Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

-55 dBm on both sides, good noise floor, bad link

Signal strenght: -55dBm on both sides
Noise floor: -94 and -92
Device: 6.6Km RocketM5 + RocketDish on both sides
MCS14+Auto
CCQ: 86% to 30%
TxPower: 12dB on both sides
bridge mode + wds
SpeedTest From AP to Station:
RX: 7.58 Mbps
TX: 35.48 Mbps
Total: 43.06 Mbps
SpeedTest From Station to AP:
RX: 32.54 Mbps
TX: 8.12 Mbps
Total: 40.65 Mbps
On Both side i see error on wifi if (both in TX):

AP:
XM.v5.2.1# ifconfig wifi0
wifi0 Link encap:Ethernet HWaddr 00:15:6D:xx:xx:8D
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:453590 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:48 Memory:b0000000-b0010000
STA:
XM.v5.2.1# ifconfig wifi0
wifi0 Link encap:Ethernet HWaddr 00:15:6D:xx:xx:2E
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:253037 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:48 Memory:b0000000-b0010000


I have a dream: mpls through ubiquiti devices.
Established Member
doush
Posts: 1,163
Registered: ‎05-20-2008
Posts: 1163
Kudos: 203
Registered: 05-20-2008

Re: -55 dBm on both sides, good noise floor, bad link

I would try
1- lower MCS rates on both sides
2- increase the Tx power a bit
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

why increase power if at the other side i have a strong signal?
lower mcs... ok.. but this is a link of 6.6km at maximium signal strenght and good noisefloor if I can not expect a best mcs there, where i can?! :\
I have a dream: mpls through ubiquiti devices.
Regular Member
kbloch
Posts: 423
Registered: ‎12-07-2009
Posts: 423
Kudos: 112
Solutions: 12
Registered: 12-07-2009

Re: -55 dBm on both sides, good noise floor, bad link

Change your RF channel. My guess is that you have interference on one end of your link and not the other. Run the site survey on both ends as a first step and see if any other netowrks are detected. If so try to switch to a frequency that is clear on both ends. If the site survey does not find anything then try the spectrum analyzer on each end.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

kbloch, i'm already on a free channel... see my noise floor
I have a dream: mpls through ubiquiti devices.
Established Member
doush
Posts: 1,163
Registered: ‎05-20-2008
Posts: 1163
Kudos: 203
Registered: 05-20-2008

Re: -55 dBm on both sides, good noise floor, bad link

When I decrease tx power on one of my PtP link which has -50dbm signal, CCQ drops a lot. Airmax capacity falls aswell. Increasing TX power, gets me better CCQ. I know this is strange.
I would definitely increase the TX power and try.
Member
jdiggity
Posts: 112
Registered: ‎07-29-2010
Posts: 112
Kudos: 9
Solutions: 1
Registered: 07-29-2010

Re: -55 dBm on both sides, good noise floor, bad link

Same as me for months. Please update if you find a solution... Here are the different threads with same problems I have tried...

www.ubnt.com/forum/search.php?searchid=546380
Emerging Member
samalama
Posts: 41
Registered: ‎05-07-2008
Posts: 41
Registered: 05-07-2008

Re: -55 dBm on both sides, good noise floor, bad link



How did you get those specs on error rates? Telnet or somewhere in the UI that I haven't seen?
Emerging Member
samalama
Posts: 41
Registered: ‎05-07-2008
Posts: 41
Registered: 05-07-2008

Re: -55 dBm on both sides, good noise floor, bad link

nevermind. I ssh'd into it and figured it out.
Established Member
kijoma
Posts: 762
Registered: ‎07-13-2008
Posts: 762
Kudos: 42
Registered: 07-13-2008

Re: -55 dBm on both sides, good noise floor, bad link

hi,
we have experienced this issue too, AP to STA (WDS) full speed. reverse direction about 8 MBps..
any ideas?
Bill Lewis
Kijoma Broadband
UK 5 GHz WISP - ISPA UK, RIPE Member
Senior Design Eng , RF / Network Eng.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

When I decrease tx power on one of my PtP link which has -50dbm signal, CCQ drops a lot. Airmax capacity falls aswell. Increasing TX power, gets me better CCQ. I know this is strange.
I would definitely increase the TX power and try.

