Nanostation Loco M2 - WDS link drops out after a few hours

aczlan · November 26, 2017, 9:25pm

I have a pair of Nanostation Loco M2 access points setup as a wireless bridge over a ~1300 foot (~400M) line of sight link. Both have LEDE Reboot 17.01.3 r3533-d0bf257c46 / LuCI lede-17.01 branch (git-17.232.21093-079f65a) installed (upgrading to 17.0.4 tonight, I forgot that I hadn't updated these yet).
One is setup as a WDS AP and the other is setup as a WDS Client, after a few hours (4-12 hours?) the link drops out but still says that is connected (if I go to status it shows the other AP as connected but I cannot ping or do anything else across the link).
If I restart the wireless service on the client, the connection comes right back and all is well for a 4-12 hours.
Anyone have any suggestions on where to look? Below are the kernel and system logs for the time period in question:

AP Config files:
/etc/config/network:

config interface 'lan'
	option type 'bridge'
	option ifname 'eth0'
	option proto 'static'
	option netmask '255.255.255.0'
	option ipaddr 'x.x.x.2'
	option gateway 'x.x.x.1'
	option dns 'x.x.x.1'

/etc/config/wireless:

config wifi-device 'radio0'
	option type 'mac80211'
	option hwmode '11g'
	option path 'pci0000:00/0000:00:00.0'
	option channel '1'
	option country 'US'
	option htmode 'HT40'
	option distance '400'

config wifi-iface 'default_radio0'
	option device 'radio0'
	option network 'lan'
	option mode 'ap'
	option ssid 'LAN_bridge'
	option wds '1'
	option key 'SECRET'
	option encryption 'psk2+tkip+ccmp'

Client Config files:
/etc/config/network:

config interface 'lan'
	option type 'bridge'
	option proto 'static'
	option netmask '255.255.255.0'
	option ipaddr 'x.x.x.3'
	option gateway 'x.x.x.1'
	option dns 'x.x.x.1'
	option _orig_ifname 'eth0'
	option _orig_bridge 'true'
	option ifname 'eth0'
	option delegate '0'

/etc/config/wireless:

config wifi-device 'radio0'
	option type 'mac80211'
	option hwmode '11g'
	option path 'pci0000:00/0000:00:00.0'
	option channel '1'
	option htmode 'HT40'
	option country 'US'
	option distance '400'

config wifi-iface
	option ssid 'LAN_bridge'
	option device 'radio0'
	option mode 'sta'
	option bssid 'AP_BSSID'
	option key 'SECRET'
	option wds '1'
	option network 'lan'
	option encryption 'psk2+tkip+ccmp'

AP Syslog:

Sun Nov 26 14:01:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 14:11:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 14:21:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 14:31:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 14:41:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 14:51:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 15:01:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 15:11:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 15:21:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
**Somewhere in here the link dropped out and I rebooted the client**
Sun Nov 26 15:28:14 2017 daemon.notice hostapd: wlan0: AP-STA-DISCONNECTED ClientMAC
Sun Nov 26 15:28:14 2017 kern.info kernel: [1893620.914221] device wlan0.sta1 left promiscuous mode
Sun Nov 26 15:28:14 2017 kern.info kernel: [1893620.919427] br-lan: port 3(wlan0.sta1) entered disabled state
Sun Nov 26 15:28:52 2017 daemon.info hostapd: wlan0: STA ClientMAC IEEE 802.11: authenticated
Sun Nov 26 15:28:52 2017 daemon.info hostapd: wlan0: STA ClientMAC IEEE 802.11: associated (aid 1)
Sun Nov 26 15:28:52 2017 kern.info kernel: [1893659.422460] device wlan0.sta1 entered promiscuous mode
Sun Nov 26 15:28:52 2017 kern.info kernel: [1893659.440920] br-lan: port 3(wlan0.sta1) entered forwarding state
Sun Nov 26 15:28:52 2017 kern.info kernel: [1893659.447086] br-lan: port 3(wlan0.sta1) entered forwarding state
Sun Nov 26 15:28:53 2017 daemon.notice hostapd: wlan0: AP-STA-CONNECTED ClientMAC
Sun Nov 26 15:28:53 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: pairwise key handshake completed (RSN)
Sun Nov 26 15:28:54 2017 kern.info kernel: [1893661.440359] br-lan: port 3(wlan0.sta1) entered forwarding state
Sun Nov 26 15:31:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)
Sun Nov 26 15:41:13 2017 daemon.info hostapd: wlan0: STA ClientMAC WPA: group key handshake completed (RSN)

