High Uplink Queuing problem

(Read 9191 times)

Sep 3^rd, 2013 at 12:26pm

We have installed HN 9460 modem on one of our customer location using Hughesnet services. On Diagnostics hourly history modem shows uplink queuing problem and problem detected on LAN. Also it shows LAN traffic high. Can anyone help me in understanding what these parameters explain in HN9460 modem. Also the end user reports high number of CRC's on cisco routers interface connected to this HN 9460 modem. Once CRC's are there drops are observed and link disconnects. The client has created a VPN with HQ and it drops once CRC's are there.Kindly share your recommendations regarding the issue. Also if some one can share any documents containing details of diagnostic codes and errors related to HN systems that could be helpful in troubleshooting remote issues.

Martin

Sep 3^rd, 2013 at 8:39pm

Dear Mr. Eric

Can you help regarding this issue.

Martin

Sep 4^th, 2013 at 7:53am

I hope someone with HN or LAN experience will help.

My experience of CRC errors on Ethernet cables is that it is due to the Ethernet cables and interfaces themselves. Look for damaged plugs, damaged spring wires inside sockets, turn off auto speed detect 10/100 Mbit/s, set both ends the same speed, if using hub all devices must be same speed, avoid mains wiring, different earth systems, switching power on nearby devices, avoid cell phone and radios nearby.

Hope others will give you more ideas.

Eric

Sep 4^th, 2013 at 1:09pm

Can't profess any specific familiarity with the 9460, but I can at least agree with Eric. Sounds like a hardware problem in the signal path. First thing I recommend is to swap cabling. Disconnect and label both at the outdoor unit; LNB Out, Transmitter In. Go to the modem and swap the cables around. Move the Signal In to the Signal out. Then return to the outdoor equipment and connect the one you labeled LNB Out to the transmitter, and the one you labeled Transmitter In to the LNB.

This will not fix the problem, as you may have shifted a dodgy segment of cable to your TX side. But in doing so, you've begun to isolate the problem. From there, it's a matter of isolating it down to a particular segment or connector. If this step does not produce any change, the troubleshooting shifts to Rx hardware.

Also, it would help to know if you're connecting this 9460 via Ka-band or Ku-band.

//greg//

Sep 5^th, 2013 at 6:38am

Yes this 9460 Modem is running on Ka-band service. The modem transmit and receive is good and locked. All of sudden we start getting high number of CRC's on cisco router interface connected to HN9460. The vpn start to disconnect and modem analysis page we see LAN errors and high uplink queuing errors.

Sep 5^th, 2013 at 8:02am

Examine the Ethernet plugs on the cable between the HN and Cisco. Check the the plastic ridges between the contacts are not crushed. Check continuity Of all wires using ohms meter with sharp probes. Look in Ethernet sockets for stuck down spring wires. Waggle the cable to see if that causes CRC errors.

Are the CRC errors associated with anything happening nearby.

Use single cable and no hub.

Set Cisco config to fixed speed.

CRC errors on Ethernet cable are related to the network interfaces at each end, the sockets, plugs and cable.

Eric

Sep 5^th, 2013 at 1:41pm

I have already checked the cables and all at site. The same issue is being faced at 10 sites of same customer. Unable to understand wt is the possible cause. Customer is not able to identify LAN issues

Sep 6^th, 2013 at 8:25am

Maybe excessive packet length. For example if Cisco sends out Ethernet packet longer than what the HN is able to process this might well cause CRC error in the HN receive ethernet interface.

If anyone else can help please read the original message and add your suggestions.

Eric

Sep 7^th, 2013 at 6:46pm

Ok. My response addressed the topic, rather that the content. Sorry about that. So now I'm unsure what one has to do with the other. So how does "uplink queuing" factor against CRC errors? I ask, because uplink queuing issues are typically a provider problem, as opposed to CRC errors being a user issue.

Redundancy issues on the provider side are more likely related to FEC. This wouldn't by any chance be a simple matter of acronym selection would it? Or am I still missing something?

//greg//

Email me:eric@satsig.net

Disclaimer, Terms of Use and Privacy

Old Forum rules

Cookie policy.