Score:0

Bad magic number: which command to verify filesystem health?

cn flag

Drives 1 and 2 are bootable (not live) 64GB USB flash drives on which I have installed Ubuntu 22.04.1 LTS. I boot the computer using Drive 1, to study the health of Drive 2 (/dev/sdb).

Drive 2 is divided into 2 MiB unallocated, followed by /dev/sdb2 (477 MiB FAT32) and /dev/sdb3 (59 GiB ext4). Both partitions are at least half-empty.

When I ran sudo fsck /dev/sdb I got "Bad magic number in super-block" results similar to those in another thread. The advice in that thread was, in effect, to run sudo fsck -f /dev/sdb3. For me, that command went through passes 1 through 5 with no reported errors. A check of sdb3 in GParted likewise reported no errors.

Those fsck results seemed inconsistent. I thought the explanation might be that the "bad magic number" message was referring to the FAT32 partition. But that wouldn't make sense: the message acknowledged that partition: "Found a dos partition in /dev/sdb." Surely the message was not telling me that a DOS partition would not function as a proper Linux partition. The message seemed rather to be highlighting a problem with sdb3.

The "bad magic number" message itself seemed to recommend running sudo e2fsck -b 8193 or 32768 (the two superblock numbers it suggested). With 32768, the command seemed to work, but it did not seem to have any effect: it purported to "fix" many problems that, on re-run, were still there.

A participant in a different discussion, responding to the same "bad magic number" message, suggested both mke2fs -n /dev/sdb3 and fdisk /dev/sdb. Not sure what these might add, or why they would be on target.

I seem to be flailing around without a clear understanding of the problem and its solution, or even whether there is a problem.

I sit in a Tesla and translated this thread with Ai:

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.