AP Kernel Log:

[1649563.614654] br-lan: port 1(eth0) entered disabled state
[1649564.612365] eth0: link up (100Mbps/Full duplex)
[1649564.617165] br-lan: port 1(eth0) entered forwarding state
[1649564.622888] br-lan: port 1(eth0) entered forwarding state
[1649566.620738] br-lan: port 1(eth0) entered forwarding state
[1652624.237670] ------------[ cut here ]------------
[1652624.242540] WARNING: CPU: 0 PID: 0 at compat-wireless-2017-01-31/net/mac80211/rx.c:4214 0x80da5cf0 [mac80211@80d80000+0x60780]()
[1652624.254355] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 113 (0x71)
[1652624.264049] Modules linked in: ath9k ath9k_common pppoe ppp_async iptable_nat ath9k_hw ath pppox ppp_generic nf_nat_ipv4 nf_conntrack_ipv6 nf_conntrack_ipv4 mac80211 ipt_REJECT ipt_MASQUERADE cfg80211 xt_time xt_tcpudp xt_state xt_nat xt_multiport xt_mark xt_mac xt_limit xt_conntrack xt_comment xt_TCPMSS xt_REDIRECT xt_LOG xt_CT slhc nf_reject_ipv4 nf_nat_redirect nf_nat_masquerade_ipv4 nf_nat nf_log_ipv4 nf_defrag_ipv6 nf_defrag_ipv4 nf_conntrack_rtcache nf_conntrack iptable_mangle iptable_filter ip_tables crc_ccitt compat ip6t_REJECT nf_reject_ipv6 nf_log_ipv6 nf_log_common ip6table_mangle ip6table_filter ip6_tables x_tables ohci_platform ohci_hcd ehci_platform ehci_hcd gpio_button_hotplug usbcore nls_base usb_common
[1652624.328632] CPU: 0 PID: 0 Comm: swapper Tainted: G        W       4.4.89 #0
[1652624.335791] Stack : 803875b4 00000000 00000001 803e0000 00000000 00000000 00000000 00000000
[1652624.335791] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.335791] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.335791] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.335791] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.335791] 	  ...
[1652624.371919] Call Trace:[<80071a8c>] 0x80071a8c
[1652624.376597] [<80071a8c>] 0x80071a8c
[1652624.380290] [<80081888>] 0x80081888
[1652624.383987] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.390133] [<800818e4>] 0x800818e4
[1652624.393854] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.399981] [<80260a14>] 0x80260a14
[1652624.403663] [<80260bf8>] 0x80260bf8
[1652624.407374] [<80ce8388>] 0x80ce8388 [ath9k_common@80ce8000+0x4800]
[1652624.413800] [<80e677f4>] 0x80e677f4 [ath9k@80e60000+0x16650]
[1652624.419701] [<80e677d4>] 0x80e677d4 [ath9k@80e60000+0x16650]
[1652624.425575] [<80e64a44>] 0x80e64a44 [ath9k@80e60000+0x16650]
[1652624.431446] [<80084448>] 0x80084448
[1652624.435130] [<80083e04>] 0x80083e04
[1652624.438821] [<800a8014>] 0x800a8014
[1652624.442508] [<8006a990>] 0x8006a990
[1652624.446194] [<80060bf4>] 0x80060bf4
[1652624.449876] 
[1652624.451557] ---[ end trace 86b60d5fad42c0e7 ]---
[1652624.456661] ------------[ cut here ]------------
[1652624.461517] WARNING: CPU: 0 PID: 0 at compat-wireless-2017-01-31/net/mac80211/rx.c:4214 0x80da5cf0 [mac80211@80d80000+0x60780]()
[1652624.473297] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 113 (0x71)