not work :\
update:
the port on rb100 where AP with errors is attached, have full duplex off. why off?
is it possible that this issue derive from ethernet side? why i see error on wifi0?
on rb1000 i see tx error on ethernet port!
I have a dream: mpls through ubiquiti devices.
Established Member
doush
Posts: 1,163
Registered: ‎05-20-2008
Posts: 1163
Kudos: 203
Registered: 05-20-2008

Re: -55 dBm on both sides, good noise floor, bad link

siddolo;
run the link with full tx power and default max MCS rates.
make the speed test from a dedicated PC or router. Also try with Airmax on and off.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

i think that problem is on ethernet side...
I have a dream: mpls through ubiquiti devices.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

ubnt staff, can i read some log messages for this error? thanks
I have a dream: mpls through ubiquiti devices.
Previous Employee
UBNT-Mike.Ford
Posts: 9,171
Registered: ‎08-17-2007
Posts: 9171
Kudos: 13
Registered: 08-17-2007

Re: -55 dBm on both sides, good noise floor, bad link

ubnt staff, can i read some log messages for this error? thanks


Hello,

Can you post a DMESG output here so we can see if any error messages are popping up?

Thanks,
Established Member
Headbang
Posts: 1,383
Registered: ‎03-16-2008
Posts: 1383
Kudos: 34
Solutions: 1
Registered: 03-16-2008

Re: -55 dBm on both sides, good noise floor, bad link

I got a call from our sister WISP today, the tech was at wits end on 2 problems. First one was easy, just a ethernet issue. Second was the problem others are having. Slow link negotiation 1 way.

I did fast walk through on this with him, 2 points we changed were ack and firmware.

I had him set manual ack to 1.33 the distance.
We dropped from 5.2.1 to 5.2 firmware.

Problem Solved

Doesn't hurt to give it a try, I don't run anything newer then 5.2 on P2P links.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

Hello,

Can you post a DMESG output here so we can see if any error messages are popping up?

