You are not logged in.
Pages: 1
I have a rtw_8822be WiFi interface. And since a couple of days my Wireless speeds are dropping to zero randomly. I tried to rfkill the device but that didn't help. Only way to bring it back is to reboot arch. Currently I have this setting enabled:
/etc/modprobe.d/rtw88_pci.conf
options rtw88_pci disable_aspm=1
options rtw88_core disable_lps_deep=1Last edited by coolpanda (2022-01-14 08:23:02)
Offline
Isn't this the same problem as your last thread?
Jin, Jîyan, Azadî
Offline
Isn't this the same problem as your last thread?
No. This time I am able to connect to WiFi, but not getting any speed on some occasions.
Offline
Well post your journal then, just like last time.
Jin, Jîyan, Azadî
Offline
Well post your journal then, just like last time.
Okay, can you tell me how can I do that using ix.io? I tried a lot of times but was unable to do so. I'm only able to upload what is being currently displayed on the screen. To display all the lines, I have to manually generate new lines using the down arrow button. I have tried saving the lines to a txt file using script file.txt but that is generating unwanted characters in the file.
Offline
Offline
I have tried these lines but it just doesn't output anything
sudo journalctl -b |& curl -F 'f:1=<-' ix.io
journalctl -b |& curl -F 'f:1=<-' ix.io
Last edited by coolpanda (2022-01-09 20:57:40)
Offline
Try https://archlinux.org/packages/communit … astebinit/ instead. Perhaps ix,io has blocked you for spamming them ![]()
The pager for journalctl can be disabled by passing the --no-pager option.
Jin, Jîyan, Azadî
Offline
Pastebinit is not working for me. But I got it using termbin. Here's the journal: https://termbin.com/o2vx
Edit: Here's another journal that I just generated when the Internet stopped working randomly (around 12:44). Then I switched to USB tethering to upload the journal to termbin (at 12:46)
Last edited by coolpanda (2022-01-10 07:28:04)
Offline
The wifi disconnects immediately after setting a MAC for the scan, try to disable https://wiki.archlinux.org/title/Networ … domization
If that doesn't help, consider a network managing service that does not scan the wifi all the time (or at least allows to configure that behavior)
Online
So I disabled MAC address randomization and everything was working fine for a couple of days, but now suddenly the issue reappeared. And the interesting thing is this time the USB wired tethering is also not working and rebooting isn't helping either. So the WiFi appears to be connected to my SSID but there is no internet whatsoever. Here's the log around the time of the issue
Last edited by coolpanda (2022-01-13 08:09:26)
Offline
Don't copy out of the pager…
Jan 13 13:07:52 archlinux NetworkManager[389]: <info> [1642059472.2376] device (usb0): interface index 3 renamed iface from 'usb0' to 'enp0s20f0u3' Jan 13 13:08:57 archlinux NetworkManager[389]: <info> [1642059537.3510] device (usb0): interface index 4 renamed iface from 'usb0' to 'enp0s20f0u3' Jan 13 13:10:18 archlinux NetworkManager[389]: <info> [1642059618.3964] device (usb0): interface index 5 renamed iface from 'usb0' to 'enp0s20f0u3'Either you re-plugged the tethered device or it signs out for powersaving/autosuspend reasons or the physical connection (cable) is wonky.
Interim you restarted the wifi and disconnected it after 45 seconds after plugging the thether device ~30secs after the connection (only recorded signal is good)
Jan 13 13:09:32 archlinux NetworkManager[389]: <info> [1642059572.4080] device (wlo1): state change: deactivating -> disconnected (reason 'user-requested'> … Jan 13 13:09:42 archlinux NetworkManager[389]: <info> [1642059582.3750] device (wlo1): Activation: starting connection 'HomeWiFi' (b01e0040-ff06-413b-976f> … Jan 13 13:09:43 archlinux wpa_supplicant[653]: wlo1: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-40 noise=9999 txrate=270000 … Jan 13 13:09:44 archlinux NetworkManager[389]: <info> [1642059584.0001] manager: NetworkManager state is now CONNECTED_GLOBAL … Jan 13 13:10:17 archlinux kernel: usb 1-3: USB disconnect, device number 9 … Jan 13 13:10:31 archlinux NetworkManager[389]: <info> [1642059631.9023] device (wlo1): state change: activated -> deactivating (reason 'user-requested', s>This are not the original symptoms.
