flourescent lights, noise interference part 2

here is a pic of my noise margin graph with SRA disabled. Obviously the line would be more likely to drop, so I deliberatly had coding gain on 0, surprisingly is still holding tho.

target snrm 7, I know is rarely near 7db

Noise Margin - No SRA

2nd graph it appears the 2nd interference isnt present so can see the 1st interference constantly coming and going.

Noise Margin 2nd day - No SRA

Edited by Chrysalis (Thu 27-May-10 19:54:25)

Just thought i would link to the first part:
Link to Part 1


Well not allot in the way of progress,But it's looking like i may have to get a billion router, so i can change the snrm, to suite conditions, ie, raise it during periods the shop lights are able to affect my connection and lower it when it has ceased,as the dslam at my exchange seems to stick to g.dmt mode, so i'm unable to re-sync using other modes, unless my isp manually changes mode, and i doubt that they would want to do that every day, , But switch of modes the other day has changed things a little in my favor, i think this because it got a slightly higher sync than it did previously,and the level of packet loss & crc errors were lower, so the throughput was more constant too, but not as good as when the lights in that shop are switched off,

the situation reversed without SRA.

the daytime on both days for packetloss is cleaner but not 100% clean.

however the large loss of snr to the night causes significant packetloss at night. This is a problem I had for many years before I started using SRA having to get the line stable for day and night without syncing really low. The sync is 5.3mbit.

Overall tho it worked better than expected on that little experiment.

with the billions you will have 2 methods of changing sync speed, both the snrm overide and coding gain settings, coding gain can be adjusted in the gui so is easier.

Coding gain, how does this work, similar to the way snr does?
Well for some reason it appears that i has dropout earlier, strange, routerstats logs show it,so does the router's log, and it's shown on the tbbqm too, but the sync,snr graphs do not show it,odd to say the least, but something has changed, as the throughput seems to have reverted back to how it was before, not good,

From router stats log, it would appear that my ip address became inactive ? as for whatever the reason it changed from the usual static ip, to the 0.0.0.0 and back to my same ip some 5 mins later ????

Edited by tommy45 (Fri 28-May-10 16:39:06)

not that sure how it works.

I believe it is more agressive on the signal utilisation similiar to how interleaving increases sync.

the simple bit is low coding gain = low sync speed and high coding gain = high sync speed, it doesnt affect the noise margin. However higher coding gain will also increase chance of crc errors and the sync dropping.

I have been checking over things again today looking for any clues of any other reasons for this E M I Possibly a new discovery
I think that even the shop's alarm is showing up on the ping plotter charts, but oddly enough i don't seem to be able to see it after closing of the shop tonight as i was walking past the shop on my way home, i noticed that the owner was in the shop but had all but one of the shops lights off, the light that was on was the office light, got home the majority of error's had stopped and so had virtually all the pkt loss,12HR time span graph and as an example of how many errors this degree of packet loss generate in 1 hr

From 15:02 to 16:02 there was 124542 corrected crc errors & 3599 uncorrected erroneous seconds but as i write this the in the past hour 18:45 till 19:45

249 corrected crc errors 175 uncorrected erroneous seconds & 2 serious erroneous seconds , So when this shop aint got its lights on my connection is not producing more than it should do given that it is on fastpath

if you do get a billion 7402 and want proper snrm and speed graphs I can graph it remotely for you on a cacti server no problem, just pm me if you want to do this after you get the router. The error count stick to using tbbq for that. Routerstats wont work on the router.

Does the router (billion) not have a built in counter for viewing the crc's erroneous secs ect? the ping plotter and tbbqm only really show packet loss, and by a percentage, where as being able to count the errors that build over a say 15mins time frame gives a reasonable idea how bad or good things are Well no packet loss or large amount of errors yesterday or today ,apart from the strange happenings earlier today all has been ok, thank's for the offer, i will probably take you up on it once i get the router ,

Edited by tommy45 (Mon 31-May-10 16:18:44)

yes you can check it manually, but only via telnet. The gui counter seems to check the wrong variable.

