Failed to flush queue
Hi, after installing this driver, everything looked working well, but after few hours, the wifi disconnected and dmesg is full of these errors:
ieee80211 phy0: rt2x00queue_write_tx_frame: Error - Dropping frame due to full tx queue 0 ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 0 failed to flush ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 2 failed to flush
And since the first lost connection I can't get it to connect again. Restarting system works just for few minutes, then it gets again into these errors. I don't even know where to look for further information about this.
Your kernel version..? And also the output of these commands:
lspci -vv |grep -i -A 9 wireless
lsusb |grep -i mediatek
4.2.8 (-300.fc23)
lspci -vv |grep -i -A 9 wireless
03:00.0 Network controller: MEDIATEK Corp. MT7630e 802.11bgn Wireless Network Adapter
Subsystem: Foxconn International, Inc. Device e074
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 17
Region 0: Memory at f7900000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [40] Power Management version 3
Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
pcilib: sysfs_read_vpd: read failed: Connection timed out
lsusb |grep -i mediatek
Bus 003 Device 005: ID 0e8d:763f MediaTek Inc.
Confirming this bug causes connection failure and the inability to reconnect(or send any tx frames probably) until the mt7630e module is removed and inserted again. I found that this suppresses the problem for a while:
nmcli radio wifi off
rmmod mt7630e
modprobe mt7630e
nmcli radio wifi on
dmesg:
[16755.465432] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
[16755.625434] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
[16755.661500] ieee80211 phy0: rt2x00queue_write_tx_frame: Error - Dropping frame due to full tx queue 0
[16756.845361] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 0 failed to flush
uname -srv:
Linux 4.4.0-kali1-amd64 #1 SMP Debian 4.4.2-3kali1 (2016-02-23)
lspci -vv |grep -i -A 9 wireless:
03:00.0 Network controller: MEDIATEK Corp. MT7630e 802.11bgn Wireless Network Adapter
Subsystem: Foxconn International, Inc. MT7630e 802.11bgn Wireless Network Adapter
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 17
Region 0: Memory at f7900000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [40] Power Management version 3
Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
pcilib: sysfs_read_vpd: read failed: Connection timed out
lsusb | grep -i mediatek:
Bus 001 Device 005: ID 0e8d:763f MediaTek Inc.
I installed the latest 2.0.3a-1 github version using dkms.
@J08nY Thanks for testing it. Though I don't recommend using the alpha branches for normal uses, those are only for testing. For normal uses, always use the release branch.:D
I used the release branch, same bug, I only installed the alpha version to try to fix it.
It appears that the frequency of this issue is somehow related to signal strength, If I have a really good signal I might go hours without being disconnected, however if I have poor signal it might be minutes.
I also have this error many times, but my wifi never disconnect. A workaround could be try to disable the power managment: sudo iwconfig wlan0 power off
where wlan0 is the name of your wifi (check with ifconfig or ip link)
It appears that the frequency of this issue is somehow related to signal strength, If I have a really good signal I might go hours without being disconnected, however if I have poor signal it might be minutes.
@J08nY I wouldn't say so, I have my laptop like 3 metres far from a wifi router, so the signal strength is really high, but the connection is lost anyway very quickly.
I also have this error many times, but my wifi never disconnect. A workaround could be try to disable the power managment: sudo iwconfig wlan0 power off
@echo-devim Tried this with no success at all. Reloading the module mt7630e on the other hand happens to work a bit more.
Also, since my first comment, I've upgraded kernel to 4.4.4-301.fc23 and tried both 2.0.3a-1 and master branches.
When I look into log using journalctl on the line when NetworkManager disconnects, it says:
<warn> Connection disconnected (reason -4)
According to NetworkManager.h this means NM_DEVICE_STATE_REASON_CONFIG_FAILED
Then it tries to reconnect:
<warn> (wlp3s0f0): link timed out.
<info> (wlp3s0f0): device state change: activated -> failed (reason 'ssid-not-found') [100 120 53]
I get the same issue. I have tried both master and 2.0.3a-1, on both Debian and Ubuntu. Below are the details for my Ubuntu laptop.
sudo lspci -vv |grep -i -A 9 wireless:
03:00.0 Network controller: MEDIATEK Corp. MT7630e 802.11bgn Wireless Network Adapter
DeviceName: WLAN
Subsystem: Hewlett-Packard Company MT7630e 802.11bgn Wireless Network Adapter
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 18
Region 0: Memory at d0600000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [40] Power Management version 3
Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
lsusb |grep -i mediatek:
Bus 003 Device 019: ID 0e8d:763f MediaTek Inc.
uname -a:
Linux hp 4.4.0-22-generic #40-Ubuntu SMP Thu May 12 22:03:46 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
Still getting the same issue with v2.0.6.
Cheers, Simon
I also have roughly this problem, only at office but not home. I can get it working by turning off wifi, and using rmmod/modprobe, but if I try that two times it brings down the system.
Yep, same for me. I have a script that disconnects using NetworkManager, then turns off the wifi, then rmmod -> sleep 1 -> modprobe and turns the wifi back on. However this seems to freeze up the whole system once every few tries. It freezes on/after modprobe mt7630e.
2016-09-21 14:47 GMT+02:00 John Vandenberg [email protected]:
I also have roughly this problem, only at office but not home. I can get it working by turning off wifi, and using rmmod/modprobe, but if I try that two times it brings down the system.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/neurobin/MT7630E/issues/11#issuecomment-248601777, or mute the thread https://github.com/notifications/unsubscribe-auth/ABPJVj1UHiGbWH_2AFOK1P-FtMWnVTmmks5qsSd0gaJpZM4G8hOC .
Just upgraded kernel to 4.7.0 and driver to 2.0.8 (commit d5023e72da59e02c8279d7ccd5f8c430f8163b03) and I seem to get far less errors. It might be just freak coincidence since I had the driver error after upgrading but currently been on wifi for hours and no error, no disconnect (note that this is indeed very unusual, before upgrade I could get maybe half an hour of basic usage until disconnect). I recommend updating (both kernel and driver) and trying it out.
@J08nY , is it still working well for you?
@jayvdb Hmm, tested it some more and seems I was just lucky that day. Issue is not solved on 4.7.0 kernel and 2.0.8 driver.
I have the same problem,
uname -r
Kernel: 4.4.0-59-generic
sudo lspci -vv |grep -i -A 9 wireless:
03:00.0 Network controller: MEDIATEK Corp. MT7630e 802.11bgn Wireless Network Adapter Subsystem: Foxconn International, Inc. MT7630e 802.11bgn Wireless Network Adapter Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-SERR- Kernel driver in use: mt7630e Kernel modules: mt7630e
lsusb |grep -i mediatek:
Bus 002 Device 006: ID 0e8d:763f MediaTek Inc.
There is a similar issue here with a kickstarter campaign to resolve it. Maybe this will help.
There seems to be some progress on the issue linked by @horacekmi: https://github.com/lede-project/source/commit/9eacb9d7fc0b4c921f8d2ec91a51f10d8c3ae12f