Print

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

[William Grimsley]

I see, I think he needs to be more careful, he could create several LOS!

Oh, my god! 2 Upstream SES!

[jid]

If you keep watching your stats with so little spare SNR  errors from time to time including the SE kind are normal and acceptable as you're willing to leave the line in its slightly more unstable state.

I would say a few upstream errors are to be expected and shouldn't cause to much of an issue unless you're noticing problems with using the connection such as packet loss.

[William Grimsley]

Fair enough.

But, I'm not "willing to leave the line in its slightly more unstable state"!

[ejs]

But I thought you are leaving it at the high bandwidth and lower SNRM state, i.e. the more unstable state, rather than manually triggering a re-sync to go back to the lower bandwidth and SNRM around 6.

[William Grimsley]

That's the router's choice not mine, the line is stable in it's current state.

[ejs]

And you have chosen to leave it like that rather than switch it off for half an hour then back on again.

[William Grimsley]

Of course I have! Why would I want to experience a decrease in rate just for a few errors on the line? Seriously, I do not get some people on here.

[jid]

But, I'm not "willing to leave the line in its slightly more unstable state"!

Therefore reboot it and let it sync at a more decent speed.

Of course I have! Why would I want to experience a decrease in rate just for a few errors on the line? Seriously, I do not get some people on here.

William, you are the one who posted the update about errors on here and we have given advice. For example those upstream SES you saw are purely down to the SNR margin being so low and the router/modem struggling to "hear" the FTTC cabs signals. Its perfectly normal behaviour, but it is not expected behaviour as you are running your line in a way its not necessarily designed to be.

[William Grimsley]

Sorry, but I'm leaving it alone, I am not going to start rebooting the router when I get a power surge.

Anyway, the Downstream ES have increased slightly since the new customer was connected.

[jid]

Sorry, but I'm leaving it alone, I am not going to start rebooting the router when I get a power surge.

Anyway, the Downstream ES have increased slightly since the new customer was connected.

The errors will increase due to increased noise.

[William Grimsley]

Understood. I hope I haven't annoyed anyone again, I took some time away after I had a bit of a angry moment on here the other week. But, I just want to leave things alone now and see how they pan out.

[NewtronStar]

Its strange how some lines can cope with lower SNRM than others unlike my own for example was syncing at the max DS of 39998 Kbps 40/10 and the SNRM was 6.8dB during the day but once the evening RFI hit just after 20:30 the SNRM started to drop and then became spiky with CRC error burst and SES

The line gave up when the SNRM hit 4.0dB and a DLM resync was activated for this 1000 meter line so must be at the very limit even with a target margin of 6dB.

[William Grimsley]

It is interesting, and makes me think it's something to do with the cabinet more than anything, I can't understand why DLM would resync some lines at a SNR Margin of 4 dB and others not, especially if all the lines in question are experiencing low errors.

[ejs]

Are you certain it's really the DLM mechanisms triggering the re-sync? Or is it merely one or other end of the connection deciding it is unable to maintain the connection due to the changed conditions, and so forces the connection to retrain? I don't think any particular retrain reason number given really proves it was the will of the DLM.

[NewtronStar]

I definitely think the DLM for my cabinet is very sensitive even a change of SSFP from MK3 to MK2 to is enough to get 3 retrains by the DLM and when the SNRM has hit 4.0dB not one retrain comes in but 2 in under 24 hours so different to other on MDWS

As I said near the start of this thread you gain a lot of knowledge from your own lines behavior  over the years but one thing is for certain no two lines behave the same way even if the stats are identical.