Well after a great 2 days of a EMI free Internet connection Sat night through to this morning when Those lights got switched on,as a test i made sure that i was downloading a test file,which was maxing out my connection, then suddenly the data stream being received stopped completely due to the switch on of the lights. it after a few mins started tricking though again, at a few hundred kbps ,
Results of errors recorded yesterday after almost 9hours and plenty of data uploaded and downloaded totaling nearly 20gb,

3760184----138922------101717--ppp0=5days09:39:56
3767214----140785-----101806--ppp0=5days18:25:50--=
Total Corrected CRC's=7030 Uncorrected:ES-1863 SES: 89

but from 08:47 to 19:04 today =1371383 Corrected CRC's , 36640 Uncorrected ES,& 29470 SES
and todays Tbbqm
My ping And for some reason the router hasn't reset it's ppp for over 6days, so maybe it was down to the exchange kit or it had been set to do it after a pre determined time ?

he seems to be in that shop alot outside of working hours.

He certainly does,particularly after he shuts shop for the day he typically will be in the shop going through paperwork(seeing how much profit he can/has made)Ka-ching ,with all or most of the lights on including some in the windows needlessly burning away some of that precious profit and more importantly stopping me using my adsl connection,lol
He may be due to go an his holidays soon (if he ain't too tight fisted) maybe then i'll get the opportunity to see how the internal telephone extensions are routed as one or two of the workers seem chatty enough,And if i was able to identify a problem and the cable could not be simply re-routed in a not so obvious way, then i could substitute the cable with a shielded type ,lol, that's assuming things are straight forward enough,

Edited by tommy45 (Wed 02-Jun-10 14:19:29)

no updates?

here I still have interference however I am now syncing low and putting up with it.

Interesting to note my upload is defenitly also affected, not by a huge amount but its down by about 50kbit and is a consistent loss.

Not anything to update, still more or less the same,except the downstream SNR seems to be affected in that it will decrease by 1db sometimes a bit more at the times the interference is present, last night well just after 12 my connection went down for around 15 mins, dunno why, and i lost around 100kbits on my d/load sync,i'm sure i'll get it back though eventually,lol packet loss today has been less than it was yesterday, but still has the adverse effect on throughput the upload speed seems to be affected on speedtest results, but if i use an ftp or the rs upload tool it still seems to max it out,except on the rare occasion of 50%+ packet loss
I'm waiting for the shop owner to go off on his holidays, then i can get a proper look at the wiring setup(internal phone extensions) i have been told that the master socket is positioned next to the mains supply fuse box or meter , and that the shop wiring is a mess never seen an electrician for years,though how true this is i don't know, but the bt master socket surely should never be positioned near the incoming mains electric supply,???

Edited by tommy45 (Tue 15-Jun-10 19:27:13)

Today there have been some development's in that ii had several losses of sync or ppp sessions, after each,the d/s sync reduced so at the minute i have lost nearly 1 mbit the other changes are that the d/s seems to be a static 4db and the u/s snr appears to of been lowered by 1db so i see 5db instead of 6db (but should be 4db)
but on the plus side of things the packet loss and the high error rate seem to be gone, so throughput also not as fast as before, is back to normal, but however it could be just some co-incidence ,only time will tell,
But i do re call a brief period after my line fault was fixed it was a tuesday ,i suffered next to no problems for the rest of that week, but the following Monday i did ? so i'm not convinced as yet that this is fixed

Edited by tommy45 (Thu 17-Jun-10 15:38:02)

Interesting thing on my line I just noticed and probably explains the weaker upload sync, the attenuation has gone up by about 5db, this is consistent.

attenuation is supposed to be a static value with maybe very small variances between syncs, a 5db move as I understand it should not happen unless something physically changes.