Thanks,



st:st_start_test:1507: Started...
spdtst:stctl_write:2578: Releasing ses, current users: 2, err: 0.
spdtst:st_nf_rx:1619: Devices: in: br0, out: ath0.
spdtst:st_nf_rx:1621: MAC: 00:15:6D:xx:xx:8D -> 00:15:6D:xx:xx:2E(6)
spdtst:st_nf_rx:1623: IP: 10.10.10.2 -> 10.10.10.5
spdtst:st_nf_rx:1624: PORT: 9 -> 9
spdtst:st_nf_rx:1626: CONNECTED
spdtst:st_receive_params:1923: Receiving PARAMS
spdtst:st_receive_params:1936: Got flags: 0x0007
spdtst:st_process_state:974: State CHANGE: 1 -> 3
spdtst:__st_state_enter:883: Entered, state: 3
spdtst:st_process_tx_params:874: Setting TX params(1): 20, Rate: 1
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_params:1890: build PARAMS
spdtst:st_build_packet:1197: built packet: 12(1460)
spdtst:st_fill_header:1183: built packet: 24
spdtst:st_fill_packet:1252: built packet: 24(1472)
spdtst:st_receive_rx_ready:1978: Receiving RX_READY
spdtst:st_process_state:974: State CHANGE: 3 -> 5
spdtst:__st_state_enter:883: Entered, state: 5
spdtst:st_process_tx_params:874: Setting TX params(2): 1, Rate: 13
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_data:1811: build DATA
spdtst:st_build_packet:1197: built packet: 1460(1460)
spdtst:st_fill_header:1183: built packet: 1472
spdtst:st_fill_packet:1252: built packet: 1472(1472)
spdtst:st_check_data_end:1993: Receiving DATA_END
spdtst:st_on_test_finish:1739: Before Qdisc reset -------------
spdtst:st_on_test_finish:1742: Qdisc stats : qlen = 0, backlog = 0, drops = 0, requeues = 0, overlimits = 0.
spdtst:st_on_test_finish:1743: Qdisc rate estimated: bps = 0, pps = 0.
spdtst:st_on_test_finish:1744: After Qdisc reset -------------
spdtst:st_on_test_finish:1747: Qdisc stats: qlen = 0, backlog = 0, drops = 0, requeues = 0, overlimits = 0.
spdtst:st_on_test_finish:1748: Qdisc rate estimated: bps = 0, pps = 0.
spdtst:st_process_state:974: State CHANGE: 5 -> 8
spdtst:__st_state_enter:883: Entered, state: 8
spdtst:st_process_tx_params:874: Setting TX params(1): 20, Rate: 1
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_results:1848: build RESULTS
spdtst:st_build_packet:1197: built packet: 80(1460)
spdtst:st_fill_header:1183: built packet: 92
spdtst:st_fill_packet:1252: built packet: 92(1472)
spdtst:st_receive_finish:2109: Receiving FINISH
spdtst:st_process_state:974: State CHANGE: 8 -> 10
spdtst:__st_state_enter:883: Entered, state: 10
spdtst:st_process_tx_params:874: Setting TX params(0): 20, Rate: 1
spdtst:__st_state_enter:930: AUTOSHUTDOWN'ing...
spdtst:st_tasklet:1399: Leaving...: asleep = 10282684, awake = 10008.
Session Data Size 1500
Local TX: time 10006853us frames 29285, dropped 9835 34294 kbps (2926 pps)
Remote TX: time 0us frames 0, dropped 0 0 kbps (0 pps)
Local RX: time 9986942us frames 7083, dropped 0 8311 kbps (709 pps)
Remote RX: time 0us frames 0, dropped 0 0 kbps (0 pps)
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0004
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0009
spdtst:st_netdevice_event:2743: Device 'wifi0' has event: 0x0009
spdtst:st_netdevice_event:2743: Device 'wifi0' has event: 0x0002
br0: port 3(ath0) entering disabled state
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0002
spdtst:st_netdevice_event:2743: Device 'br0' has event: 0x0009
br0: port 1(eth0_real) entering disabled state
spdtst:st_netdevice_event:2743: Device 'br0' has event: 0x0002
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0009
ag7240_ring_free Freeing at 0x8363f800
ag7240_ring_free Freeing at 0x83677000
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0002
spdtst:st_netdevice_event:2743: Device 'eth1_real' has event: 0x0009
ag7240_ring_free Freeing at 0x82ed6800
ag7240_ring_free Freeing at 0x82c0b000
spdtst:st_netdevice_event:2743: Device 'eth1_real' has event: 0x0002
spdtst:st_netdevice_event:2743: Device 'eth0' has event: 0x000A
Ebtables v2.0 unregistered
eth0: port 3(ath0) entering disabled state
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0006
spdtst:st_netdevice_event:2743: Device 'wifi0' has event: 0x0006
ath_pci: driver unloaded
eth0: port 1(eth0_real) entering disabled state
eth0: port 2(eth1_real) entering disabled state
spdtst:st_netdevice_event:2743: Device 'eth0' has event: 0x0006
ath_dev: driver unloaded
wlan_me: driver unloaded
wlan: mac acl policy unregistered
ath_rate_atheros: driver unloaded
ath_hal: driver unloaded
ubnt_poll: Uninitializing packet classifier...
ubnt_poll: Uninitializing module
wlan: driver unloaded
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0006
ag7240_cleanup Freeing at 0x83d38000
spdtst:st_netdevice_event:2743: Device 'eth1_real' has event: 0x0006
ag7240_cleanup Freeing at 0x810b4000
AG7240: cleanup done
AG7240: Length per segment 1540
AG7240: Max segments per packet 1
AG7240: Max tx descriptor count 80
AG7240: Max rx descriptor count 252
AG7240: fifo cfg 3 01f00140
I have a dream: mpls through ubiquiti devices.
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link