[1652624.482988] Modules linked in: ath9k ath9k_common pppoe ppp_async iptable_nat ath9k_hw ath pppox ppp_generic nf_nat_ipv4 nf_conntrack_ipv6 nf_conntrack_ipv4 mac80211 ipt_REJECT ipt_MASQUERADE cfg80211 xt_time xt_tcpudp xt_state xt_nat xt_multiport xt_mark xt_mac xt_limit xt_conntrack xt_comment xt_TCPMSS xt_REDIRECT xt_LOG xt_CT slhc nf_reject_ipv4 nf_nat_redirect nf_nat_masquerade_ipv4 nf_nat nf_log_ipv4 nf_defrag_ipv6 nf_defrag_ipv4 nf_conntrack_rtcache nf_conntrack iptable_mangle iptable_filter ip_tables crc_ccitt compat ip6t_REJECT nf_reject_ipv6 nf_log_ipv6 nf_log_common ip6table_mangle ip6table_filter ip6_tables x_tables ohci_platform ohci_hcd ehci_platform ehci_hcd gpio_button_hotplug usbcore nls_base usb_common
[1652624.547552] CPU: 0 PID: 0 Comm: swapper Tainted: G        W       4.4.89 #0
[1652624.554704] Stack : 803875b4 00000000 00000001 803e0000 00000000 00000000 00000000 00000000
[1652624.554704] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.554704] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.554704] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.554704] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.554704] 	  ...
[1652624.590833] Call Trace:[<80071a8c>] 0x80071a8c
[1652624.595502] [<80071a8c>] 0x80071a8c
[1652624.599197] [<80081888>] 0x80081888
[1652624.602893] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.609038] [<800818e4>] 0x800818e4
[1652624.612760] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.618896] [<80260a14>] 0x80260a14
[1652624.622578] [<80260bf8>] 0x80260bf8
[1652624.626277] [<80ce8388>] 0x80ce8388 [ath9k_common@80ce8000+0x4800]
[1652624.632718] [<80e677f4>] 0x80e677f4 [ath9k@80e60000+0x16650]
[1652624.638615] [<80e677d4>] 0x80e677d4 [ath9k@80e60000+0x16650]
[1652624.644489] [<80e64a44>] 0x80e64a44 [ath9k@80e60000+0x16650]
[1652624.650366] [<80084448>] 0x80084448
[1652624.654046] [<80083e04>] 0x80083e04
[1652624.657738] [<800a8014>] 0x800a8014
[1652624.661423] [<8006a990>] 0x8006a990
[1652624.665109] [<80060bf4>] 0x80060bf4
[1652624.668791] 
[1652624.670471] ---[ end trace 86b60d5fad42c0e8 ]---
[1652624.675578] ------------[ cut here ]------------
[1652624.680433] WARNING: CPU: 0 PID: 0 at compat-wireless-2017-01-31/net/mac80211/rx.c:4214 0x80da5cf0 [mac80211@80d80000+0x60780]()
[1652624.692211] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 113 (0x71)
[1652624.701903] Modules linked in: ath9k ath9k_common pppoe ppp_async iptable_nat ath9k_hw ath pppox ppp_generic nf_nat_ipv4 nf_conntrack_ipv6 nf_conntrack_ipv4 mac80211 ipt_REJECT ipt_MASQUERADE cfg80211 xt_time xt_tcpudp xt_state xt_nat xt_multiport xt_mark xt_mac xt_limit xt_conntrack xt_comment xt_TCPMSS xt_REDIRECT xt_LOG xt_CT slhc nf_reject_ipv4 nf_nat_redirect nf_nat_masquerade_ipv4 nf_nat nf_log_ipv4 nf_defrag_ipv6 nf_defrag_ipv4 nf_conntrack_rtcache nf_conntrack iptable_mangle iptable_filter ip_tables crc_ccitt compat ip6t_REJECT nf_reject_ipv6 nf_log_ipv6 nf_log_common ip6table_mangle ip6table_filter ip6_tables x_tables ohci_platform ohci_hcd ehci_platform ehci_hcd gpio_button_hotplug usbcore nls_base usb_common
[1652624.766474] CPU: 0 PID: 0 Comm: swapper Tainted: G        W       4.4.89 #0