An HR fault could conceivably cause attenuation to vary.

talking of attenuation the d/s on my connection has decreased by 0.5db yesterday following the number of re-syncs, but i know the netgear reports a higher value, as the real attenuation is 28db, i only know this cause i asked dan,

I may of found a way of avoiding this noise/interference during the opening times of the shop, that is wait until it causes the downstream snr to decrease below the set target of 4db, then re boot the router, it syncs at a slightly lower rate, then to re gain the higher rate repeat once the snr rises once the interference stops, so it is similar to adjusting the snr values , worked today, well nearly, still around 200kbps shy of the highest sync achieved, the only thing that i am annoyed about is the loss of ppp i have been getting recently it dropped early this morning as i was playing a game on line, i think uko had some probs as later the web page and e-mail server were down, for a while

Edited by tommy45 (Fri 18-Jun-10 21:16:52)

I have suspected HR for months, even years. But cant get it investigated.

openreach/BTw is now of the mindset that everyone's equipment or wiring is too blame and have got too greedy on callouts which has scared most isp's off.

looking at my graphs the attenuation increased march april time. It has always jumped around but before was around 49-52db range and is now 54-57 range.

Edited by Chrysalis (Fri 18-Jun-10 21:41:28)

the ppp thing is a netgear problem.

The cause could be as a result of another line being commissioned or repaired, i can remember joking with the engineer he just said got to keep us busy somehow,lol

indeed, I remember robertos noticing his sync dropping so he went to his cabinet and found that every time the engineer even slightly touched the cables it caused a loss of sync. That is how sensitive and fragile things are.

Well what ever it is it's a pain in the a**

I have my upload back to full speed again but had to give it more juice. Using the -1db power instead of -2db power.

In reply to a post by Chrysalis:

I have suspected HR for months, even years. But cant get it investigated.....

They can be notoriously difficult to track down. I used to absolutely loathe intermittent electronic faults. They almost always generated customer animosity.

Thats one part of my connection that seems resilient never changes, i think if they did annex m i would not have issues getting the max sync or somewhere close to it ,here are some graphs of what the rx crc's do normally when no interference from shop RX Serious errored secondsRX corrected crc errorsRX uncorrected errored seconds the errors are set to 15min intervals

well I spoke to my BT guy, it seems another 50+ lines on my cabinet failed around the time my current issues started and people were pair swapped, he reckons I am one of them. I told him I noticed no real downtime before the recent noise issues started but obviously the change in attenuation indicates something has happened.

As soon as I report this to ukonline the first thing that will happen is my profile is reset and SRA gone. So this wont be chased up officially given now that ukonline is now my backup connection with a planned cease.

Edited by Chrysalis (Sat 26-Jun-10 11:47:24)

Not good,but if your line was to fail completely or develop a voice fault then possibly you may get another pair swap, and as a result the adsl could be better as a result?

just goes to show just how vulnerable the connection (line) we rent is ,if it isn't a hr fault then you get a bumbling bt or engineer swapping pairs in a cab at random,and fail to check if those other lines are adsl active or not Also they don't even tell you about, or ask you prior to that if your having any issues, with your line,

As far as my connection goes i have seen a improvement in throughput speeds during the daytime( shop open) and every now and then i do get my max speed, this is only once thew serious erroneous seconds stop being generated or there are very low in numbers , there is still the noise or whatever it is that will cause my downstream snr to drop from a static 4.0 DB to 3,9-2.8DB, this is as a direct result of the initial switch on of the lights in that shop, sometimes it will fully recover during the day(shop open) other times it wont, the packet loss levels have not been as high over the past week,and at times there has been none(from my PC to the exchange) but there is still some picked up by the tbbqm
Maybe something has been done by bt,or easynet? or it's just co incidental something that was used by the shop isn't being used now? but it's certainly performing better now than it has done in past months

Edited by tommy45 (Sat 26-Jun-10 11:59:55)

I have no doubt if voice failed the fault would be fixed, the difference between BT's attitude to voice and adsl faults is night and day.

