Suspend do not work - instant wake-up

I have a fresh installation of Tumbleweed, up for a couple of weeks. Anytime the system goes to suspend due to inactivity or manual selection, it goes to the process, turns off and immediately back on to login screen.
All status is preserved, so no restart or whatever. How to help him ‘sleep well’? :slight_smile:
AMD 5600x / 6900xt / 32gb ram.

Did some googling before and maybe this information will be helpful:


cat /sys/power/state
freeze mem disk

dmesg | grep -i acpi
 5224.753950] ACPI: PM: Preparing to enter system sleep state S3
 5225.090202] ACPI: PM: Saving platform NVS memory
 5225.107755] ACPI: PM: Low-level resume complete
 5225.107773] ACPI: PM: Restoring platform NVS memory
 5225.110610] ACPI: \_SB_.PLTF.C002: Found 2 idle states
 5225.116812] ACPI: \_SB_.PLTF.C004: Found 2 idle states
 5225.120515] ACPI: \_SB_.PLTF.C006: Found 2 idle states
 5225.124536] ACPI: \_SB_.PLTF.C008: Found 2 idle states
 5225.128535] ACPI: \_SB_.PLTF.C00A: Found 2 idle states
 5225.132585] ACPI: \_SB_.PLTF.C001: Found 2 idle states
 5225.136621] ACPI: \_SB_.PLTF.C003: Found 2 idle states
 5225.140627] ACPI: \_SB_.PLTF.C005: Found 2 idle states
 5225.144638] ACPI: \_SB_.PLTF.C007: Found 2 idle states
 5225.148688] ACPI: \_SB_.PLTF.C009: Found 2 idle states
 5225.152715] ACPI: \_SB_.PLTF.C00B: Found 2 idle states
 5225.155380] ACPI: PM: Waking up from system sleep state S3
 5225.676972] Modules linked in: ccm rfcomm snd_seq_dummy snd_hrtimer snd_seq nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat af_packet nf_tables ip6table_nat ip6table_mangle ip6table_raw ip6table_security iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 iptable_mangle iptable_raw iptable_security nfnetlink ip6table_filter ip6_tables iptable_filter bpfilter cmac algif_hash algif_skcipher af_alg qrtr bnep nls_iso8859_1 nls_cp437 vfat fat intel_rapl_msr intel_rapl_common iwlmvm mac80211 edac_mce_amd libarc4 btusb snd_hda_codec_realtek btrtl kvm_amd snd_hda_codec_generic btbcm snd_hda_codec_hdmi ledtrig_audio btintel kvm iwlwifi btmtk snd_hda_intel irqbypass pcspkr gigabyte_wmi wmi_bmof bluetooth efi_pstore snd_usb_audio snd_intel_dspcfg k10temp i2c_piix4 snd_intel_sdw_acpi cfg80211 snd_hda_codec snd_usbmidi_lib snd_rawmidi snd_hda_core snd_seq_device mc snd_hwdep snd_pcm snd_timer
 5225.676992]  ecdh_generic snd rfkill soundcore thermal joydev gpio_amdpt gpio_generic tiny_power_button acpi_cpufreq button fuse configfs dmi_sysfs ip_tables x_tables hid_steam hid_generic usbhid amdgpu crct10dif_pclmul crc32_pclmul polyval_clmulni polyval_generic gf128mul ghash_clmulni_intel drm_ttm_helper ttm aesni_intel iommu_v2 gpu_sched xhci_pci xhci_pci_renesas drm_buddy xhci_hcd crypto_simd drm_display_helper cryptd nvme cec usbcore ccp nvme_core rc_core sp5100_tco wmi btrfs blake2b_generic libcrc32c crc32c_intel xor raid6_pq sg dm_multipath dm_mod scsi_dh_rdac scsi_dh_emc scsi_dh_alua msr efivarfs


Any help appreciated.

Are you sure it “turns off” (power down)?

Are you sure it is the login screen and not the unlock screen?

Do you do this from a desktop session? When yes, which desktop?