[1652624.773641] Stack : 803875b4 00000000 00000001 803e0000 00000000 00000000 00000000 00000000
[1652624.773641] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.773641] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.773641] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.773641] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.773641] 	  ...
[1652624.809756] Call Trace:[<80071a8c>] 0x80071a8c
[1652624.814426] [<80071a8c>] 0x80071a8c
[1652624.818120] [<80081888>] 0x80081888
[1652624.821816] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.827961] [<800818e4>] 0x800818e4
[1652624.831683] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652624.837810] [<80260a14>] 0x80260a14
[1652624.841493] [<80260bf8>] 0x80260bf8
[1652624.845192] [<80ce8388>] 0x80ce8388 [ath9k_common@80ce8000+0x4800]
[1652624.851632] [<80e677f4>] 0x80e677f4 [ath9k@80e60000+0x16650]
[1652624.857530] [<80e677d4>] 0x80e677d4 [ath9k@80e60000+0x16650]
[1652624.863403] [<80e64a44>] 0x80e64a44 [ath9k@80e60000+0x16650]
[1652624.869275] [<80084448>] 0x80084448
[1652624.872960] [<80083e04>] 0x80083e04
[1652624.876638] [<800a8014>] 0x800a8014
[1652624.880331] [<8006a990>] 0x8006a990
[1652624.884015] [<80060bf4>] 0x80060bf4
[1652624.887697] 
[1652624.889376] ---[ end trace 86b60d5fad42c0e9 ]---
[1652624.894482] ------------[ cut here ]------------
[1652624.899340] WARNING: CPU: 0 PID: 0 at compat-wireless-2017-01-31/net/mac80211/rx.c:4214 0x80da5cf0 [mac80211@80d80000+0x60780]()
[1652624.911125] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 113 (0x71)
[1652624.920817] Modules linked in: ath9k ath9k_common pppoe ppp_async iptable_nat ath9k_hw ath pppox ppp_generic nf_nat_ipv4 nf_conntrack_ipv6 nf_conntrack_ipv4 mac80211 ipt_REJECT ipt_MASQUERADE cfg80211 xt_time xt_tcpudp xt_state xt_nat xt_multiport xt_mark xt_mac xt_limit xt_conntrack xt_comment xt_TCPMSS xt_REDIRECT xt_LOG xt_CT slhc nf_reject_ipv4 nf_nat_redirect nf_nat_masquerade_ipv4 nf_nat nf_log_ipv4 nf_defrag_ipv6 nf_defrag_ipv4 nf_conntrack_rtcache nf_conntrack iptable_mangle iptable_filter ip_tables crc_ccitt compat ip6t_REJECT nf_reject_ipv6 nf_log_ipv6 nf_log_common ip6table_mangle ip6table_filter ip6_tables x_tables ohci_platform ohci_hcd ehci_platform ehci_hcd gpio_button_hotplug usbcore nls_base usb_common
[1652624.985380] CPU: 0 PID: 0 Comm: swapper Tainted: G        W       4.4.89 #0
[1652624.992546] Stack : 803875b4 00000000 00000001 803e0000 00000000 00000000 00000000 00000000
[1652624.992546] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.992546] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.992546] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.992546] 	  00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[1652624.992546] 	  ...
[1652625.028662] Call Trace:[<80071a8c>] 0x80071a8c
[1652625.033331] [<80071a8c>] 0x80071a8c
[1652625.037027] [<80081888>] 0x80081888
[1652625.040722] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652625.046868] [<800818e4>] 0x800818e4
[1652625.050589] [<80da5cf0>] 0x80da5cf0 [mac80211@80d80000+0x60780]
