’s dead.
Well, on the positive side I don’t have to think about all the unfinished pieces of music anymore that I was collecting since I got the unit.
I think I have enough weird tracker adventures for now. Starting with a unit, that froze every twenty minutes, a radio that never worked and now a destroyed SD-card for no apparent reason than being in the tracker over night.
The SD-Card that came with my tracker died on day 1 when I was transferring files onto it from my Mac - same symptoms as yours - luckily I didn’t lose any compositions. Maybe some bad cards got shipped. The card I bought to replace it has so far been rock solid.
Upgrading to 1.8 for me was fine, but I have had sd card corruption like this maybe 2 or 3 times just generally from reading and writing to the sd card in windows, im not really sure if it’s an issue with the tracker specifically, but it’s happened a lot in comparison to for example my camera’s sd card where i don’t think ive ever had it corrupt
if there’s any way to reduce the chances of this happening just in general it would be pretty great for me, i’ve lost projects and had to spend a couple hours repopulating all my samples a few times now (but again, upgrading for 1.8 for me was fine, but in general this has happened, with the strange characters for names and everything)
I used the original card for some month now. It died with 1.8
I wanted to show a friend how cool the Tracker is and he witnessed how the machine asked to insert a card while it was in there. This demonstration proved pretty much the opposite…
I took out the card, put in the Mac, Mac couldn’t find it either, no chance of reviving.
I take out the other card now and go back to the previous firmware… just to be safe.
ger@Gers-MBP-2 TRKR231221 % sudo fsck_msdos /dev/rdisk2s1
Password:
** /dev/rdisk2s1 (NO WRITE)
** Phase 1 - Preparing FAT
** Phase 2 - Checking Directories
** Phase 3 - Checking for Orphan Clusters
5576 files, 12870528 KiB free (1608816 clusters)
MARK FILE SYSTEM CLEAN? no
***** FILE SYSTEM IS LEFT MARKED AS DIRTY *****
I put the card into the tracker (still running 1.8.0) turned it on. Waited 30s and turned it off without doing anything. Brought the SD card back to computer to restore the rest of the backup and ran fsck first thing. It now reports “Free space in FSInfo block (1608802) not correct (1608837)”.
ger@Gers-MBP-2 ~ % cd /Volumes/TRKR231221
ger@Gers-MBP-2 TRKR231221 % sudo fsck_msdos /dev/rdisk2s1
** /dev/rdisk2s1 (NO WRITE)
** Phase 1 - Preparing FAT
** Phase 2 - Checking Directories
** Phase 3 - Checking for Orphan Clusters
Free space in FSInfo block (1608802) not correct (1608837)
Fix? no
5573 files, 12870696 KiB free (1608837 clusters)
MARK FILE SYSTEM CLEAN? no
***** FILE SYSTEM IS LEFT MARKED AS DIRTY *****
I downgraded to 1.7.1. Did nothing else. Brought the SD back to the computer. Reran fsck which “Found 331 orphaned clusters”.
ger@Gers-MBP-2 ~ % sudo fsck_msdos /dev/rdisk2s1
** /dev/rdisk2s1 (NO WRITE)
** Phase 1 - Preparing FAT
** Phase 2 - Checking Directories
** Phase 3 - Checking for Orphan Clusters
Found orphan cluster(s)
Fix? no
Found 331 orphaned clusters
Free space in FSInfo block (1614491) not correct (1613144)
Fix? no
5637 files, 12905152 KiB free (1613144 clusters)
I compared everything on the SD card to the backup. The differences:
.fseventsd/fseventsd-uuid changed
a project.mt file changed (I assume the open project)
a new file Workspace/patterns/patternsMetadata
I fixed the orphan clusters and put the SD back in the tracker (now 1.7.1). Turned it on, then off and brought the SD card back to the computer. I saw “Free space in FSInfo block (1618250) not correct (1618251)” but no orphaned clusters.
Plus 23 recordings freshly made last night with the new 133 second recording limit.
OK, I got so wound up with this sudden data loss that in pure caveman anger I whipped out the SanDisk 256 Ultra micro sd card while the Tracker was switched on.
I was greeted with please insert sd card.
So I did.
Now the data is back, all projects loading.
This is most bizarre. I also cleaned the gold contacts on the SD Card.
1.8 firmware seems to freak out a few systems here and there.
Hi @here, Thank you for reporting this bug. We are sorry you’re experiencing issues with the 1.8.0 firmware. We will do our best to resolve it and notify you when it’s ready.
Just updated to 1.8 with the original 16gb SD card that came with and all my samples wiped too. Luckily I always back up everything but this seems to be a prevalent issue with 1.8. Hopefully this is resolved on next update.
Hi, I am also having this issue. Not sure if it was to do with 1.8, all I know is that the tracker can read the SD card but when I put it in the PC it says it’s corrupted and unreadable, so I’m now completely unable to interface with my PC - I can try another card but it means I’ve lost hours of work, since this occurred when I was about to backup what I’d done.
Thanks for the reply - yes, I thought it was weird. I was using the little card-reader-dongle that came with the Polyend. I bought another card reader today (‘Mogood’ brand - I tested it and it seems to work fine with other SD cards), and the computer still thinks it is ‘corrupted’; I also tried it in different USB ports to see if that was the issue - no dice.
I’ve ordered another micro SD card to see if it’s the card itself.
If so I’ll have to try putting the folder structure on it, putting it in the tracker, saving my work to the new card (since the tracker appears to retain the last project you worked on in the memory), and then putting the rest of my backup onto the new card - maybe this will fix the issue and rescue my work.
So! I got a new card and tried my above method - nope. Upon setting up the folder structure in the new card and putting it in, and it said there was no SD card.
In the end, I copied out my work by hand from the screen, put the old SD card in to the PC and formatted it, copied my backup onto the formatted SD card, then went back to the tracker and put it back to Firmware 1.7.1 and it worked. Then I re-entered my patterns from my notes. Annoying, but it seems fixed.
PS I then took the plunge and tried updating the firmware again to 1.8.0, and the error is not recurring - yet! It it fails again, I’ll come back and report it since it would imply it’s the firmware that’s at fault.
PPS - The error is nor recurring, which is great. I wonder if the problem is actually the card-reader-dongle given out free with the Trackers? Since I’ve been using a better card-reader, the fault has not recurred, and actually, several times before I’ve had issues with the card being unreadable, but not since my new card-reader…