It does ‘turn off’. There is a power ‘click’ and all fans/leds (motherboard, gpu) goes off, and then instantly happens power ‘click’, fans starts to work and after a few secs I see password prompt.

You’re right, it is more like unlock screen. My session is preserved, I need to provide the password and I’m back to the same screen as before attempting suspend.

I do that from desktop session, Gnome / Wayland.

Thanks for the clarification. If you come back to the session as it was, it is no login, it is just the unlock, to make sure it is the same person as before that uses the session.

Using the correct wording is important for the understanding of others about the details of your problem.

Host erlangen has a 5600X too:

erlangen:~ # inxi -zSCMG 
System:
  Kernel: 5.18.11-1-default arch: x86_64 bits: 64 Console: pty pts/2 
    Distro: openSUSE Tumbleweed 20221204 
Machine:
  Type: Desktop System: Micro-Star product: MS-7C56 v: 2.0 serial: N/A 
  Mobo: Micro-Star model: B550-A PRO (MS-7C56) v: 2.0 serial: <filter> 
    UEFI: American Megatrends LLC. v: A.90 date: 03/17/2022 
CPU:
  Info: 6-core model: AMD Ryzen 5 5600X bits: 64 type: MT MCP cache:L2: 3 MiB 
  Speed (MHz):avg: 2200 min/max: 2200/3700 cores:1: 2200 2: 2200 3: 2200 
    4: 2200 5: 2200 6: 2200 7: 2200 8: 2200 9: 2200 10: 2200 11: 2200 12: 2200 
Graphics:
  Device-1: AMD Lexa PRO [Radeon 540/540X/550/550X / RX 540X/550/550X] 
    driver: amdgpu v: kernel 
  Display: x11 server: X.Org v: 21.1.4 with: Xwayland v: 22.1.5 driver:X:
    loaded: amdgpu unloaded: fbdev,modesetting,vesa dri: radeonsi gpu: amdgpu 
    resolution: 3840x2160~60Hz 
  API: OpenGL v: 4.6 Mesa 22.2.4 renderer: AMD Radeon RX 550 / 550 Series 
    (polaris12 LLVM 15.0.5 DRM 3.46 5.18.11-1-default) 
erlangen:~ #

Suspend to RAM works like a charm. I happened to test a HDMI cable exhibiting spurious errors on suspend/resume this morning:

erlangen:~ # journalctl --boot -6 --unit systemd-suspend.service -g 'starting|finished' 
Dec 06 06:47:10 erlangen systemd[1]: Starting System Suspend... 
Dec 06 06:48:12 erlangen systemd[1]: Finished System Suspend. 
Dec 06 08:50:49 erlangen systemd[1]: Starting System Suspend... 
Dec 06 08:51:48 erlangen systemd[1]: Finished System Suspend. 
Dec 06 08:54:19 erlangen systemd[1]: Starting System Suspend... 
Dec 06 08:55:20 erlangen systemd[1]: Finished System Suspend. 
Dec 06 08:58:47 erlangen systemd[1]: Starting System Suspend... 
Dec 06 08:59:56 erlangen systemd[1]: Finished System Suspend. 
Dec 06 09:01:44 erlangen systemd[1]: Starting System Suspend... 
Dec 06 09:04:55 erlangen systemd[1]: Finished System Suspend. 
Dec 06 09:20:28 erlangen systemd[1]: Starting System Suspend... 
Dec 06 09:27:03 erlangen systemd[1]: Finished System Suspend. 
Dec 06 09:30:21 erlangen systemd[1]: Starting System Suspend... 
Dec 06 09:30:41 erlangen systemd[1]: Finished System Suspend. 
Dec 06 09:42:55 erlangen systemd[1]: Starting System Suspend... 
Dec 06 09:47:00 erlangen systemd[1]: Finished System Suspend. 
Dec 06 09:47:41 erlangen systemd[1]: Starting System Suspend... 
Dec 06 09:48:34 erlangen systemd[1]: Finished System Suspend. 
erlangen:~ # 

You may want to tell more about your machine.

