Suspend acting like Hibernation

General Help
1

I have noticed since using Zorin 17 that when the computer suspends itself, waking up from suspend takes 10-15 seconds for any response (I.e. monitor turns on and you can see the password login field and move the mouse, but can't click anywhere for 10-15 seconds).

Almost like it is taking additional time to move the data from the swap to RAM instead of it remaining in RAM. Any thoughts on how to get rid of this delay. I didn't experience this issue with Kubuntu which I was using prior.

image
image1042×747 54.6 KB

Some basic details on the system (desktop)

Summary

OS: Zorin OS jammy 17 x86_64
Host: B560 DS3H AC-Y1
Kernel: Linux 6.8.0-48-generic
Shell: bash 5.1.16
DE: GNOME 43.9
WM: Mutter (X11)
CPU: 11th Gen Intel(R) Core(TM) i5-11600K (12) @ 4.90 Gz
GPU: Intel UHD Graphics 750 @ 1.30 GHz [Integrated]
Memory: 5.14 GiB / 15.41 GiB (33%)
Swap: 976.02 MiB / 2.00 GiB (48%)
Disk (/): 106.10 GiB / 915.32 GiB (12%) - ext4

2

B560 DS3H Suspend Delay

Based on the provided search results, it appears that the B560 DS3H AC-Y1 motherboard, specifically with the Y1 BIOS version, may experience a delay when waking from suspend in Linux.

One possible solution mentioned in the search results is to enable the disable-wakeup.service systemd service. This service can be enabled using the command sudo systemctl enable disable-wakeup.service and started using sudo systemctl start disable-wakeup.service. This service disables wake-up events for devices, which may help resolve the delay issue.

Another potential solution is to modify the /etc/systemd/system/disable-wakeup.service file using sudo vim /etc/systemd/system/disable-wakeup.service. This file controls the disable-wakeup service and can be customized to suit specific system configurations.

Additionally, some users have reported success with setting the GP30 value in /proc/acpi/wakeup using the command sudo sh -c 'echo "GP30" > /proc/acpi/wakeup'. This command sets the wake-up device to GP30, which may help resolve the delay issue.

It’s also worth noting that the search results suggest that updating the BIOS to a newer version may help resolve the issue. However, this is not explicitly mentioned as a solution for the delay on waking from suspend.

It’s recommended to try the disable-wakeup.service and /proc/acpi/wakeup modifications first, as they are specifically mentioned in the context of the B560 DS3H AC-Y1 motherboard and Linux. If these solutions do not resolve the issue, further troubleshooting or BIOS updates may be necessary.

Entity 1: disable-wakeup.service - a systemd service that disables wake-up events for devices.

Entity 2: /proc/acpi/wakeup - a file that controls wake-up devices and can be modified to set specific wake-up devices.

Note: The provided search results are based on user experiences and may not be exhaustive or universally applicable. It’s essential to test and verify any solutions in your specific environment before considering them effective.

[On a personal note, another reason why systemd sucks.]

1 Like
3

Thanks for the info. disable-wakeup.service does not exist on my system. Not sure why as I did a standard fresh install of 17.2.

I tried setting the GP30 value in /proc/acpi/wakeup and will report back.

This makes more sense now as to why I started seeing with Zorin as I switched to Zorin when my old motherboard died and I replaced it with this one. From what I've read B560 DS3H AC-Y1 is BIOS locked and can't be updated. :roll_eyes:

As of last month I was on the latest BIOS version available.

1 Like
4

Unfortunately, the GP30 entry doesn't exist in my /proc/acpi/wakeup file and probably explains why I still see no difference. I also don't have a service called disable-wakeup.service so I'm kind of stuck on whether or not this problem is resolvable without new hardware.

5

Sorry I can't help you further friend. It was all I could find, as well as what you have discovered about the BIOS not able to be updated beyond what you have done. My offspring are Windows 11 users, despite my protestations, that said the eldest needs it for education software she purchased during lockdown, and the youngest has it at work so guess they need to be familiar with it. I have had to fork out for a TPM 2.0 chip for their motherboards as there was no BIOS update, unlike my AMD rig which had a BIOS update for it, but I leave it disabled.

6

if system go into suspend/hibernate , it writes content to swap.
ideally swap size should be same as ram , otherwise it would struggle
also look into disabling USB auto suspend .