Update on our progress: We went with the BH20 on our main backhaul, and one of the legs out of the Micro. The main backhaul has been having less than 100% on the uplink, sometimes as low as 80%. The other leg is reliably 99-100% on up and down. Jitter on both has been 7/8.
We have the main backhaul set at 25%DL. We used to run it at 50% but had gone to the 25% setting about a month ago, and it improved performance. We tried 75% and it was dismal. Here is the Link test on the main backhaul:
Stats for LUID: 2 Test Duration: 2 Pkt Length: 1522
Downlink RATE: 3686400 bps
Uplink RATE: 10334208 bps
Downlink Efficiency: 100 Percent
Max Downlink Index: 100
Actual Downlink Index: 100
Expected Frag Count: 14400
Actual Frag Count: 14400
Uplink Efficiency: 93 Percent
Max Uplink Index: 100
Actual Uplink Index: 93
Expected Frag Count: 40368
Actual Frag Count: 43248
Is this acceptable? I have seen it as high as 99% on the uplink, but it is usually high 80s to 93%. Performance seems OK…
Thanks! Carol and Frank
I always shoot for higher link efficiencies with backhauls. Here’s one of my BH20s:
Current Results Stats for LUID: 2 Test Duration: 10 Pkt Length: 1522
Downlink RATE: 6907084 bps
Uplink RATE: 6534758 bps
Aggregate RATE: 13441842 bps
Pkt Xmt (Act/Exp): 5621/0
Pkt Rcv (Act/Exp): 5309/0
Downlink Efficiency: 99 Percent
Downlink Index (Act/Max): 99/100
Frag Count (Act/Exp): 135896/134904
Uplink Efficiency: 99 Percent
Uplink Index (Act/Max): 99/100
Frag Count (Act/Exp): 127914/127632
Otherwise it looks like you’re probably good to go. You could probably tweak the performance with GPS synchronization and channel coordination, but other than that your setup seems to be performing about right.
Thanks for the info. We have GPS sync through the Micro, and we have good separation on all our frequencies coming out of the Micro. Will continue to watch the link to make sure it does not degrade further. We do not have shielded ethernet at this time, are thinking of changing over if that would help…
Thanks! Carol
Are you seeing errors on the Ethernet interface of the radio, or if your switch is managed on the switch itself?
I just checked the ethernet stats and there are errors on the main BH20 backhaul leg (2.46 one side reflectored), I confess I don’t know the significance though. Here are the errors on the Master:
inoctects Count 2028542147
inucastpkts Count 46351010
Innucastpkts Count 217561
indiscards Count 0
inerrors Count 93
inunknownprotos Count 0
outoctets Count 1516161015
outucastpktsCount 59636702
outnucastpkts Count 292100
outdiscards Count 0
outerrors Count 0
RxBabErr 1
TxHbErr 0
EthBusErr 0
CRCError 91
RxOverrun 2
LateCollision 0
RetransLimitExp 0
TxUnderrun 0
CarSenseLost 0
and the Slave, which are fewer:
inoctects Count 1532385410
inucastpkts Count 59629379
Innucastpkts Count 292037
indiscards Count 0
inerrors Count 4
inunknownprotos Count 0
outoctets Count 1949162819
outucastpktsCount 46336149
outnucastpkts Count 217257
outdiscards Count 0
outerrors Count 0
RxBabErr 0
TxHbErr 0
EthBusErr 0
CRCError 0
RxOverrun 4
LateCollision 0
RetransLimitExp 0
TxUnderrun 0
CarSenseLost 0
The other BH20 leg, from Micro to remote site (3.86 miles, one side reflectored) has no errors.
Thanks for any help on this.
Carol
93 errors out of 2 billion octets is ok.
What’s up with the CRC errors?
Wow - didn’t see those, good catch.
Possibly a bad cable, connector, or negotiation mismatch (flapping).
Yup
Gut feeling says the same.
Check your cabling and if you are plugged into a managed switch force it’s speed and duplex on the port.