You are not logged in.
As you can see in this screenshot of a virtual machine screen, during the boot systemd fails to activate swap blaming a timeout.
Yet the device is working, otherwise the rest of the system would not boot, and in fact I can use it from the command line.
In fstab I have this line:
/dev/mapper/encb1 none swap defaults 0 0
I tried to look for that .device file, but as far I can say it is not in the disk.
What can I try? Any insight? Thanks
Last edited by ezzetabi (2013-12-30 16:00:57)
Offline
(Posting pictures of text is generally frowned upon here.)
What is your drive setup?
Offline
(I did not really have a choice since it is a vm...)
/dev/mapper/encb1 is a partition inside a cryptsetup block (plain dm-crypt) special.
Last edited by ezzetabi (2014-01-04 21:45:57)
Offline
I believe this is the excepted behavior under systemd. There have been at least one other thread on this topic.
CPU-optimized Linux-ck packages @ Repo-ck • AUR packages • Zsh and other configs
Offline
the expected behavior is waiting of a device that is already good for use?
Offline
The most recent releases of systemd have special GUIDs for swap and /home partitions. If these are set, then it will try to automount them. Of course, the logic is supposed to be that entries in the fstab should override this, but judging by the forum threads that have been popping up over this of late, this does not seem to be the case.
BTW, had you done a little searching for this issue, you would have found a number of threads.
Offline
I see, I am probably misunderstanding the problem since I cannot find anything. So it is a bug of systemd that will be probably fixed soon. Thanks.
Offline
I read around, but still I have the problem. The GUID problem does not seem to apply as I am using a msdos partition table, or at least this is what partprobe -s states when finding them. And neither are marked as Swap, jsut as Linux.
This is the fdisk output. (This is a real machine, no more pictures! )
Disk /dev/mapper/encssd: 119.2 GiB, 128035676160 bytes, 250069680 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x8b7370dc
Device Boot Start End Blocks Id System
/dev/mapper/encssd1 2048 50333695 25165824 83 Linux
/dev/mapper/encssd2 50333696 250069679 99867992 83 Linux
Offline
Not sure anyone actually read the first post. This is about lvm activation, not the gpt generator.
There's plenty out there about this, including resolved bug reports on our own flyspray. Nothing here is working as intended.
Offline
Sorry guys, I posted this in the "Newbie corner" for a reason. The LVM you are speaking about is the logical volume manager?
I do not think it is correlated, I have no lvm2 in my mkinitcpio configuration.
But, perhaps, since I have partitions inside a plain dm-crypt device is the same problem?
You see?
$ lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 931.5G 0 disk
└─encplate 254:3 0 931.5G 0 crypt
sdb 8:16 0 119.2G 0 disk
└─encssd 254:0 0 119.2G 0 crypt
├─encssd1 254:1 0 24G 0 dm [SWAP]
└─encssd2 254:2 0 95.2G 0 dm
sr0 11:0 1 1024M 0 rom
encplate and encssd2 do not appear mounted because I used zfs.
Offline
Just using the UUID in fstab solved the issue
UUID=0bea8e5e-6d77-49d8-b391-3922d5e7991c swap swap defaults,discard 0 0
Offline