[1652625.056714] [<80260a14>] 0x80260a14
[1652625.060410] [<80260bf8>] 0x80260bf8
[1652625.064115] [<80ce8388>] 0x80ce8388 [ath9k_common@80ce8000+0x4800]
[1652625.070555] [<80e677f4>] 0x80e677f4 [ath9k@80e60000+0x16650]
[1652625.076442] [<80e677d4>] 0x80e677d4 [ath9k@80e60000+0x16650]
[1652625.082330] [<80e64a44>] 0x80e64a44 [ath9k@80e60000+0x16650]
[1652625.088198] [<80084448>] 0x80084448
[1652625.091882] [<80083e04>] 0x80083e04
[1652625.095561] [<800a8014>] 0x800a8014
[1652625.099258] [<8006a990>] 0x8006a990
[1652625.102938] [<80060bf4>] 0x80060bf4
[1652625.106607] 
[1652625.108295] ---[ end trace 86b60d5fad42c0ea ]---
[1656961.531085] device wlan0.sta1 left promiscuous mode
[1656961.536375] br-lan: port 3(wlan0.sta1) entered disabled state
[1657000.400508] device wlan0.sta1 entered promiscuous mode
[1657000.416849] br-lan: port 3(wlan0.sta1) entered forwarding state
[1657000.423019] br-lan: port 3(wlan0.sta1) entered forwarding state
[1657002.415840] br-lan: port 3(wlan0.sta1) entered forwarding state
[1729152.802304] eth0: link down
[1729152.805769] br-lan: port 1(eth0) entered disabled state
[1729180.803335] eth0: link up (100Mbps/Full duplex)
[1729180.808173] br-lan: port 1(eth0) entered forwarding state
[1729180.813898] br-lan: port 1(eth0) entered forwarding state
[1729182.811672] br-lan: port 1(eth0) entered forwarding state
[1893620.914221] device wlan0.sta1 left promiscuous mode
[1893620.919427] br-lan: port 3(wlan0.sta1) entered disabled state
[1893659.422460] device wlan0.sta1 entered promiscuous mode
[1893659.440920] br-lan: port 3(wlan0.sta1) entered forwarding state
[1893659.447086] br-lan: port 3(wlan0.sta1) entered forwarding state
[1893661.440359] br-lan: port 3(wlan0.sta1) entered forwarding state
[1894771.793057] device wlan0.sta1 left promiscuous mode
[1894771.798391] br-lan: port 3(wlan0.sta1) entered disabled state
[1894777.535302] device wlan0.sta1 entered promiscuous mode
[1894777.547462] br-lan: port 3(wlan0.sta1) entered forwarding state
[1894777.553631] br-lan: port 3(wlan0.sta1) entered forwarding state
[1894779.546485] br-lan: port 3(wlan0.sta1) entered forwarding state
[1895148.287966] device wlan0.sta1 left promiscuous mode
[1895148.293178] br-lan: port 3(wlan0.sta1) entered disabled state
[1895148.349721] device wlan0 left promiscuous mode
[1895148.354492] br-lan: port 2(wlan0) entered disabled state
[1895150.539163] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[1895150.663905] device wlan0 entered promiscuous mode
[1895153.075046] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[1895153.081886] br-lan: port 2(wlan0) entered forwarding state
[1895153.087660] br-lan: port 2(wlan0) entered forwarding state
[1895154.299110] device wlan0.sta1 entered promiscuous mode
[1895154.316512] br-lan: port 3(wlan0.sta1) entered forwarding state
[1895154.322716] br-lan: port 3(wlan0.sta1) entered forwarding state
[1895155.086003] br-lan: port 2(wlan0) entered forwarding state
[1895156.315941] br-lan: port 3(wlan0.sta1) entered forwarding state

