[Solved] System gets borked after suspending

the_gremlin · 2022-08-12 17:37:40

I have a hp 15s eq-2021nv laptop, and whenever I cause it to S3 suspend, be it through systemctl suspend or though closing the laptop lid, whenever I resume usage, I find that I cannot open open programs, when using commands such as ls or cd, i get bash: <command>: command not found and whenever i try to use programs like cargo, htop, firefox i get an input/output error. This can only be fixed by force-shutting down the laptop by holding down the power button and restatring it.

I would appreciate if anyone could help me solve this problem as a laptop that breaks whenever the lid closes is pretty useless.

Last edited by the_gremlin (2022-08-14 08:50:07)

the_gremlin · 2022-08-12 17:56:57

One interesting thing I noticed is that in journalctl, there were never any messages about the laptop being "un-suspended", i don't know if this is how it actually works, but i would expect to have some message about things reactivating
for example:

Aug 12 20:35:38 menthol audit[2228]: CRED_DISP pid=2228 uid=1000 auid=1000 ses=1 msg='op=PAM:setcred grantors=pam_faillock,pam_permit,pam_faillock acct="root" exe="/usr/bin/>
Aug 12 20:35:38 menthol kernel: audit: type=1106 audit(1660325738.413:79): pid=2228 uid=1000 auid=1000 ses=1 msg='op=PAM:session_close grantors=pam_systemd_home,pam_limits,p>
Aug 12 20:35:38 menthol kernel: audit: type=1104 audit(1660325738.414:80): pid=2228 uid=1000 auid=1000 ses=1 msg='op=PAM:setcred grantors=pam_faillock,pam_permit,pam_failloc>
Aug 12 20:38:50 menthol polkitd[477]: Registered Authentication Agent for unix-process:2280:113129 (system bus name :1.68 [/usr/bin/pkttyagent --notify-fd 5 --fallback], obj>
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.3706] manager: sleep: sleep requested (sleeping: no  enabled: yes)
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.3708] manager: NetworkManager state is now ASLEEP
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.3709] device (wlo1): state change: activated -> deactivating (reason 'sleeping', sys-iface-state: 'managed')
Aug 12 20:38:50 menthol polkitd[477]: Unregistered Authentication Agent for unix-process:2280:113129 (system bus name :1.68, object path /org/freedesktop/PolicyKit1/Authenti>
Aug 12 20:38:50 menthol dbus-daemon[348]: [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' req>
Aug 12 20:38:50 menthol systemd[1]: Starting Network Manager Script Dispatcher Service...
Aug 12 20:38:50 menthol dbus-daemon[348]: [system] Successfully activated service 'org.freedesktop.nm_dispatcher'
Aug 12 20:38:50 menthol systemd[1]: Started Network Manager Script Dispatcher Service.
Aug 12 20:38:50 menthol audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=NetworkManager-dispatcher comm="systemd" exe="/usr/lib/systemd/systemd" >
Aug 12 20:38:50 menthol kernel: audit: type=1130 audit(1660325930.379:81): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=NetworkManager-dispatcher comm="systemd" exe=>
Aug 12 20:38:50 menthol kernel: wlo1: deauthenticating from c0:94:ad:9e:a5:a3 by local choice (Reason: 3=DEAUTH_LEAVING)
Aug 12 20:38:50 menthol wpa_supplicant[426]: wlo1: CTRL-EVENT-DISCONNECTED bssid=c0:94:ad:9e:a5:a3 reason=3 locally_generated=1
Aug 12 20:38:50 menthol wpa_supplicant[426]: wlo1: CTRL-EVENT-DSCP-POLICY clear_all
Aug 12 20:38:50 menthol wpa_supplicant[426]: wlo1: CTRL-EVENT-REGDOM-CHANGE init=CORE type=WORLD
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4506] device (wlo1): supplicant interface state: completed -> disconnected
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4509] device (wlo1): state change: deactivating -> disconnected (reason 'sleeping', sys-iface-state: 'manage>
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4621] dhcp4 (wlo1): canceled DHCP transaction
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4622] dhcp4 (wlo1): activation: beginning transaction (timeout in 45 seconds)
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4623] dhcp4 (wlo1): state changed no lease
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.4640] device (wlo1): set-hw-addr: set MAC address to 72:A7:8B:90:2A:3B (scanning)
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.6285] device (wlo1): supplicant interface state: disconnected -> interface_disabled
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.6288] device (wlo1): state change: disconnected -> unmanaged (reason 'sleeping', sys-iface-state: 'managed')
Aug 12 20:38:50 menthol NetworkManager[350]: <info>  [1660325930.6323] device (wlo1): set-hw-addr: reset MAC address to B4:B5:B6:79:E4:E7 (unmanage)
Aug 12 20:38:50 menthol systemd[1]: Reached target Sleep.
Aug 12 20:38:50 menthol systemd[1]: Starting System Suspend...
Aug 12 20:38:50 menthol wpa_supplicant[426]: wlo1: CTRL-EVENT-DSCP-POLICY clear_all
Aug 12 20:38:50 menthol systemd-sleep[2291]: Entering sleep state 'suspend'...
Aug 12 20:38:50 menthol kernel: PM: suspend entry (s2idle)
-- Boot 74cbdc4cf08249f1846267ea6800daea --
Aug 12 20:40:25 menthol kernel: Linux version 5.18.16-zen1-1-zen (linux-zen@archlinux) (gcc (GCC) 12.1.1 20220730, GNU ld (GNU Binutils) 2.38) #1 ZEN SMP PREEMPT_DYNAMIC Wed>
<normal boot process follows>