What metric do you use for "my Wireless speeds"? Syncthing (only)?
Can you at those times still ping "google.com"? "8.8.8.8"? The "_gateway"?
There's
kernel: rtw_8822be 0000:02:00.0: firmware failed to leave lps state and that matches https://github.com/lwfinger/rtw88/issues/53 and https://forum.manjaro.org/t/wifi-drops- … orks/55937
The latter matches your symptoms, both speculate on aspm being the cause and the manjaro thread claims "pcie_aspm.policy=performace" to fix it.
Online
Don't copy out of the pager…
Didn't have any other choice as termbin requires internet to work.
What metric do you use for "my Wireless speeds"? Syncthing (only)?
By that I mean I can't connect to any websites (just keeps loading, no error code either)
Can you at those times still ping "google.com"? "8.8.8.8"? The "_gateway"?
Yes, I tried pinging "google.com" and the output was something starting from 64 bytes... which I think it was able to successfully ping "google.com"
There's
kernel: rtw_8822be 0000:02:00.0: firmware failed to leave lps stateand that matches https://github.com/lwfinger/rtw88/issues/53 and https://forum.manjaro.org/t/wifi-drops- … orks/55937
The latter matches your symptoms, both speculate on aspm being the cause and the manjaro thread claims "pcie_aspm.policy=performace" to fix it.
# /sys/module/pcie_aspm/parameters/policy
[default] performance powersave powersupersaveFrom above, I think it should be "pcie_aspm.policy=performance" and not "pcie_aspm.policy=performace". The person from manjaro thread maybe just mistyped it. Anyways, This seems to have fixed it for me too. (also, after applying this fix, rebooting, "un-applying" it, doing sudo update-grub, rebooting again, all this makes it work just fine. So I have no clue to how all of this is working. But again I think it should stop working again anyways after some time.)
But don't know it's (pcie_aspm.policy=performance) implications on power savings on my system. I can use any type of power saving at this time as my battery is even below 50% of the original charge capacity.
Last edited by coolpanda (2022-01-13 10:09:34)
Offline
redirect the output into a file and upload that (or cnp from an editor) or avoid the pager as HoaS told you in #8
Do not copy out of the pager - it can make the journal borderline useless.
The person from manjaro thread maybe just mistyped it.
Yup. And the person blindly copy and failed it ;-)
Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.
There's gonna be *some* battery impact because the policy affects the entire bus and every device attached (how much they make use of that is a different question)
And yes: "unapplying" the parameter will provoke the same error again.
What's the output of
modinfo rtw_8822beOnline
redirect the output into a file
Man, this was so easy. sudo journalctl -b > output.txt. I was figuring this out for so long ![]()
What's the output of modinfo rtw_8822be
modinfo: ERROR: Module rtw_8822be not found.
Offline
And you already have
options rtw88_pci disable_aspm=1
lsmod | grep rtwwill likely say that you're using rtw88_pci so this is moot unles the parameter didn't apply
systool -vm rtw88_pciOnline
$ lsmod | grep rtw
rtw88_8822be 16384 0
rtw88_8822b 229376 1 rtw88_8822be
rtw88_pci 36864 1 rtw88_8822be
rtw88_core 258048 2 rtw88_pci,rtw88_8822b
mac80211 1232896 2 rtw88_pci,rtw88_core
cfg80211 1077248 2 rtw88_core,mac80211$ systool -vm rtw88_pci
Module = "rtw88_pci"
Attributes:
coresize = "36864"
initsize = "0"
initstate = "live"
refcnt = "1"
srcversion = "A3BC4EA4BE60B47A855B228"
taint = ""
uevent = <store method only>
Parameters:
disable_aspm = "Y"
disable_msi = "N"
Sections:disable_aspm is set to "Y" so it is applied. But on the other hand, setting pcie_aspm.policy to performance also worked for me which shouldn't have?
Last edited by coolpanda (2022-01-13 14:58:57)
Offline
It prevents aspm on the bus - seems you'll have to go w/ that ![]()
Online
Pages: 1