search

NixOS / nixops-aws

GNU Lesser General Public License v3.0
52 stars 39 forks source link

nvme mountpoint change after rebooting the server #45

Open PsyanticY opened 5 years ago

PsyanticY commented 5 years ago

deploying and m5.large with an encrypted and non encrypted volume then rebooting the system cause the mountpoint to get swapped with the volume attaching point

[nix-shell:~/src/nixos-hardening/nix/nixops]$ nixops ssh-for-each -d nixos-hardened-image lsblk
test> NAME        MAJ:MIN RM SIZE RO TYPE  MOUNTPOINT
test> nvme0n1     259:0    0  30G  0 disk  
test> └─nvme0n1p1 259:2    0  30G  0 part  /
test> nvme1n1     259:1    0  50G  0 disk  /data2
test> nvme2n1     259:3    0  40G  0 disk  
test> └─nvme2n1   254:0    0  40G  0 crypt /data

[nix-shell:~/src/nixos-hardening/nix/nixops]$ nixops reboot -d nixos-hardened-image 
test> rebooting EC2 machine...
test> waiting for the machine to finish rebooting....[down]............................[up]
test> uploading key ‘ca.crt’...
test> uploading key ‘luks-nvme2n1’...

[nix-shell:~/src/nixos-hardening/nix/nixops]$ nixops ssh-for-each -d nixos-hardened-image lsblk
test> NAME        MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
test> nvme1n1     259:0    0  40G  0 disk 
test> nvme2n1     259:1    0  50G  0 disk 
test> nvme0n1     259:2    0  30G  0 disk 
test> └─nvme0n1p1 259:3    0  30G  0 part /

Notice how nvme1n1 was 50G then became 40G

ibarrick commented 3 years ago

@PsyanticY Did you ever figure this out? I had the same thing happen.