Print

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

[BritBrat]

Edit:- P.S. how are you gathering data for that chart?

"JD Auto SpeedTester" put it into Google UK with the "" and you will find it.

I'm using routerstats lite running on a remote PC and www.logmein to monitor the stats remotely.
Here's one I prepared earlier !

I am doing the same thing with the speed tester but using RemotelyAnywhere.

However the relatively gentle noise slope downwards over many hours doesn't seem to be directly related to a piece of plant being switched on.
It's almost as so the modem is gradually getting hot and bothered and running out of breath - i.e. the descrimination logic is becoming overloaded.

Did you say you had another router?

If you do try the other to rule out the router as faulty if you get the same results.

Do you have many portforward/firewall rules in the 834GT?

If so are they being loged, turn off loging.

[MollyCoddle]

@waltergmw

Sorry W, I only see a flat 4100k connection there. I did not manage to capture the re-sync last night.

@BritBrat

Thanks for the tip

Thanks both, MC

Edit:- BritBrat you little editer 

>>If you do try the other to rule out the router as faulty if you get the same results.
A.---A USRobotics, I've tried swapping many times, I get the same SNRM action as those charts posted. The only difference between them is the way the USR does not bring PPP back up after one of my disputed re-sync events.

>>Do you have many portforward/firewall rules in the 834GT?
A.---No rules, I took them out weeks ago, just in case.

[BritBrat]

I think you need to try the other router again directly into the test socket as you have it now.

As you have reduced the noise on the line.

[MollyCoddle]

>>I think you need to try the other router again directly into the test socket as you have it now.
A.---Done, I have attached the SNRM and CRC for yesterday/overnight -- the re-sync was caught on the GT's log:-

Wed, 2009-03-11 01:37:47 - LCP down.
Wed, 2009-03-11 01:37:55 - Initialize LCP.
Wed, 2009-03-11 01:37:55 - LCP is allowed to come up.
Wed, 2009-03-11 01:38:02 - CHAP authentication success

Also attached the freq. response before the router swap, however I cannot get Routerstats or DMT to graph the freq. response (with the USR) so until I get the Netgear I can't post a graph of this. Noise and sync graphing are OK.

Feedback much appreciated,  MC

[attachment deleted by admin]

[attachment deleted by admin]

[BritBrat]

That looks a lot better.

I would leave it now for a day or two and see what happens.

In the mean time start pluging phone equipment in one at a time to see if you get the noise back and isolate the faulty device.

It may have been the router faulty and causing the noise.

however I cannot get Routerstats or DMT to graph the freq. response (with the USR) so until I get the Netgear I can't post a graph of this. Noise and sync graphing are OK.

Try routerstats lite.

Although it does look like it may not do that USR:



[attachment deleted by admin]

[MollyCoddle]

>>That looks a lot better.

A.----Those charts are for the Netgear. I inserted the USR 9ish today.

>>In the mean time start pluging phone equipment in one at a time to see if you get the noise back and isolate the faulty device.

A.----Just in case I have confused the issue, the helicopter dissappeared before I removed the telephone and microfilter out of circuit. This had surprised me as this noise had been present at one volume or another since Sept08.

Anyway, I will stick to the plan and introduce things after I have left the USR for a few days, as it only went in Wed a.m.

I will also try the lite version.

Thanks again, MC

[MollyCoddle]

Overnight Update.

There have been 're-syncless' nights previously, however the attached charts of HEC and ES indicate a router resilience to errors this night.

Chart notes:-
1. The SNRM chart is a different time-scale to the HEC and ES charts.
2. The SNRM chart shows a re-sync event early in the day - that is the swapping back to the USR.
3. The HEC and ES charts exhibit a change roughly a third of a way through - this is me realising that RouterStats can plot for increments.
4. All sample once per minuite.

Even so, are not these errors rather high?

MC

[attachment deleted by admin]