Just before I joined ukonline my voice died very late on a friday, I reported it on saturday morning and a BT engineer showed up at my door on the sunday, was fixed within 30 minutes.

I keep hearing reports that btor work on weekends, but a while back when my line died,when i rang em to find out when bt where gonna fix it i was told probably would be the following week, as it was a bank holiday Tuesday would be the earliest , so i said why not saturday/sunday, i got told that or don't work sat/sun or bank holidays? but at the time that exchange down south was down the one that set on fire and then was flooded, but why that would affect the or services from the north west is a bit rich to say the least

voice will be prioritised for obvious reasons. On top of this I expect other priority levels exist as well.

this latest noise (the one that requires me to sync very low and occurs at weekends) is not HR related by the look of it.

yesterday I had a almost mirror image of packetloss on my VM and ukonline graphs. Also I have been having a few suspect VM dropouts on healthy stats, which follow packet loss. Which I had previous packetloss graphs show signs of matching, yesterdays is very conclusive.

It would seem this interference which has been forcing me to have 20+db snr margins on the adsl to keep it useable is also affecting my cable. What kind of devices you reckon are capable of also interfering cable? given cable frequency ranges does this new info narrow it down?

In reply to a post by Chrysalis:

voice will be prioritised for obvious reasons. On top of this I expect other priority levels exist as well.

It was a pstn fault ,had no line it was totally dead

Well just to update you, the level of packet loss seems to of certainly overall become a lot less, apart from the random 60-80% surge of it, the serious errors are also virtually non at all or only a small amount, so a near max throughput can be achieved, so getting a accurate speedtest result is still hit or miss, but if i download a test file using the IE addon IE pro which has a download manager that you can have upto 10 connections to the source fie file i get a near full speed download,but as a result of the errors,it won't max out the connection fully, i see this by monitoring the ping response times, during shop opening times it is very low compared to when shop is shut and the owner ain't sat in there writing his book, or uploading porn,lol

what is strange is that once the workers have gone home, i see an increase of errors, including of serious errors and throughput slows to an erratic crawl and sometimes the packet loss will increase to a really high amount and no internet rec or send of data is possible but stays in sync

Edited by tommy45 (Tue 29-Jun-10 18:08:28)

Just had a thought, could this packet loss &errors that cripple my rx throughput be due to the Internet connection in the shop such as cross talk? i doubt that it will be the case but is it possible & is it likely or not,

cross talk is possible yes. cross talk depending on severity can cause major issues.

The reason thought of that as a possible cause or part of it, is that the shop owner has a web site, and frequently updates it, also he receives a lot of e-mail business related, all i was thinking is that he may have some high speed upload connection and that when sending data is creating this crosstalk, i'm just trying to explore other possibilities,with a bit of luck i should be able to have a look at how the telephone wiring is set up ,there maybe something obvious as a telephone cable being next to a mains power cord

Well today he (shop owner ) stayed back for over 1hr 30mins after closing time
causing crc errors and packet loss so after i became really [censored] about it i decided to go and ask him how much longer he intended staying in the shop for as i wanted to use my bb !!!! the shop was in darkness no lights(groundfloor) and the van was parked in it's usual spot so he definitely in there , any way i came home and all interference stopped prior to me checking routerstats ect ,he knows he is causing me problems,and i bet he also has a good idea what is the actual cause, but doesn't want to change anything to stop it, of course it could be just a co incidence that it suddenly stops as i go near the shop,but i doubt it, Today highest level of pkt loss for nearly 2weeks, plus there is some at a earlier time, 100% the thin part, was i think a easynet dns issue as everything data flow stopped pages would not load could not download anything, but remained connected or appeared to, then there is a wider part, that i think is down to tbb.com as r stats was able to ping the bbc without issue at this time

Edited by tommy45 (Tue 06-Jul-10 19:55:18)

