Wifi clients disconnect after update to 3.11.1


#1

Hello,
from last update to 3.11.1 I have some troubles with WiFi. Clients on wlan can’t connect to WiFi after some time. I have to reboot my Omnia. Then it works again.
I found this in /var/log/messages:

 2019-01-10 06:29:06 info kernel[]: [   14.610873] ip_gre: GRE over IPv4 tunneling driver
2019-01-10 06:29:06 info kernel[]: [   14.629887] Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
2019-01-10 06:29:06 info kernel[]: [   14.638562] PPP generic driver version 2.4.2
2019-01-10 06:29:06 info kernel[]: [   14.646152] ip6_tables: (C) 2000-2006 Netfilter Core Team
2019-01-10 06:29:06 info kernel[]: [   14.659067] nf_conntrack version 0.5.0 (16384 buckets, 65536 max)
2019-01-10 06:29:06 info kernel[]: [   15.352245] cryptodev: driver 1.8 loaded.
2019-01-10 06:29:06 info kernel[]: [   16.622352] Loading modules backported from Linux version wt-2017-01-31-0-ge882dff19e7f
2019-01-10 06:29:06 info kernel[]: [   16.630378] Backport generated by backports.git backports-20160324-13-g24da7d3c
2019-01-10 06:29:06 info kernel[]: [   17.052384] pci 0000:00:02.0: enabling device (0140 -> 0142)
2019-01-10 06:29:06 info kernel[]: [   17.058263] ath10k_pci 0000:02:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
2019-01-10 06:29:06 warning kernel[]: [   17.234532] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/pre-cal-pci-0000:02:00.0.bin failed with error -2
2019-01-10 06:29:06 warning kernel[]: [   17.245269] ath10k_pci 0000:02:00.0: Falling back to user helper
2019-01-10 06:29:06 err kernel[]: [   17.255437] firmware ath10k!pre-cal-pci-0000:02:00.0.bin: firmware_loading_store: map pages failed
2019-01-10 06:29:06 warning kernel[]: [   17.264538] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/cal-pci-0000:02:00.0.bin failed with error -2
2019-01-10 06:29:06 warning kernel[]: [   17.274922] ath10k_pci 0000:02:00.0: Falling back to user helper
2019-01-10 06:29:06 err kernel[]: [   17.285106] firmware ath10k!cal-pci-0000:02:00.0.bin: firmware_loading_store: map pages failed
2019-01-10 06:29:06 info kernel[]: [   17.302466] ath10k_pci 0000:02:00.0: qca988x hw2.0 target 0x4100016c chip_id 0x043202ff sub 0000:0000
2019-01-10 06:29:06 info kernel[]: [   17.311733] ath10k_pci 0000:02:00.0: kconfig debug 0 debugfs 1 tracing 0 dfs 1 testmode 1
2019-01-10 06:29:06 info kernel[]: [   17.321345] ath10k_pci 0000:02:00.0: firmware ver 10.2.4-1.0-00041 api 5 features no-p2p,raw-mode,mfp,allows-mesh-bcast crc32 f43fa422
2019-01-10 06:29:06 warning kernel[]: [   17.366081] ath10k_pci 0000:02:00.0: board id is not exist in otp, ignore it
2019-01-10 06:29:06 warning kernel[]: [   17.373198] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/QCA988X/hw2.0/board-2.bin failed with error -2
2019-01-10 06:29:06 warning kernel[]: [   17.383662] ath10k_pci 0000:02:00.0: Falling back to user helper
2019-01-10 06:29:06 err kernel[]: [   17.393762] firmware ath10k!QCA988X!hw2.0!board-2.bin: firmware_loading_store: map pages failed
2019-01-10 06:29:06 info kernel[]: [   17.403468] ath10k_pci 0000:02:00.0: board_file api 1 bmi_id N/A crc32 bebc7c08
2019-01-10 06:29:06 info kernel[]: [   18.539255] ath10k_pci 0000:02:00.0: htt-ver 2.1 wmi-op 5 htt-op 2 cal otp max-sta 128 raw 0 hwcrypto 1
2019-01-10 06:29:06 debug kernel[]: [   18.631545] ath: EEPROM regdomain: 0x0
2019-01-10 06:29:06 debug kernel[]: [   18.631550] ath: EEPROM indicates default country code should be used
2019-01-10 06:29:06 debug kernel[]: [   18.631553] ath: doing EEPROM country->regdmn map search
2019-01-10 06:29:06 debug kernel[]: [   18.631558] ath: country maps to regdmn code: 0x3a
2019-01-10 06:29:06 debug kernel[]: [   18.631561] ath: Country alpha2 being used: US
2019-01-10 06:29:06 debug kernel[]: [   18.631563] ath: Regpair used: 0x3a
2019-01-10 06:29:06 info kernel[]: [   18.887014] Linux video capture interface: v2.00

Can you look at it, please?

Thanks


#2

I expect it’s the same as in this CZ thread: Padající Wi-Fi připojení


#3

Tak to si Michelle pocte…;-\


#4

I had checked she had posted in Czech already :wink:


#5

Hello,
We cannot be sure, if this is the same problem or not. From the mentioned thread only one guy sent email to Support. However, we don’t have diagnostics, so it’s really difficult to tell what’s happening as we haven’t touch default Wi-Fi drivers in releases 3.11 and 3.11.1.

From the OP, I cannot see any issue what will cause disconnecting clients and if it happens on 2.4 GHz or 5 GHz. The output is from ath10k, which meant that it couldn’t find the firmware files and fallbacking to other firmware files and in the end it found the firmware.


#6

It happened again. I can see new message in /var/log/messages:

2019-01-11 15:43:51 warning ddns-scripts[18017]: myddns_ipv4: Get registered/public IP for mydomain.cz failed - retry 145/0 in 60 seconds

I’m using myddns for change A record for my domain.

and I can have full log of this message:

2019-01-11 15:51:28 warning dnsmasq-dhcp[3727]: DHCP packet received on vethMJYJ00 which has no address
2019-01-11 15:51:45 warning dnsmasq-dhcp[3727]: DHCP packet received on vethMJYJ00 which has no address

One note: I’m also using squid for some PC’s. And all PC’s which use squid as proxy work! Currently I’m not in the place of the Omnia.


Logs spam "DHCP packet received on eth2 which has no address"
#7

This seems more of a routing issue and perhaps only WiFi related if vethMJYJ00 is enslaved with the same bridge as the WiFi ifaces in question.

Does vethMJYJ00 belong to squid or some other daemon? Those clients with connectivity working through squid are connecting over LAN or WLAN?


and nit-picking - are the WLAN clients being really disconnected from the AP or rather remaining connected to the AP but traffic having ceased? If latter it might be another symptom of a routing issue.


#8

I know it is weird but I solved this issue by enabling DNS forwading (and TLS Cloudflare). So it was some kind of DNS issue I suppose…