Thanks in advance

Aaron Z

aczlan · November 30, 2017, 9:36pm

I was disconnected again today, here is the status on the Luci Wireless Overview, at this point any connections (be it ping, telnet, ssh, http, etc) to any devices at the far end of the link fail, yet it says that it is connected. If I disable and re-enable the wireless, it works again.
ConnectedButNotReally

Aaron Z

cyablo · January 24, 2018, 7:59am

Same problem here with TP-LINK WDR3600 <-> TP-LINK CPE210 WDS bridge. Both ar71xx / ath9k devices. WDS bridge with WDR3600 <-> A5-V11 Ralink is rocksolid by the way.

cyablo · February 1, 2018, 10:33am

I tried to use a spare TP-LINK 841N v11 as dedicated WDS Staion for the TP-LINK CPE210 Client... this is even worse, the bridge breaks after at least 1 hour and the Station has to be restarted. Both devices on latest stable.

mk24 · February 1, 2018, 7:59pm

I have not had this problem. I run WDS on various ath9k devices including those with separate CPU / wifi chip (MR3420 v1, Nanobeam M5 XW) and AR9341 SoC (WNR2000v4, TL-WR841). The Nanobeam XW will also happily WDS to the ath10k chip in a Netgear R6100.

Are you using HT40, because I always have used HT20. The original post suggests that it is a bug related to a specific rate selection. Check your log files are you getting the same message? Does the logged error occur on the AP, the client, or both?

cyablo · February 2, 2018, 9:29am

I'm using HT20 on 2.4 GHz. The Kernel Warning is only showed in the Station's Log, Client has no log entries like this.

cyablo · February 2, 2018, 8:53pm

Even more Problems with 841N <-> CPE210, most of the time the CPE210 Client now locks up and has to be power cycled. Restarting the Station does not help. Would like to try snapshots builds on CPE210 but since months they fail to build.

BTW: It was nearly fine with an earlier snapshot in May 2017: WiFi on WDR3600 unstable after 2-3 Days

So, some changes caused the problem to come back.

Already filled a ticket within the bug tracker months ago but got no response on that.

aczlan · February 3, 2018, 4:38am

I opened a ticket as well, but not much response: https://bugs.lede-project.org/index.php?do=details&task_id=1246&only_watched=1
Someone suggested trying batman-adv, that did not work at all and made a mess of things. I am now trying the openwrt-ar71xx-generic-ubnt-rocket-m-squashfs-sysupgrade.bin image and if that doesnt work, its back to the stock firmware and off to Amazon to buy a 3rd AP.

Aaron Z

mk24 · February 3, 2018, 3:12pm

@aczlan What hardware is in your Loco 2's exactly? Did they have "XM" and "XW" versions? Is it 32 MB RAM? I don't have one but I could try setting up a test on similar hardware. Like I said, I've never had this problem while running WDS on a lot of ath9k devices.

I don't think Rocket firmware will work on a Nanostation Loco M2, because the Ethernet port configuration is different. Anyway it would still be the same ath9k wifi stack.

aczlan · February 3, 2018, 3:32pm

They are the XM version (bought them in 2011). 32MB of RAM. Both the Nanostation Loco M2 and the Rocket M2 have a single Ethernet port. The Nanostation M2 (the non-Loco version) has 2 Ethernet ports.
I installed the Rocket firmware last night, reset both to Openwrt/LEDE defaults and configured them again, waiting to see if it drops out again.

Aaron Z

cyablo · February 5, 2018, 12:58pm

Build a snapshot for the CPE210 myself and updated it from stable lede, also switched back to WDR3600 as station, now station and client running both latest snapshots with 4.9.77 kernel. Will report if that helps.

aczlan · February 5, 2018, 2:05pm

I have a CPE210 en-route from Amazon for a separate project at work (looking at it as a possible replacement for the Nanostation Loco M2 now that Ubiquity is less than friendly about 3rd party firmware). Perhaps I will try a link with a CPE210 and a Nanostation.

Aaron Z

cyablo · February 7, 2018, 11:06am

FYI: Current setup with WDR3600 <-> CPE210 both on new snapshot build with 4.9.77 kernel, stable since 2 days.

aczlan · February 7, 2018, 12:03pm

I will have to try that tonight. I had switched to "mesh" network and replaced wpad-mini with wpad-full, but it still flaked out last night.