Just a small update which could mean something or not, but ,the owner is away holidays, so i asked the guy who he left in charge if i could have a quick look around at the telephone wiring extensions, and that i wasn't going to touch etc anything, just basically do a visual survey to see if anything obvious stood out such as a power cable running parallel with a phone extension or similar, but he would not let me, saying that i would have to see the boss, i get the impression that they know what is the cause,and don't for whatever the reason want anything changing or me to find out what it is, i have spoken to a bt engineer recently about it he was bb trained too, he had never heard of it before,but said that they may be up to no good, whatever he meant by that I'm not sure, maybe they have something they shouldn't have plugged into the phone line? the other thing i have noticed is that the downstream snr level no longer seems to be completely recovering to the set 4db always 3 db or less even after shop emi has ceased and is now going as low as 1.75db no disconnects Well i'm sick of this now, yet another hour past shop opening hours today before i'm able to use my bb fully i need to find a way to convince the shop keeper that his lights are the cause of my problem, but how? ideas welcome

Edited by tommy45 (Mon 12-Jul-10 19:34:41)

Well for whatever the reason is things seem to be getting worse.i am getting upto 60% packet loss at times the downstream snr has dipped to 1.2 at times
currently the connection is generating around 900 erroneous seconds every 15mins
and 45,000 crc errors every 15 mins (all errors are downstream) and even browsing is at a snails pace or fails completely

So as the snr has crept in a downwards direction from the set 4db to at best around 2.5db i'm not wanting to re sync as i will loose quite a bit of sync speed and the problem will still be present

I've only 'dipped in' to your thread but thought I would offer this after reading your last post

If you are getting so many errors you may find that your actual 'throughput' is better with a lower sync when not tweaking your Target SNR ! I wouldn't only focus on your actual sync - if your connection runs cleaner with a lower sync your actual broadband experience may be better...

If i was able to tweak the snr level via the router i would go for a lower sync during the times that this outside interference is occurring and the then again raise it for all other times,using snrm but i can't as my router don't support that, i would need a more expensive billion router to do that,
The snr used to go back up after the interference stopped each day, but for some reason although it will increase a little won't return to the set value again,

but even syncing at 8mb on adsl2+ i still see packet loss and errors, where as adsl g.dmt i don't this interference is what should be cut down not my sync speed, it's really annoying that the two companies that you pay so you can get a broadband service bt isp, are not interested isp just offer to mask the problem, in which i would end up with a service that would be unacceptable to me and bt don't give a dam about their network being degraded by this and other interference but expect payment for this broken service

Sorry when I read "set 4db" that looked as if your had tweaked from the default 6db Target!?

NB. You don't need an expensive router to tweak - a £10 eBay DG834GT or similar will do...

If my isp used dslams at the exchange that were very compatible with the broadcom chip set then i would already have one, but as my isp uses a conexant based dslam s I bought a DG834G-V5(conexant chipset) instead hence the 19mbit sync i get at 4db with a attenuation of 28-29db (isp default snr is 5db) if i used a broadcom or other router i would be looking at 14-15mbit tops(snr at 3db) also with the packet loss & errors so this would defeat the object , like i said the billion will allow the snr to changed via telnet but they are expensive,and as for e bay second hand routers,mmm not really a reliable way as you don't know that what you are buying actually is in a fully working condition or not,at least with new you get a the warranty

Edited by tommy45 (Thu 15-Jul-10 20:34:14)

Post deleted by tommy45

I was/i'm getting the 45-50%+ level of packet loss again, but found a way of reducing it,or it appears to reduce, that is to put 10 files in the rapiduploader tool and run something like 5 or 10 instances of it, maxes out the upload and the packet loss gets reduced or appears to do,???
the other thing that i have been giving some thoughts to is the fact that during a quiet line test i am able to hear a constent humming.buzzing noise similar to that of which a cooling fan would make,it has always done this, but i would also say that the interference from the shop has always been there as well . My thoughts are bt main cable or branch of to shop ect too close to mains cable in street, hence the constant noise in the background, but is this regarded as a fault by bt?

