Boot frozen after updating to 6.11.2

Update:

• I took out the flash drive and plugged it into a Windows 10 PC, copied the entire contents to the desktop (as an extra backup)
• Downloaded the Unraid USB Creator and chose Stable/Unraid 6.9.2
• Copied the "config" folder over and tried again

 

Same result as before, hanging at the same line. I waited 5 mins before Ctrl+Alt+Del

Next I tried restoring from a USB flash drive backup

• Got the backup files/folders from GDrive
• Plugged my flash drive into Windows 10 PC
• Manually selected and deleted all files on the flash drive (did not format)
• Copied all files over and plugged it in

 

Next I got the below screen and boot was stuck/frozen.

 

I searched google for: "SYSLINUX 6.03 EDD unraid" and there is a mention on these forums of USB2 vs USB3 -- my mobo (GA-7PESH2) only has USB2.0, no problem there. I found a mention of UEFI vs EFI and the folder. I ran the "make_bootable.bat" as admin and replugged in my drive and tried again.

 

It booted successfully. I ran upstairs and logged into the webUI and my heart sank, see screenshot two. Ugh.. Looks like my drive configuration was lost/messed up? This was my absolute nightmare, I don't want my array corrupted. I am going to keep debugging this issue. Need to be extra careful now that my array is in question..

 

Since I was able to boot the unraid web UI I was able to download my Diagnostics.zip and make a manual flash backup (just incase). Attached is the diagnostics if someone who knows what to look for wouldn't mind helping me out (thanks in advance).

unraid-diagnostics-20221105-1458.zip

Using the CLI on unraid webUI I was able to look through the file contents in /config/ on the USB drive.

I found a few interesting files:

• DISK_ASSIGNMENTS.txt
  • Has all my disks and correctly lined up with where they should be
  • e.g. Disk: disk1  Device: WDC_WD100EMAZ-00WJTA0_XXXXXXXX  Status: DISK_OK
  • All disks have "DISK_OK" at the end (for whatever thats worth)
  • The file 1:1 matches my backups (manual and GDrive from 6 days ago)
• disk.cfg
  • This looks like a config file of 22 disks (1-22)
• disk.log
  • Nothing much in here, some setting/config (warranty="24")

 

No other files which I went through looked relevant to this issue. Given I want to get my server backup and have no where else to get more experienced help I think I will try to manually (in the UI) assign the drives back to their correct places based on the DISK_ASSIGNMENTS.txt file.

 

Fingers crossed.

After carefully filling in the dropdowns and triple checking I noted a few warning messages:

• On the two parity drives I saw: "All existing data on this device will be OVERWRITTEN when array is Started"
• At the bottom near "Start" array I saw:
  • Stopped. Configuration valid.
  • Start will record all disk information, bring the array on-line, and start Parity-Sync.
    The array will be immediately available, but unprotected until Parity-Sync completes.
    [] Parity is already valid.
• Since I luckily did my last parity just a few days ago I am going to check "Parity is already valid.". Might regret this later..
  • Last checked on Sun 30 Oct 2022 10:34:25 PM PDT (six days ago)

---

 

The array started successfully, all my docker containers were there and auto-started successfully as well! I am going to kick off another parity check just incase.

 

Not sure what to think at the moment but am glad my data is seemingly still intact. A few immediate thoughts:

• Very glad I had an offsite/cloud USB backup
  • Just checked my config and it happens every 24 hours, this seems OK
  • I am going to donate to Duplicati who was doing my backups
• I feel that I cannot upgrade to latest and get the updates and security fixes due to these core OS/boot issues
• While I love the ~7 years I have been using unraid I do feel uncomfortable with unraid as my daily driver for my server given today's issue and previous boot issues I have been having for 3+ years now (see my post history for more, tldr: all rsyslogd.pid hangs during boot)
• Lastly, I would glad pay for professional help which actually fixed my issues. Not just got me back up and running but identified the root cause and fixed it to prevent it in the future. I mentioned this in the recent survey/poll which went out

 

I don't even know if anyone will read any of this but hopefully this thread helps someone else in the future. Good luck y'all.

53 minutes ago, tone said:

DISK_ASSIGNMENTS.txt

This file is just for your information. Unraid doesn't use it, instead config/super.dat is read to get your disk assignments. If the file doesn't exist nothing is assigned.

4 minutes ago, trurl said:

This file is just for your information. Unraid doesn't use it, instead config/super.dat is read to get your disk assignments. If the file doesn't exist nothing is assigned.

Thanks for responding. Looking at my previous backup files there is:

• super.dat (4KB)
• super.old (4KB)

Both look like binary or some human unreadable file. Do you know of a way to view the contents?

Disable autostart in Disk Settings, shut down, restore config/super.dat to flash from your backup, reboot. Then you can confirm the disk assignments are correct before starting the array.

I already have it working again from manually reassigning all the devices. After today's fiasco I don't want to take anymore chances experimenting with a working server. Is there a way to check the file contents without fully restoring it?

Changed Status to Open

Changed Priority to Urgent

I just wanted to share that I had the same issue. I plugged in my UNRAID flash drive to my PC and was about to reformat the whole thing when I noticed that one file had a strange extension.

 

The file 'syslinux.cfg' was incorrectly named 'syslinux.cfg-'. On a whim, I tried removing the dash at the end, plugged the flash drive back into my server (same mobo as OP, GA-7PESH2), and everything booted with no issues. It looks like the file got renamed incorrectly in the UNRAID update process.

On 11/7/2022 at 2:03 PM, mike_y said:

The file 'syslinux.cfg' was incorrectly named 'syslinux.cfg-'. On a whim, I tried removing the dash at the end, plugged the flash drive back into my server (same mobo as OP, GA-7PESH2), and everything booted with no issues. It looks like the file got renamed incorrectly in the UNRAID update process

 

I just tried to upgrade again today (albeit to 6.11.5) but same result. I looked and syslinux.cfg had a "-" at the end of it:

syslinux.cfg-

 

I tried renaming it as you suggested and booting again. Got a different error this time, no boot still though:

Quote

boot:

Failed to load COM32 file menu.c32

 

Try recreating the flash drive as a new install with whatever version from the Download page, then copy config folder from your backup.

 

Boot from USB2 port.

Thanks for responding @trurl. Tried what you suggest with 6.11.5 and it didn't work. When I tried with 6.9.2 (original OS version) it booted again...

 

Must be something after 6.9.2 with my hardware setup?

Are you booting in SAFE mode?

No, just normal. Why?

To rule out any plugins. Booting in SAFE mode is actually a requirement for bug reports.