1

I have a new install of Debian 13 with a LUKS-encrypted hard drive (/dev/sda) with LVM. I am trying to add another encrypted drive to the LVM volume group as a new physical drive. For this, I am trying to use a USB drive.

I partitioned the USB drive (/dev/sdb) and initialized /dev/sdb1 as a LUKS container. (I noticed that without a partition table, the UUID of the USB drive kept changing.) I added a new line to /etc/crypttab, then ran update-initramfs -u

sda5_crypt UUID=<refers to /dev/sda5> none luks,discard,x-initrd.attach sdx_crypt UUID=<refers to /dev/sdb1> none luks,discard,x-initrd.attach 

While at boottime I'm prompted to unlock sda5_crypt, I'm not prompted to do so for sdx_crypt. It is still not unlocked even if sdx_crypt is used (e.g. it is a PV in the volume group).

cryptdisks_start sdx_crypt works fine, but I think it reads /etc/crypttab directly.

Is /etc/crypttab obsolete for what happens at boot? Or can it be that /dev/sdb(1) is not ready at this point? I see no error messages that would indicate this.

I notice that on a Debian 12 machine where multiple encrypted volumes work OK, the options in crypttab are luks,discard,initramfs — it seems to me this is the old way of specifying what volumes should be processed by initramfs.

1 Answer 1

3

It seems in Debian 13 it is still necessary to add the initramfs flag to lines in crypttab for them to be copied into initramfs. I realized it after some digging into the initramfs preparation scripts. What used to pick up the first line from crypttab I don't know.

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.