Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[biohead]

Hi all,

Last time I tried to get some detailed line stats, I couldn't seem to get anything working. It may have been how my networking interface was setup, as it was in an XP VM, but I was fairly sure everything was indeed correct. Anyway, I've now bought one of those lovely little HP Microservers and set it up to run WHS2011 (may change that to WS2012 when its out) and would like to have another bash at it.

Would someone be as kind to do a write up please? I'm sure I'm not the only one who will benefit from one. I'll probably have another bash at it myself this week and do one if no-one else can.

Thanks

[tristan]

I'm planning to tidy up the scripts I've been using (all bash + sed + awk + gnuplot) in conjunction with "the" linestats script to monitor my line and make them more general and understandable to people who aren't me!

If the HP microserver can run cygwin then you're essentially good to go. I'll forward them on if you like?

[biohead]

If you could do that please that would be appreciated.

[biohead]

Just had another bash at it (completely missed the guide included in the zip file!  ) but wondering if I'm doing something daft. The error is that it's trying to write to a nonexistant pipe.

Do I have to change any of the routers default settings, or hook upto LAN2 instead of 1?

[Bald_Eagle1]

Are you trying to use the Windows or Linux scripts?

To get stats from the modem, a connection is required to the HG612's LAN2 port.

The easiest way to do this is to leave the modem's LAN1 port connected to the router as normal for internet access & connect a spare ROUTER LAN port back to the modem's LAN2 port to get the modem's stats.

The modem's default IP Address is 192.168.1.1 (this is how the scripts are set up initially, but can be easily edited to suit).

My router's IP address is 192.168.1.254 & no router settings were changed to get it all working together.

[biohead]

Using Windows (WHS2011).

Current setup is the FTTC modem downstairs at the master socket, then a stretch of Cat6 to connect it to the Apple Time Capsule router upstairs. It is using LAN1 on the modem. The WHS2011 box is sat right next to the time capsule, so I've plugged it in there as I can't run another cable down to the modem. My time capsule is setup for the entire network, including all DHCP reservations and port forwarding etc.

I can't really run another stretch of cat6 for the forseeable future, so I'm limited to hooking LAN2 upto a laptop temporarily. Is there a way around this yet?

[Bald_Eagle1]

Just a thought.......

Is there any reason why the router couldn't also be downstairs next to the modem, the long Cat6 cable then being used as a long "patch" cable from the router to the WHS2011 box upstairs?

[biohead]

Lack of space and power sockets are the main culprits, but I do also have some other wired devices in the router room too.

May take a bit of ingenuity this one...

[biohead]

Managed to quickly swap the patch cable around at both ends and get some current stats logged. Would've been nice to get it over a period of time but others in the house use the net connection frequently.



Those stats are on a line roughly 350-400m from the cab (length of footpath).

[Bald_Eagle1]

That doesn't look like too bad a connection you have there.

I can see from the pbParams data that you are connected to an ECI DSLAM.
Were you initially supplied with the ECI modem?

Interleaving (D:) looks a little high downstream at a depth of 1121 & OFF for upstream (depth of 1).
That probably explains the gap between attainable DS rate of 83156 Kbps & actual sync speed of 70789 Kbps.
Do you suffer from "noise" interference at all?
If you do get cabling things sorted, the ongoing graphs may provide an indication of that.

Lower down the Plink.log, you will see INP & delay values.
Linked with Interleaving depth & RSCorr error counts, these provide some connection stability to help to combat any "interference", but at the expense of some sync & throughput speed.


I see you are using the Windows version of the script(s).

You may wish to replace the current GRAPH6.BAT with the one attached to this message.

It produces smaller montages (can be posted as attachments in the Kitz forum), displays some of the data slightly differently & the output has been slightly "prettified".

Personally, I find the portrait version slightly easier to view in a forum thread.

[biohead]

Indeed I was supplied with an ECI modem (Which runs so much cooler than the HG!), so presumed I was on an ECI DSLAM. How could you tell from pbParams, as I can't see anything obvious which tells me so.

Noise wise, I'm not actually sure. All the cabling is underground so it's not too easy to trace. Back on ADSL though, I'd always have issues whenever myself or even a neighbour was using equipment such as electric lawnmowers or hedge trimmers.
Next time I have a bit of time, I've got a dirt cheap ddwrt wireless router somewhere - I'll possibly be able to set this up and that wireless network solely to get the stats... if it's possible. Not sure if I'll need the router in switch mode or router mode.

My INP and delay values down are 3.00 and 8.00 respectively.

[Bald_Eagle1]

Indeed I was supplied with an ECI modem (Which runs so much cooler than the HG!), so presumed I was on an ECI DSLAM. How could you tell from pbParams, as I can't see anything obvious which tells me so.

Discovery Phase pbParams band plan tones from a Huawei DSLAM:-
Discovery Phase (Initial) Band Plan
US: (0,95) (868,1207) (1972,2783)
DS: (32,859) (1216,1963) (2792,3959)


From an ECI DSLAM:-
Discovery Phase (Initial) Band Plan
US: (0,95) (880,1195) (1984,2771)
DS: (32,859) (1216,1959) (2792,4083)

HG612s run coolest when wall mounted or at least stood on end to let the heat escape through the vents a little easier.
Which version do you have?
3B's are the latest versions.

I have used B & 2B versions that only become slightly warm when operated stood up on my desk.

[biohead]

It is wall mounted, fortunately the mounts are the exact same as the ECI version - I wouldn't have put it past Openreach to have both modems with a different mount! Its a 2V model - in all fairness it's not massively hot, but theres a definite difference between it and the ECI.

Looking back at my stats and a few other peoples, I'm quite happy with it's performance for the distance it is from the cab.