AG7240CHH: Mac address for unit 0
AG7240CHH: 00:15:6d:xx:xx:8d
spdtst:st_netdevice_event:2743: Device 'eth0' has event: 0x0005
AG7240CHH: Mac address for unit 1
AG7240CHH: 02:15:6d:xx:xx:8d
spdtst:st_netdevice_event:2743: Device 'eth1' has event: 0x0005
wlan: 0.8.4.2 (Atheros/multi-bss)
ubnt_poll: Initializing polling module...
ubnt_poll: Initializing packet classifier...
ubnt_poll: OS timer frequency 1000 Hz
ath_hal: 0.9.17.1 (AR5416, DEBUG, REGOPS_FUNC, 11D)
ath_rate_atheros: Copyright (c) 2001-2005 Atheros Communications, Inc, All Rights Reserved
wlan: mac acl policy registered
wlan_me: Version 0.1
Copyright (c) 2008 Atheros Communications, Inc. All Rights Reserved
ath_dev: Copyright (c) 2001-2007 Atheros Communications, Inc, All Rights Reserved
rssi-leds 1.1 loaded
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x000A
spdtst:st_netdevice_event:2743: Device 'eth1_real' has event: 0x000A
spdtst:st_netdevice_event:2743: Device 'eth0' has event: 0x0005
bridge: can't decode speed from eth0_real: 0
device eth0_real entered promiscuous mode
bridge: can't decode speed from eth1_real: 0
device eth1_real entered promiscuous mode
ath_pci: 0.9.4.5 (Atheros/multi-bss)
PCI: Setting latency timer of device 0000:00:00.0 to 64
ar5416Attach: Mac Chip.Rev 0x80.2 PCI devid 0x2a
ath_hal: Ubiquiti Merlin
ath_hal: NF adjust values: 5G - -114/-116/-120/-114/-116/-120, 2G: -110/0/0/-110/0/0
spdtst:st_netdevice_event:2743: Device 'wifi0' has event: 0x0005
wifi0: Atheros 9280: mem=0x10000000, irq=48 hw_base=0xb0000000
ubnt_poll: Attaching hooks...
ubnt_poll: ack_mode set to 0
ubnt_poll: Using antenna array type: None, ports=1, default=1
ubnt_poll: Setting timing params to min:1000 max:3000 offset:3000 tx_time: 3000
ubnt_poll: Setting aggregation params to non-txop:0 tx-op:0 scheduling:0
ubnt_poll: Setting TXOP Window parameters to size:25 timeout-perc:5
ubnt_poll: Module started in AP mode
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0005
Ebtables v2.0 registered
ag7240_ring_alloc Allocated 1280 at 0x83bbf800
ag7240_ring_alloc Allocated 4032 at 0x83473000
ATHRS26: resetting s26
ATHRS26: s26 reset done
Setting PHY...
AG7240: enet unit:0 phy:4 is up...RGMii 100Mbps full duplex
AG7240: done cfg2 0x7135 ifctl 0x10000 miictrl
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0001
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0004
ag7240_ring_alloc Allocated 1280 at 0x8359b800
ag7240_ring_alloc Allocated 4032 at 0x82f4d000
Setting PHY...
spdtst:st_netdevice_event:2743: Device 'eth1_real' has event: 0x0001
spdtst:st_netdevice_event:2743: Device 'br0' has event: 0x000A
spdtst:st_netdevice_event:2743: Device 'wifi0' has event: 0x0001
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0001
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0004
device ath0 entered promiscuous mode
AG7240: unit 0: phy 4 not up carrier 1
AG7240: enet unit:0 phy:4 is down...
br0: port 3(ath0) entering learning state
spdtst:st_netdevice_event:2743: Device 'br0' has event: 0x0001
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0004
br0: topology change detected, propagating
br0: port 3(ath0) entering forwarding state
AG7240: enet unit:0 phy:4 is up...RGMii 100Mbps full duplex
AG7240: done cfg2 0x7135 ifctl 0x10000 miictrl
spdtst:st_netdevice_event:2743: Device 'ath0' has event: 0x0004
br0: port 1(eth0_real) entering learning state
spdtst:st_netdevice_event:2743: Device 'eth0_real' has event: 0x0004
br0: topology change detected, propagating
br0: port 1(eth0_real) entering forwarding state
spdtst:stctl_write:2497: got write string: (10)533 slave
.
spdtst:__st_create_ses:473: creating session: 533(count: 1).
