I re-tested the previous update, hoping that the update from 08-01-2021 somehow fixed the existing serious problems.
The error log is not as bad as before and existing errors are negligible to a point, and shouldn’t mess with the system usage.
However, I have still the same serious issue with LIMITED CONNECTION. Wifi icon is showed with exclamation mark, megasync is syncing, but sites aren’t loading, no matter of what connection I would use.
If I don’t deal with it, I won’t be able to update Manjaro anymore and this would force me to use different distro, so this is a very serious issue for me.
Again, this no-full-Internet problem is tied to the latest updates. Restoring system from backup resolves the issue. Here is the info I gathered so far:
the issue exists on all users so it’s not a matter of user configs
there was no errors or warnings during update
the issue exists on kernels 5.9 and 5.4, so it’s not a kernel problem
I downgraded linux-firmware to version from December and the issue persists, so it’s not a network card driver problem
I’m trying to downgrade network-manager, so far I can’t find any older version in backups, I only downgraded network-manager-qt, but I suspect it’s not enough, any idea how to find older versions of manjaro packages to download?
I assume you’ve already tried to delete your WiFi config and create it again? (after update)
I’d suggest to make another thread to check further on your issue.
No. I assume that since the default test user also has the problem, it’s not a matter of configs. However, maybe the old config should be updated with pacnew, if there is one.
Now that I know which package is at fault, I can go back and forth to test it. I had to find the networkmanager stable version on the Manjaro site, because I didn’t have it in pacman cache. Now, I can easily downgrade again if I hit a dead end or lack of time.
I’ll post my findings here, if there will be any.
Ah, right. Still, all connections had this issue so this is rather not a problem of connection configs. It would be rather odd to have the same config issue in all stored connections. Ah, I created a NEW CONNECTION with my new phone and the issue was the same.
NetworkManager.conf is basically empty:
# Configuration file for NetworkManager.
# See "man 5 NetworkManager.conf" for details.
I’ll check nmcli statuses and maybe I’ll find out something.
Here is how it looks now, when everything is OK (downgraded networkmanager):
michaldybczak alienware-PC ~ nmcli
wlp60s0: connected do FunBox-4BA3
"Qualcomm Atheros QCA6174"
wifi (ath10k_pci), ******************* hardware, MTU 1500
default IPv4
inet4 192.168.1.18/24
route4 0.0.0.0/0
route4 192.168.1.0/24
inet6 ****************************
route6 fe80::/64
route6 ff00::/8
p2p-dev-wlp60s0: disconnected
"p2p-dev-wlp60s0"
wifi-p2p, hardware
enp59s0: not available
"Qualcomm Atheros Killer E2400"
ethernet (alx), ******************** sprzęt, MTU 1500
lo: not managed
"lo"
loopback (unknown), 00:00:00:00:00:00, hardware, MTU 65536
DNS configuration:
servers: 192.168.1.1
domains: home
interface: wlp60s0
*** signs are to cover some personal, detailed data.
EDIT: I updated networkmanager and the issue appears, like with a switch. Looking on devices and connections, all looks correct, no issues, errors, warnings. It looks all the same as with the downgraded version. No idea what to do further.
This seems to disable the quiet kernel boot options set in the grub menu entries for the several kernel versions.
At least I now get the usual kernel messages on boot and shutdown.
Pulseaudio 14.2 fixed audio issues on of my laptops. My Acer V3-575G still has audio issues, however I am not sure whether these are pulsaudio, kernel or hardware related. It is impossible to get the audio to run, when I am using the kernel 5.10.7-3-MANJARO. With both 5.11 rc3 and 5.4 LTS the audio works fine.
[ 10.182361] snd_hda_intel 0000:00:1f.3: azx_get_response timeout, switching to polling mode: last cmd=0x00170500
[ 11.192532] snd_hda_intel 0000:00:1f.3: No response from codec, disabling MSI: last cmd=0x00170500
[ 12.195869] snd_hda_intel 0000:00:1f.3: azx_get_response timeout, switching to single_cmd mode: last cmd=0x00170500
[ 12.196025] azx_single_wait_for_response: 17 callbacks suppressed
[ 12.713050] snd_hda_codec_realtek hdaudioC0D0: Unable to sync register 0x2b8000. -5
[ 12.713518] snd_hda_codec_realtek hdaudioC0D0: Unable to sync register 0x2b8000. -5
[ 13.049084] snd_hda_codec_hdmi hdaudioC0D2: Unable to sync register 0x2f0d00. -5
[ 17.126092] nouveau 0000:01:00.0: Enabling HDA controller
[ 17.229468] nouveau 0000:01:00.0: bus: MMIO read of 00000000 FAULT at 6013d4 [ IBUS ]
[ 17.236338] rfkill: input handler enabled
[ 18.935766] azx_single_send_cmd: 491 callbacks suppressed
[ 19.211791] snd_hda_codec_hdmi hdaudioC0D2: Unable to sync register 0x2f0d00. -5
These errors are well known to me. They started appearing with kernel 5.4 and on some distros (including Manjaro) I was able to get rid of them (and get the audio to run) by setting TLP setting RUNTIME_PM_ON_AC and RUNTIME_PM_ON_BAT to ON.
I do not know why the issue appeared again, after the last testing update on the 14th of January. Until then I have been using kernel 5.10. without issues.
EDIT:
Disabling power saving on the audio driver fixes the issue:
sudo tee /etc/modprobe.d/snd-hda-intel.conf <<<'options snd-hda-intel power_save=0'
before the fix
