TensorDockT
TensorDock
Jun 1

GRUB/EFI dpkg error!

GRUB/EFI dpkg error is happening consistently across multiple regions (Vancouver, Detroit) and Ubuntu versions (22.04, and what appears to be a 22.04 base even when 20.04 is selected) specifically for the 2x RTX A6000 configuration. error message: mount: /var/lib/grub/esp: special device /dev/disk/by-id/scsi-0QEMU_QEMU_HARDDISK_hd0-part15 does not exist. This indicates a problem with their base image or provisioning templates for this hardware. You need to provide a VM where sudo apt-get update && sudo apt-get install coreutils runs without dpkg errors. What the Documentation/Similar Issues Suggest Could Be Tried (Advanced - Use with Caution, or ideally, by TensorDock Support): The ullright.org link shows a complex process to manually fix debconf settings for grub-pc and grub-efi by: Backing up debconf selections for GRUB. Removing lines related to grub-efi if the system is not actually EFI-booted (or correcting the EFI device path if it is). Purging and re-applying these modified debconf settings. Running dpkg --configure -a and potentially dpkg-reconfigure grub-pc (or grub-efi-amd64 if it's truly an EFI system). This is generally beyond what a user should have to do on a fresh cloud VM. It requires correctly identifying if the system is BIOS or EFI, finding the correct boot device paths, and carefully editing configuration database entries.
PendingPending