1

I was trying to install a new package on 16.04 last night, but received an error indicating /boot was full. I've run into this before, and the solution was to purge files, so I decided to just purge one, apparently the wrong one.

(I saw vmlinuz-4.15.*-generic and vmlinuz-4.8.*-generic. I was late, so I deduced that 4.8 was old and should keep the 15s...)

Anyway, on reboot, I got file not found, you need to load the kernel first.

I found directions on recovering a deleted kernel, but I've confused myself. I have dual 3 drives, one with 12.04 (which is where I am know) and Windows (7? I don't actually use it), and then a second drive with 16.04, and a really old one with maybe Windows XP.

parted -l shows:

Model: ATA WDC WD1600JB-00G (scsi)
Disk /dev/sda: 160GB Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End    Size   Type     File system  Flags
 1      32.3kB  160GB  160GB  primary  ntfs         boot


Model: ATA WDC WD1002FAEX-0 (scsi)
Disk /dev/sdb: 1000GB Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End     Size    Type      File system     Flags
 1      1049kB  530GB   530GB   primary   ntfs            boot
 2      530GB   1000GB  470GB   extended
 5      530GB   996GB   466GB   logical   ext4
 6      996GB   1000GB  4293MB  logical   linux-swap(v1)


Model: ATA WDC WD1003FZEX-0 (scsi) 
Disk /dev/sdc: 1000GB Sector size (logical/physical): 512B/4096B
Partition Table: msdos

Number  Start   End     Size    Type      File system     Flags
 1      1049kB  8000MB  7999MB  primary   linux-swap(v1)
 2      8001MB  1000GB  992GB   extended
 5      8001MB  33.0GB  25.0GB  logical   ext4
 6      33.0GB  33.5GB  499MB   logical   ext4
 7      33.5GB  43.5GB  9999MB  logical   ext4
 8      43.5GB  53.5GB  9999MB  logical   ext4
 9      53.5GB  1000GB  947GB   logical   ext4

On boot, grub offers choices as being on sdb5. Sdb5 only list vmlinuz-3.*, so that would be the the install for 12.04. The files for 16.04 seem to be on sdc6. Edit: It looks like sdc5 may be the mount point for that drive, since the listing for that shows /root, /boot, /bin and so on.

So,after all that info dump, my question is:

Which partitions to I mount to what? /dev/sdb5 or /dev/sdc5 to /mnt and then /dev/sdc6 to /mnt/boot? or...? (And also, when I install the kernel, will it figure out I need 4.8, or will I need to specify it?)

Thanks.

Samuel Philipp
  • 417
user3088551
  • 11
  • 2
  • It's usually easier to choose a different kernel for booting from the GRUB menu. – user535733 Mar 07 '19 at 23:36
  • Do you have a Legacy/MBR or UEFI/GPT system? Also, it might be easiest to re-install. – Aaron Franke Mar 08 '19 at 00:03
  • user535733 - That's what I finally did, about midnight last night - ran update-grub from my old 12.04 install. Of course, that was after I'd use a the live disc to explore the various partitions, and guess (I think correctly) that sdc5 was the root partition I was looking for, and then after following all the recovery steps, have apt-get inform me the kernel was already present and up-to-date. Thanks for the hint! – user3088551 Mar 08 '19 at 19:02

0 Answers0