spdtst:st_get_ses_cmd:545: getting session: 533(count: 2).
spdtst:stctl_write:2530: Got ses, current users: 2.
spdtst:st_process_state:974: State CHANGE: 0 -> 1
spdtst:__st_state_enter:883: Entered, state: 1
spdtst:st_process_tx_params:874: Setting TX params(1): 20, Rate: 1
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_start_test:1507: Started...
spdtst:stctl_write:2578: Releasing ses, current users: 2, err: 0.
spdtst:st_nf_rx:1619: Devices: in: br0, out: ath0.
spdtst:st_nf_rx:1621: MAC: 00:15:6D:xx:xx:8D -> 00:15:6D:xx:xx:2E(6)
spdtst:st_nf_rx:1623: IP: 10.10.10.2 -> 10.10.10.5
spdtst:st_nf_rx:1624: PORT: 9 -> 9
spdtst:st_nf_rx:1626: CONNECTED
spdtst:st_receive_params:1923: Receiving PARAMS
spdtst:st_receive_params:1936: Got flags: 0x0007
spdtst:st_process_state:974: State CHANGE: 1 -> 3
spdtst:__st_state_enter:883: Entered, state: 3
spdtst:st_process_tx_params:874: Setting TX params(1): 20, Rate: 1
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_params:1890: build PARAMS
spdtst:st_build_packet:1197: built packet: 12(1460)
spdtst:st_fill_header:1183: built packet: 24
spdtst:st_fill_packet:1252: built packet: 24(1472)
spdtst:st_receive_rx_ready:1978: Receiving RX_READY
spdtst:st_process_state:974: State CHANGE: 3 -> 5
spdtst:__st_state_enter:883: Entered, state: 5
spdtst:st_process_tx_params:874: Setting TX params(2): 1, Rate: 13
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_data:1811: build DATA
spdtst:st_build_packet:1197: built packet: 1460(1460)
spdtst:st_fill_header:1183: built packet: 1472
spdtst:st_fill_packet:1252: built packet: 1472(1472)
spdtst:st_check_data_end:1993: Receiving DATA_END
spdtst:st_on_test_finish:1739: Before Qdisc reset -------------
spdtst:st_on_test_finish:1742: Qdisc stats : qlen = 0, backlog = 0, drops = 0, requeues = 0, overlimits = 0.
spdtst:st_on_test_finish:1743: Qdisc rate estimated: bps = 0, pps = 0.
spdtst:st_on_test_finish:1744: After Qdisc reset -------------
spdtst:st_on_test_finish:1747: Qdisc stats: qlen = 0, backlog = 0, drops = 0, requeues = 0, overlimits = 0.
spdtst:st_on_test_finish:1748: Qdisc rate estimated: bps = 0, pps = 0.
spdtst:st_process_state:974: State CHANGE: 5 -> 8
spdtst:__st_state_enter:883: Entered, state: 8
spdtst:st_process_tx_params:874: Setting TX params(1): 20, Rate: 1
spdtst:__st_state_enter:906: not DONE...
spdtst:__st_state_enter:909: Registering timeout: 5000
spdtst:st_build_results:1848: build RESULTS
spdtst:st_build_packet:1197: built packet: 80(1460)
spdtst:st_fill_header:1183: built packet: 92
spdtst:st_fill_packet:1252: built packet: 92(1472)
spdtst:st_receive_finish:2109: Receiving FINISH
spdtst:st_process_state:974: State CHANGE: 8 -> 10
spdtst:__st_state_enter:883: Entered, state: 10
spdtst:st_process_tx_params:874: Setting TX params(0): 20, Rate: 1
spdtst:__st_state_enter:930: AUTOSHUTDOWN'ing...
spdtst:st_tasklet:1399: Leaving...: asleep = 9763931, awake = 10017.
Session Data Size 1500
Local TX: time 10015965us frames 29947, dropped 9827 35038 kbps (2989 pps)
Remote TX: time 0us frames 0, dropped 0 0 kbps (0 pps)
Local RX: time 10004548us frames 10986, dropped 0 12868 kbps (1098 pps)
Remote RX: time 0us frames 0, dropped 0 0 kbps (0 pps)
XM.v5.2.1#
I have a dream: mpls through ubiquiti devices.
Previous Employee
UBNT-Mike.Ford
Posts: 9,171
Registered: ‎08-17-2007
Posts: 9171
Kudos: 13
Registered: 08-17-2007

Re: -55 dBm on both sides, good noise floor, bad link

Hello,

Asking Edmundas to take a look.

Thanks,
Member
siddolo
Posts: 115
Registered: ‎10-28-2009
Posts: 115
Registered: 10-28-2009

Re: -55 dBm on both sides, good noise floor, bad link

do you have news?
I have a dream: mpls through ubiquiti devices.