Well the problem's i have been experiencing with my line seem to of subsided recently, for the past 2 weeks i have little or no packet loss,and only a small number of serious error seconds, sometimes the Dslam will cause a re-sync and it will sync at a lower speed,at first i just left it to it's own ,after around 6days the Dslam caused a re-sync and it was around the normal sync 18 mbps+ @ the 4 db snrm and i was surprised as throughput was more or less normal,no 10-20%+ packet loss ,

observations: at a lower sync when downloading a test file in order to max out the connection, the ping increases to 200 ms+ where as with a higher sync it may only reach 200 ms is this a normal thing?

and the other thing i noticed is the crc error and erroneous seconds rate is higher with the higher sync too, but these don't seem to be having a detrimental impact on throughput,except maybe when using a single stream to connect, throughput can at times fluctuate wildly,there is also some disturbance of the downstream snr as it can cause it to drop off from the almost static 4 db by 0.5 db at times,but it is recovering when the source stops emitting its EMI, which i think is down to some machinery inside a engineering works across the road

so it's not 100% fixed but it's a hell of a lot better than it was a few months back

In reply to a post by tommy45:

... emitting its EMI, which i think is down to some machinery inside a engineering works across the road...

What makes you think it's still not the shop you've been previously mentioning?!

In reply to a post by b4dger:

In reply to a post by tommy45:

... emitting its EMI, which i think is down to some machinery inside a engineering works across the road...

What makes you think it's still not the shop you've been previously mentioning?!

Because it has occurred after the shop has closed as well as during the operating hours of the shop, there also could be multiple sources at several locations at play,but the packet loss,was the shop for sure,the engineering place is 24hrs 6 days a week or can be Dependant on work they have, i also at times can hear faint muffled noised from said property along with vibrations that are transmitted through the ground that i can hear when in bed(from machinery) the Emi or whatever affects the SNR is or seems to be random, so unable to pin it down, but it did only happen during shop hours, but it would affect it more sometimes by 1db or more for short times (while something was in use) there is also a new tanning studio opened recently,that has late night opening that is near to the shop that was responsible for the issues or some of them

Edited by tommy45 (Thu 21-Oct-10 18:36:09)

ok so some time on a new isp and some new thoughts on this.

this problem started probably about a month or so after dan at ukonline pushed my upstream snrm down to 4db and uncapped it.

on my new isp I am seeing what I consider a lot of upstream crc errors, I am getting more crc errors than on my downstream which I find bizzare.

they not going up on a regular basis, eg. on my downstream I may get a crc once every 30 seconds or so. Yet I am yet to witness the upstream count go up, it seems to jump up every now and then so eg. it can be at 100, I go away for a bit and then see its jumped up to 160, I then watch it for an hour and it doesnt budge, go away for 10 minutes and its jumped to 220, and so on.

To me its a sign I have a fault on the line that I am getting these upstream errors (864 sync 6db snrm).

I am also diagnosing my 2nd interference as silent noise, by silent noise I mean there is noise on the line that causes crc errors but the router doesnt detect it so the snrm doesnt fluctuate. I know its there when the crc error rate explodes, fixable by bumping the margin 2-3db. So still weird issues but nowhere near as severe now with the lower upstream sync.

Edited by Chrysalis (Wed 19-Jan-11 18:35:40)

Sounds like the router may be faulty.