Aaron Z

cyablo · February 8, 2018, 9:08am

CPE210 just rebooted by itself 15 hours ago, don't know what happened. maybe I have to use external log server if this happens a lot. Anyway, rebooting is still better than getting unstable and stopping to work

Edit:

In fact it did not 'really' rebooted, just like the Uptime got a reset. Log also looks strange, there is some massive gap but no entries that indicate a hard reboot:

Mon Feb  5 13:31:06 2018 kern.info kernel: [   24.623836] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: 
link becomes ready
Mon Feb  5 13:31:06 2018 kern.info kernel: [   24.630655] br-lan: port 2(wlan0) entered blocking state
Mon Feb  5 13:31:06 2018 kern.info kernel: [   24.636062] br-lan: port 2(wlan0) entered forwarding state
Wed Feb  7 18:07:03 2018 kern.notice kernel: [   43.842261] random: crng init done
Thu Feb  8 02:14:35 2018 daemon.notice netifd: Network device 'eth0' link is down
Thu Feb  8 02:14:35 2018 kern.info kernel: [29295.850860] eth0: link down
Thu Feb  8 02:14:35 2018 kern.info kernel: [29295.855541] br-lan: port 1(eth0.1) entered disabled state

cyablo · February 12, 2018, 7:42pm

Alright, now it broke. after 8 Days the WDR3600 went unstable again and needed to be rebooted.

aczlan · February 12, 2018, 7:58pm

Mine is as well, but it happens multiple times per day (with the same blocking, forwarding/down sequence). My wife says that its time to go back to the stock firmware posthaste as she is tired of it cutting out in the middle of watching something.

Aaron Z

i4freegr · February 13, 2018, 1:32am

Nano M2 & CPE210 meshing with batman-adv openwrt 15 12 nodes & 3 uplinks working fine since 18 months

Only problem with Nano M2 eth interface stability.
Nodes LOS 2km 300bps raw speed

we are about to switch to LEDE and form adhoc mesh to 802.11s

In case you need some help, let me know

greetings
3zl

aczlan · February 13, 2018, 2:39am

Do you have an example of your config files that will work with LEDE/OpenWrt? Is it just the base image with batman-adv installed?
I tried batman-adv, but couldn't get an encrypted link to work, couldn't find useful documentation for the config file and didn't have any more time to fuss with it.

Aaron Z

i4freegr · February 13, 2018, 1:29pm

here you go (please read some docs abt batman-adv)

batman-adv configurations

/etc/config

batman-adv : example working for me or leave it as default configured

 config mesh 'bat0'
	option aggregated_ogms '1'
	option ap_isolation '0'
	option bonding '0'
	option fragmentation '0'
	option gw_bandwidth '2000/512'
	option gw_mode 'client'
	option gw_sel_class '20'
	option log_level '2'
	option orig_interval '1000'
	option bridge_loop_avoidance '1'
	option distributed_arp_table '1'
	option multicast_mode '1'
	option network_coding '0'
	option hop_penalty '30'
	option isolation_mark '0'

network:

config interface 'lan'
.....
	option ifname 'eth0.1 bat0'  #add  bat0 virt interface to br-lan bridge

config interface 'radio0_batnet'
	option mtu '1536'
	option auto '1'
	option proto 'batadv'
	option mesh 'bat0'	#corresp to batman-adv-configuration

wireless

# enable only ONE  mesh interfacee config
#802.11s preferred for us now


# mesh interface adhoc configurations

config wifi-iface 'default_radio0_adhoc'	
	option device 'radio0'	#your radio name
	option ifname 'radio0_mesh'	
	option network 'radio0_batnet'	#corrsponding too network-config-interface
	option mode 'adhoc'
	option encryption 'none'	#config first witout encr
	option bssid 'CA:FE:00:CO:FF:EE'	#all nodes MUST have same MAC-like bssid !!
	option ssid 'i4f-127d46.5.mesh'		#name of your joice
	option disabled '0'

	
# mesh interface 802.11s configurations
	
config wifi-iface 'default_radio0_mesh'
	option device 'radio0'
	option ifname 'radio0_mesh'
	option network 'radio0_batnet'
	option mode 'mesh'
	option encryption 'none'
	option mesh_id 'CA:FE:00:CO:FF:EE'
	option ssid 'i4f-127d46.5.mesh'
	option mesh_fwding '0'
	option disabled '1'
	
	
# config come direct from working network...

greetings
3zl