Sure, if anything else is required please let me know - my Linux-fu is pretty basic, though:


VFIEVOX3:~/> inxi -zSCMG 
System:
  Kernel: 6.0.10-1-default arch: x86_64 bits: 64 Desktop: GNOME v: 43.1
    Distro: openSUSE Tumbleweed 20221204
Machine:
  Type: Desktop Mobo: Gigabyte model: B550I AORUS PRO AX
    serial: <superuser required> UEFI: American Megatrends LLC. v: F14e
    date: 10/13/2021
CPU:
  Info: 6-core model: AMD Ryzen 5 5600X bits: 64 type: MT MCP cache: L2: 3 MiB
  Speed (MHz): avg: 3024 min/max: 2200/4650 cores: 1: 2953 2: 2200 3: 3579
    4: 2960 5: 3825 6: 2200 7: 2200 8: 3193 9: 3592 10: 2872 11: 3731 12: 2990
Graphics:
  Device-1: AMD Navi 21 [Radeon RX 6800/6800 XT / 6900 XT] driver: amdgpu
    v: kernel
  Display: wayland server: X.org v: 1.21.1.4 with: Xwayland v: 22.1.5
    compositor: gnome-shell driver: gpu: amdgpu resolution: 3840x1600~120Hz
  API: OpenGL v: 4.6 Mesa 22.2.4 renderer: AMD Radeon RX 6900 XT (navi21
    LLVM 15.0.5 DRM 3.48 6.0.10-1-default)

FIEVOX3:~> journalctl --boot -6 --unit systemd-suspend.service -g 'starting|finished' 
Hint: You are currently not seeing messages from other users and the system.
      Users in the 'systemd-journal' group can see all messages. Pass -q to
      turn off this notice.
-- No entries --


FIEVOX3:~> sudo journalctl --boot -6 --unit systemd-suspend.service -g 'starting|finished' 
[sudo] password for root: 
-- No entries --


Upload full output of “journalctl -b” to https://susepaste.org

here:
https://susepaste.org/97532992

This needs to be run as root, otherwise no system messages are shown.

sorry, here is updated version:
https://susepaste.org/37768971

A combination of Gigabyte Aorus B450 + openSUSE Tumbleweed never resumed from suspend to RAM. Other distributions suspended/resumed fine. I switched to a ASUSTeK PRIME B450-PLUS, which readily suspends and resumes. Might be a similar issue with your B550.

In this case, I think it actually resumes fine. The issue is, it wont stay suspended. The system seems to suspend and turn off fine, but immediately wakes up the moment after. So I guess there is some incorrect event waking it up, but this is kinda fresh installation (and on previous one it was behaving similar, but I wasn’t concerned as I was playing around) and nothing extraordinary connected (wired mouse + kb, speakers, usb DAC).

I have seen this (more or less) in other systems. Maybe look into the BIOS settings if reboot on Power On (or similar wording) is on. My explanation is that it switches off, sees the power, interprets that as the return of power after a power outage (for some reason) and powers on again.

Can we somehow see what event triggered waking up?
same system sleeps correctly on windows os.
I will check the BIOS settings but it is pretty standard as well.

If yes, it is not obvious.

Dec 06 10:03:56 localhost.localdomain kernel: smpboot: CPU 11 is now offline
Dec 06 10:03:56 localhost.localdomain kernel: ACPI: PM: Low-level resume complete

So resume immediately follows suspend (non-boot CPU are offlined on suspend).

Another likely culprit is WoL.

Try disabling for each network device using:

ethtool -s <device_name> wol d

I have the same problem, although it only started sometime in the last month or two; before that, suspend worked fine.

Why would Wake-on-LAN be an issue? Wouldn’t that require some other rogue device on the network to be continually spamming the network with WoL packets? According to tcpdump I have no such issue on my LAN (very unsurprisingly, as I don’t use WoL and no other device would even know that my desktop can be woken up in this way).