That's interesting, as i to have a new isp. and have Annex M i get between 1.85 -1.7 mbit upload on a 3db snr fast path, the error's on my upstream seem higher than the downstream, but my upstream snr will fluctuate,where as the downstream is more or less flat-line's on 3db or a bit higher,
The interference can still be seen on the tbbqm, and sometimes but not always on ping plotter as packet loss,+jitter, but doesn't seem to affect throughput as much, in testing i set the snr to 1db got a 19.5mbit sync and 1.4mbit upstream on annex a as it doesn't sync back on annex m if i tweak the snr, the good thing is my throughput is a lot better that used to be at one point,
The other observation i can report is the latency seems to reversed,
when on uko if i maxed out my downstream my latency would be around 180-200ms whilst downloading (to bbc.co.uk) and the upstream would be less always, however now on be it's the other way around upstream is 400ms+ whilst downstream is 40-50ms regardless of adsl mode ?
which i think is a bit odd,
whilst on that subject of latency another strange thing is that apart from the bbc my pings to tbb are higher than they used to be, bbc i get 13ms as i used to on uko, my ping to the be exchange kit is lower now only by 3-4ms but not constant & will fluctuate to old uko value,of 10-12ms
Yesterday i notice that pings to tbb and other sites inc uts.ubi.com (gaming server) went lower for several hours then returned to the usual higher value lower was co-incidentally just less than the normal times i would get on uko

/usr/sbin/adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 8000
Channel: FAST, Upstream rate = 1788 Kbps, Downstream rate = 16780 Kbps
Link Power State: L0
Mode: ADSL2+ AnnexM EU-56
Channel: Fast
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.3 3.6
Attn(dB): 33.5 16.3
Pwr(dBm): 18.4 13.1
Max(Kbps): 19850 1750
Rate (Kbps): 16780 1788
G.dmt framing
K: 255(0) 51
R: 0 0
S: 1 1
D: 1 1
ADSL2 framing
MSGc: 55 16
B: 254 50
M: 1 1
T: 2 3
R: 0 0
S: 0.5249 0.9807
L: 3886 416
D: 1 1
Counters
SF: 338147 338315
SFErr: 959 277293
RS: 0 0
RSCorr: 0 0
RSUnCorr: 0 0

HEC: 337 82073
OCD: 0 0
LCD: 0 0
Total Cells: 214246282 2967109221
Data Cells: 16652395 105300039
Drop Cells: 0
Bit Errors: 0 8942386

ES: 808 197809
SES: 0 0
UAS: 49 1768
AS: 5415

INP: 0.00 0.00
PER: 16.01 16.18
delay: 0.13 0.24
OR: 30.47 10.87

Bitswap: 1641 839

Total time = 1 hours 31 min 21 sec
SF = 338147
CRC = 959
LOS = 0
LOF = 0
ES = 808
Latest 1 day time = 1 hours 31 min 21 sec
SF = 338147
CRC = 959
LOS = 0
LOF = 0
ES = 808
Latest 15 minutes time = 1 min 21 sec
SF = 5074
CRC = 20
LOS = 0
LOF = 0
ES = 19
Previous 15 minutes time = 15 min 0 sec
SF = 56253
CRC = 200
LOS = 0
LOF = 0
ES = 168
Previous 1 day time = 0 sec
SF = 0
CRC = 0
LOS = 0
LOF = 0
ES = 0
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF = 56191
CRC = 184
LOS = 0
LOF = 0
ES = 152
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF = 56191
CRC = 125
LOS = 0
LOF = 0
ES = 108
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF = 56254
CRC = 175
LOS = 0
LOF = 0
ES = 155

Edited by tommy45 (Wed 19-Jan-11 19:51:44)

you mean routers all 6 of them. 3 billion routers, 2 netgear and 1 speedtouch

the errors arent of a critical amount (performance is fine), but in the past I remember getting almost 0 errors on the upstream maybe 10 or so a day.

today I have had 2450 errors on downstream and 2732 errors on upstream.

Edited by Chrysalis (Wed 19-Jan-11 22:07:50)

Well, something's causing it. I thought you'd moved house, ISP, etc but kept the same router...

no just changed isp.

so common factor that hasnt been changed is line between me and exchange. now on new dslam, different router. In the past filters and cables have been ruled out.

now its not critical tho and I can lve with it, before it was much worse when I was pushing the upstream harder.