I ran systemct suspend at around 20:38:50, and you can see all the noise about NetworkManager and everything else suspending, however the logs don't contain anything about the system waking up from being suspended, even though i did wake it up, they just skip straight to the next boot i did.

seth · 2022-08-12 19:56:32

Seems like the root device (HDD/SSD) doesn't wake up (and so the system journal cannot be synced to disk, so it doesn't survive a reboot)

Run "demsg -w" in a terminal, send the system to sleep and see whether dmesg prints any errors on resume.
You can also/additionally run "sudo journalctl -f", though it's likely the same error messages (iff any) for both logs.

the_gremlin · 2022-08-12 22:20:28

Both the output by dmesg and journal ctl was spammed by c2h class 1 function 2 not support multiple times, but other than that that they both only had stuff relating to networking systems going to sleep, journalctl with NetworkManager and dmesg with wlo1 deauthing

Last edited by the_gremlin (2022-08-12 22:25:28)

the_gremlin · 2022-08-12 22:27:04

I also noticed that the hard drive LED on my laptop was going nuts after waking up so this looks more and more like a problem with the disk itself, since nothing is showing up on the logs

Edit: This same problem also happens when waking up from s2idle, and while at the time of writing this i havent looked at dmesg for this specific case yet, im positive the output is going to be the same as with S3 suspend. (Which it is)

Last edited by the_gremlin (2022-08-12 22:58:36)

ewaller · 2022-08-12 23:26:51

the_gremlin wrote:

I also noticed that the hard drive LED on my laptop was going nuts after waking up so this looks more and more like a problem with the disk itself, since nothing is showing up on the logs
Edit: This same problem also happens when waking up from s2idle, and while at the time of writing this i havent looked at dmesg for this specific case yet, im positive the output is going to be the same as with S3 suspend. (Which it is)

Maybe https://bbs.archlinux.org/viewtopic.php … 1#p2049571 ?

the_gremlin · 2022-08-12 23:35:13

ewaller wrote:

the_gremlin wrote:
I also noticed that the hard drive LED on my laptop was going nuts after waking up so this looks more and more like a problem with the disk itself, since nothing is showing up on the logs
Edit: This same problem also happens when waking up from s2idle, and while at the time of writing this i havent looked at dmesg for this specific case yet, im positive the output is going to be the same as with S3 suspend. (Which it is)
Maybe https://bbs.archlinux.org/viewtopic.php … 1#p2049571 ?

This looks like it has to do with Intel CPU's, however I have an AMD chip, a Ryzen 3 5000 series, which i forgot to mention, so this unfortunately isn't applicable here.

Last edited by the_gremlin (2022-08-12 23:36:02)

seth · 2022-08-13 05:42:55

c2h class 1 function 2 not support

Is that verbatim?
Also there should at least be some IO errors in dmesg when you try to access the disk (ie. run firefox) because we know that this causes an IO error.

According to https://support.hp.com/id-en/document/c07953649 this is an nvme?
Which one (check the SMART data) and is this maybe https://wiki.archlinux.org/title/Solid_ … ST_support ?

the_gremlin · 2022-08-13 11:59:48

seth wrote:

c2h class 1 function 2 not support
Is that verbatim?

