internet keeps dropping after 10pm

[#1]

went to config rx adjust axis auto and unclicked then set to 15 is that correct ?

also chapped guy upstairs , his is working fine and all he is using is a FIRESTICK in his tv 

[#2]

Nope, its the Tx Noise graph that needed changing

[#3]

could some 1 check these settings, im getting confused again thanx

[#4]

the latest graph , been running 2 hours 

thanx

is it possible to get it to capture graphs auotomatically thro the night 

cheers

[#5]

Even a graph of just the upstream sync speed would be useful, or some more stats.

Looking at this latest Rx Noise Margin graph, it should now be obvious why I did not do a more detailed analysis of the few blips Townman was interested in earlier - because they are clearly insignificant when compared to the disconnections on the latest graph.

[#6]

---

@glesgajohn wrote:

is it possible to get it to capture graphs automatically thro the night

I thought my set up guide covered this?  Right click on any graph you want to automatically capture and select "Scheduled Capture [always]" and "Make available to webserver" - the first will snap-shot the graphs every 2 hours, the second provides a "current snap shot" to make the current view immediately available.

All graphs are (by default) stored in the router stats folder.

@ejs - looking at these plots, I suspect more than one source of RFI.

@Chris - any chance of a look see here please?  Especially see #52 & #55.

---

[#7]

---

@Townman wrote:

looking at these plots, I suspect more than one source of RFI.

---

---

Or a faulty line. Get the line fixed, and all the things that you think are sources of REIN or RFI will no longer have any significant affect on the line.

[#8]

hi guys thanks for getting back to me, i left router stats running all night (still running ) these are some pics thanks

[#9]

@ejs - the problem is that the engineers do not seem to be able to find anything to fix - which is why I've aske for @CRT review.

@glesgajohn - A word of caution ... I have just been reminded that RS (certainly V6.8k and below) can get its knickers elastic twisted and report no synch (connection down) when it is not.  The tell-tale sign is that the summary tab contains garbage.  Just click the stop and start recording buttons and plotting will resume again.

[#10]

@glesgajohn,

Can you please change the y axis maximums back to the settings I originally advised.  Sometimes one has to be careful that the measuring tools do not give misleading results.  At sometime of the early hours RS misreported the Tx Noise level - it cannot of have been in the 1,000sdB.

The consequence of setting the graph axis to "automatic" is that an aberration measurement resets the graph maximum (to 1,000s) and the real data (single digits) can no longer be discerned.

The Rx Noise graphs are at least (if not more) important than the Tx Noise graphs - if you want to post Tx, please also post Rx.  You will find that setting automatic (scheduled) capture useful.

If you fancy poking, setting an alarm for when Rx / Tx synch is zero will alert you to both disconnects and false readings (if) version 6.9 suffers the same issue as 6.8k I drew to your attention a few minutes ago.

[#11]

also they sent me a new router, y i dont know

[#12]

@glesgajohn

All of the DS plots are ZERO - looks like version 6.9 does suffer the same problems as 6.8k.  Here are the alert settings I use.

[#13]

@Townman

I thought a graph of the sync speeds would be useful, upstream particularly, to see when the sync speed changes. I can't understand why you think the upstream is not important, when we know the upstream keeps dropping to abnormally low sync speeds here, and normally the upstream is more reliable than the downstream. So if it's affecting the upstream, whatever it is, it must be bad.

As far as I can tell, which is not that far because the graphs aren't quite clear, it looks like the connection is dropping, and re-connecting at much the same downstream speed as it was before. This does not fit with the REIN hypothesis, if it were REIN, I'd expect it to re-connect at a lower downstream speed if it re-connects when the REIN is present, if the REIN were affecting the downstream. So if the problem really is some due to interference and the line itself is OK, which I still don't think is likely, it'd be more like SHINE, short bursts of interference which knock out the connection, which have gone by the time it re-connects.

[#14]

@ejs,

I guess that we all "play" from our experience - thank you for shining light on your focus on the US metrics - what you have expanded on now makes sense and helps us sing from the same hymn sheet.

@glesgajohn

Please post the following RS graphs (ideally as a time related group)...

• Rx Noise (this will alos have the Rx synch speed)
• Tx Noise
• Tx Synch speed

Note to all, we need to be careful not to take all zero plots as being indicative of a synch drop - attached is my graph from this evening ... the apparent drops is RS loosing the plot (please forgive the pun! ) not a synch drop reconnecting at the same speed.

[#15]

is this them ..i must be doing something wrong, it doesnt seem to be capturing images