On 19/05/2021 06.52, J Leslie Turriff wrote:
Looking at the messages that come up during boot, I see a 30 second delay while systemd tries to Resume from Hibernation. Looking at the Services settings in YaST, I see that all services related to this are disabled. How do I tell systemd not to attempt this during boot? (Does this happen when initrd is active, or after switching to real /?)
Resume from hibernation happens very early and is done by the kernel, not systemd. The kernel loads and at some point very early checks the swap partition to see if there is an hibernation image there. If not, it continues. It is controlled by the "resume=..." option in the kernel line.
| From YaST => System => Services: | │systemd-hibernate │Manually │Inactive (Dead) │Hibernate │ | │systemd-hybrid-sleep │Manually │Inactive (Dead) │Hybrid Suspend+Hibernate │ | │systemd-suspend │Manually │Inactive (Dead) │Suspend │ | | From journalctl: | May 18 21:48:39 pinto systemd: dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x2dpart3.device: Job dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x> | May 18 21:48:39 pinto systemd: Timed out waiting for device dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x2dpart3.device. | May 18 21:48:39 pinto systemd: Dependency failed for Resume from hibernation using device /dev/mapper/cr_ata-ST1000DM010-2EP102_Z9AFBY9T-part3. | May 18 21:48:39 pinto systemd: systemd-hibernate-resume@dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x2dpart3.service: Job systemd-hibernate-resume@de> | May 18 21:48:39 pinto systemd: dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x2dpart3.device: Job dev-mapper-cr_ata\x2dST1000DM010\x2d2EP102_Z9AFBY9T\x> | | From /etc/crypttab: | swap UUID=618ca366-638c-4bf8-bd86-917f0eac83db /dev/urandom | # cr_ata-ST1000DM010-2EP102_Z9AFBY9T-part3 UUID=618ca366-638c-4bf8-bd86-917f0eac83db /dev/urandom
Despite the messages, systemd can not control resume from hibernation, has to be something else but related. I would check fstab. Whatever the partition is used for, it has to be opened.