Yes, although i don't think its related as i didn't see it today, and dmesg displays no io errors from running dmesg -w and writing to a log, the only messages are network-related.

The drive is a kioxia kgb 40nzv nvme, and changing the specified kernel parameter to 0 didn't have any effect, though i dont know how any other values might help me.

Last edited by the_gremlin (2022-08-13 12:00:06)

seth · 2022-08-13 19:03:28

Are you sure you effectively added the parameter?

cat /proc/cmdline
systool -vm nvme_core

Also try to pass "pcie_aspm=off"

The drive is a kioxia kgb 40nzv nvme

https://www.google.com/search?q=%22kioxia+kgb%22 returns zero hits, are you sure that's the accurate name?

Top be sure: you're not getting any errors in "dmesg -w" *while* running firefox produces IO errors??

the_gremlin · 2022-08-13 23:49:28

I made an incredibly stupid blunder. I was writing the dmesg output to a file so I could have it handy, but the problem is that the drive suddenly changes to read-only, so nothing gets written to said file.

Now that i am outputting dmesg directly to terminal, Im getting a bunch of errors:

ACPI: EC: interrupt blocked
ACPI: button: The lid device is not compliant to SW_LID
ACPI: EC: interrupt unblocked
pci 0000:00:00.2: can't derive routing for PCI INT A
pci 0000:00:00.2: PCI INT A:no GSI
<amdgpu and other driver output omited>
nvme nvvme0: Device not ready; aborting reset, CSTS=0x3
nvme nvme0 : Removing after probe failure status: -19
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 2, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 2, rd, 0, flush 0, corrupt, gen0
nvme0n1: detected capcity change from 500118192 to 0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 3, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 4, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 5, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 6, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 7, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 8, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 9, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3): bbdev /dev/nvme0n1p3 errs: wr 10, rd, 0, flush 0, corrupt, gen0
BTRFS error (device nvme0n1p3) in btrfs_commit_transaction:2418: errno=-5 IO failure (Error while writing out transaction)
BTRFS info (device nvme0n1p3: state E): forced readonly
BTRFS warning (device nvme0n1p3: state E): Skipping commit of aborted transaction
BTRFS error (device nvme0n1p3: state EA): in cleanup_transaction:1982: errno=-5 IO failure
< a lot of errors about journald failing to write stuff because the disk is readonly, omitted>

Its really frustrating how i didn't think about this, but then again im just glad i caught it.

(i wrote this by hand cause so any apparent errors are mistakes i made)

Last edited by the_gremlin (2022-08-14 00:01:45)

seth · 2022-08-14 06:11:34

Archimedes and his bath tub, Newton napping under an apple tree… at least anecdotally "I discovered that by sheer luck" is no cause for frustration.

The output makes far more sense and you're perfectly matching https://wiki.archlinux.org/title/Solid_ … nd_support

the_gremlin · 2022-08-14 08:49:17

seth wrote:

Archimedes and his bath tub, Newton napping under an apple tree… at least anecdotally "I discovered that by sheer luck" is no cause for frustration.
The output makes far more sense and you're perfectly matching https://wiki.archlinux.org/title/Solid_ … nd_support

Yeah this fixes it. Thanks a bunch!

Arch Linux

#1 2022-08-12 17:37:40

[Solved] System gets borked after suspending

#2 2022-08-12 17:56:57

Re: [Solved] System gets borked after suspending

#3 2022-08-12 19:56:32

Re: [Solved] System gets borked after suspending

#4 2022-08-12 22:20:28

Re: [Solved] System gets borked after suspending

#5 2022-08-12 22:27:04

Re: [Solved] System gets borked after suspending

#6 2022-08-12 23:26:51

Re: [Solved] System gets borked after suspending

#7 2022-08-12 23:35:13

Re: [Solved] System gets borked after suspending

#8 2022-08-13 05:42:55

Re: [Solved] System gets borked after suspending

#9 2022-08-13 11:59:48

Re: [Solved] System gets borked after suspending

#10 2022-08-13 19:03:28

Re: [Solved] System gets borked after suspending

#11 2022-08-13 23:49:28

Re: [Solved] System gets borked after suspending

#12 2022-08-14 06:11:34

Re: [Solved] System gets borked after suspending

#13 2022-08-14 08:49:17

Re: [Solved] System gets borked after suspending

Board footer