Sstrhsrtj

I use full disk encryption which uses an LVM. It uses like this in lsblk:

NAME                                          MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT

sda                                             8:0    0 465,8G  0 disk  

├─sda1                                          8:1    0   200M  0 part  /boot/efi

├─sda2                                          8:2    0   500M  0 part  /boot

└─sda3                                          8:3    0 465,1G  0 part  

  └─luks-3f530000-b2c3-4ba3-9e85-1a96494cc25d 253:0    0 465,1G  0 crypt 

    ├─fedora_martin--friese-root              253:1    0    50G  0 lvm   /

    ├─fedora_martin--friese-swap              253:2    0   7,8G  0 lvm   [SWAP]

    └─fedora_martin--friese-home              253:3    0 407,3G  0 lvm   /home

The same setup I had with Ubuntu for several non-LTE releases as well. On the first installation, it worked fine on both Ubuntu (say 14.10) and now Fedora 22. When I upgraded the Ubuntu I occasionally had with issue (around 15.04 I'd say). On Fedora 22 it was not a problem until the upgrade to Fedora 23. Now I have the A stop job is running errors.

In total there a couple errors that cycle through (except that the time is increasing):

• (1 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / 1min 30s)


• (2 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / 1min 30s)


• (3 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / 1min 30s)


• (4 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / 1min 30s)


• (5 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / 1min 30s)


• More of them. I did not get to make pictures of all of them. But with the pattern I assume that number 6 and 7 are the same type.


• (8 of 8) A stop job is running for LVM PV NVB… on /dev/dm-0 (20 s / no limit)

The last one has no time limit, which is interesting. After the timeout is over, it dumps a couple more messages and leaves me with the following:

http://chaos.stw-bonn.de/users/mu/uploads/2015-11-27/LVM-Fehler-7.jpg

From the last line it seems that it cannot properly unmount and luksClose the encrypted volumes.

My setup is just plain Awesome WM (and has been that on Fedora 22 as well). I usually log off in Awesome wM and then click the poweroff button in SDDM, the display manager. That way I do not have to enter systemctl poweroff and I have the hope that all my user programs are closed before the shutdown starts.

What can I do to find the thing that blocks the proper closing of the LVM crypto devices?

Update 2016-01-05

I have uploaded a log from the whole shutdown. In the end it does say that it was not able to close the device as it was busy but I do not see why it has been busy. What is the cause for this?

This seems related to https://bugzilla.redhat.com/show_bug.cgi?id=1299178.
If I read this properly it's because the setup of LVM on top of encrypted device isn't well supported by systemd !? This surprises me because, well, it seems like an obvious setup to me. Encrypt a big partition, then split it into as many logical partitions as desired, seems easier than the other way around...