You are here

LQ/NLQ on DtD Links?

7 posts / 0 new
Last post
w6bi
w6bi's picture
LQ/NLQ on DtD Links?
Under what conditions would the LQ and/or NLQ on a DtD link not be 100/100?

Orv W 6BI
AE6XE
AE6XE's picture
OLSR determines LQ and NLQ
OLSR determines LQ and NLQ based on broadcast UDP packets.  If a packet goes missing, then the LQ or NLQ drops.   The OLSR packets have sequence #s to determine if one has been lost.   Packets can be lost over dtdlink with a flooded network, hardware failures, etc.   In the out-of-box settings for DtDLink, the settings are different than RF.   It has been thought the drop threshold of the link for dtdlink would occur if LQ went below 95%, but I had once observed a live link with lower %LQ, so not so sure -- needs more investigation.    Are the links over dtdlink dropping or showing lower than 95% LQ or NLQ? 
w6bi
w6bi's picture
Will investigate
I believe we've seen them lower, but I can't find an example at the moment.  I'll keep my eyes open.

 
w6bi
w6bi's picture
Found one...
    100% 100% 57.7    
N6FL-VC-OjaiEast-W-Sector-2G.local.mesh   (dtd)   100% 100%      
N6FL-VC-Reeves-Rd-to-Black-Mtn-Link-5G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-Camarillo-Hills-SW-Sector-5G.local.mesh   (dtd)   84% 100%      
WD6EBY-VC-RasnowPk-N-Sector-5G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SimiEast-2G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SimiEast-3G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SimiEast-5G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SouthMtn-E-Sector-2G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SouthMtn-W-Sector-2G.local.mesh   (dtd)   100% 100%      
WD6EBY-VC-SouthMtn-W-Sector-5G.local.mesh   (dtd)   100% 100%     IperfSpeed
WD6EBY-VC-Sulphur-ResourceServer.local.mesh   (dtd)   100% 100%  
Support File Attachments: 
ke6bxt
ke6bxt's picture
There are two ways to make a

There are two ways to make a dtd connection 

The "real" way is by using a cat5e cable either directly from one node to another or from a node, to a switch, and from the switch to another node. 
Another way is by using a (maybe part 15, maybe TDMA) RF link to "look like" a direct cat5e connection.  These RF links can have packet loss just like any "non-dtd" RF path.

In your example, these nodes are not all at the same site so they are probably using  a RF path with packet loss.


Here is some information from Eric's map:

WD6EBY-VC-SimiEast-3G (3.4GHz)
34.257059, -118.638
SSID: AREDN-20-v3
Linked to:
Node Name (cost to/from) distance bearing
WD6EBY-VC-SimiEast-2G (DTD) 0.11 mi 270°
WD6EBY-VC-SimiEast-5G (DTD) 0.23 mi 270°
KE6WEZ-VC-SimiWest-3G (1/1) 7.92 mi 264.7°
WD6EBY-VC-RasnowPk-N-Sector-5G (DTD) 16.48 mi 247.3°
WD6EBY-VC-Camarillo-Hills-SW-Sector-5G (DTD) 18.58 mi 269.7°
WD6EBY-VC-SouthMtn-E-Sector-2G (DTD) 23.22 mi 282°
WD6EBY-VC-SouthMtn-W-Sector-2G (DTD) 23.37 mi 281.9°
WD6EBY-VC-SouthMtn-W-Sector-5G (DTD) 23.43 mi 281.7°
WD6EBY-VC-Sulphur-ResourceServer (DTD) 31.7 mi 289.7°
WD6EBY-VC-SulphurMtn-to-Oxnard-5G (DTD) 31.76 mi 290.1°
N6FL-VC-OjaiEast-W-Sector-2G (DTD) 33.19 mi 294.7°
N6FL-VC-Ojai-East-Sector-5G (DTD) 33.21 mi 294.7°
N6FL-VC-Reeves-Rd-to-Black-Mtn-Link-5G (DTD) 33.23 mi 294.7°
AE6XE
AE6XE's picture
The support data shows 100%
The support data shows 100% for all the dtd links.   One exception from the the SimiEast-3G was showing 94.4%.     It would be expected that you have some packet loss over the part 15 links in use.  This would be expected.     You might remove the "option mode ether" entry in the olsr config file on al these nodes to have olsr configured accordingly.   However, If there are any mesh RF links between any of these nodes, then this would be problematic.   
w6bi
w6bi's picture
Thanks
Thanks, Joe.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer