[Solved] Suspend to Disk: systemd-hibernate.service Fails

Question

Howdy-ha, folks.  I know this looks pretty mundane, but for the life of me I can't get hibernation to work.  Each time I run systemctl hibernate the screen will go blank as though preparing to suspend, then come back on; journalctl shows the same error messages:
PM: Cannot find swap device, try swapon -a.
PM: Cannot get swap writer
Failed to start Hibernate.
Dependency failed for Hibernate.
Service sleep.target is not needed anymore. Stopping.
Unit systemd-hibernate.service entered failed state.
This happens when swap is mounted, and "swapon" says as much.  I've been working on this for about four hours now, and none of what I've found seems to pertain to my situation (most of the "solutions" were "Increase partition size').
What I've tried:
- Three different kernels---linux-ck custom build and linux-ck-sandybridge from graysky's repo (both with BFQ and CFQ: see below), as well as the stock Arch kernel
- A 2 Gb swap file at /swapfile on my SSD;
- A 4 Gb swap partition on /dev/mmcblk0 (an SD card, and my preferred method/location, if possible);
- An 8 Gb swap partition on HDD /dev/sdb2
Reasons for wanting hibernate rather than suspend-to-RAM:
- Maximize battery life.
- There's some kind of issue with Linux-ck and suspending to RAM; it'll work once, then freeze in a kernel panic the second time, forcing a hard reset.  The stock kernel suspends to RAM, but still won't hibernate
- I put a lot of time into building a custom Linux-ck kernel, and would rather not have to start the configuration over with the stock kernel if possible.
I'm working with limited space on my boot drive and, as mentioned, increasing the amount of swap has no bearing on this.  The resulting phenomenon and error messages are the same, regardless of size and medium.  I've exhausted my ideas on where to look, and what to even look for.  Thanks a lot to everyone in advance (especially you,karol; don't think I don't see you lurking back there )
Last edited by ANOKNUSA (2013-11-17 05:32:14)

Answer 1

I rebooted, so trying to recreate the same situation:
$ free -h
total used free shared buffers cached
Mem: 7.7G 7.2G 525M 0B 32M 2.3G
-/+ buffers/cache: 4.8G 2.9G
Swap: 3.7G 0B 3.7G
top reports nearly the same thing:
top - 20:00:13 up 22 min, 1 user, load average: 0.12, 0.22, 0.14
Tasks: 155 total, 1 running, 154 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.1 us, 0.3 sy, 0.1 ni, 99.5 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem: 8061380 total, 7500096 used, 561284 free, 34432 buffers
KiB Swap: 3872236 total, 0 used, 3872236 free, 2417172 cached
htop, however, reports about 1.5 Gb more free RAM.  After reboot, swappiness is still set to 100 with the same results.
EDIT:
ANOKNUSA wrote: I'll test it again with the other locations mentioned in the OP, and get back to you.
Even though I rebooted, I haven't done this yet, so ignore it for the moment.
Last edited by ANOKNUSA (2013-08-28 01:06:10)