Kitz Forum
Broadband Related => FTTC and FTTP Issues => Topic started by: z1ts on May 03, 2017, 09:53:30 AM
-
Good morning fellow Kitizens!
Looking for some insight into my business FTTC service - I have a TP-LINK TD-W9980 combined VDSL modem/router on a very long line.
Attached is a screenshot from my router. Question is:- do I have a problem??!! Websites take ages to load, videos buffer like mad, web browsing is generally painful and I've got an office of 10 people moaning at me :(
Some insights into what could be up would be very much appreciated :)
Thanks in advance
(Very willing to supply more info if you tell me what would help!)
z1ts
-
Today's Speedtest:-
www.speedtest.net/my-result/6267144234 (http://www.speedtest.net/my-result/6267144234)
(Ping looks ridiculously high to me....)
-
was that done while the connection was idle? was it done over wifi?
-
Hi z1ts
I could be wrong, but as it's daisy, my gut feeling is congestion. However, this is a gut feeling and nothing to prove it
You could try pathping BBC.co.uk which would help narrow my gut feeling
You could also try gateway hopping, to see if that improves it for you
Many thanks
John
-
Thanks both,
@Dray - Speedtest was taken on wired connection, with virtually no other services accessing the web at the time.
@d2d4j - could you step-by-step me through pathping bbc.co.uk and gateway hopping? :-[ (Sorry!)
-
Can you try it again when you know the connection is idle? I just tried Warwicknet and got a ping of 18ms. Is the Coventry server the best for you?
-
Just tried it again:-
www.speedtest.net/my-result/6267218015 (http://www.speedtest.net/my-result/6267218015)
Notice it's tried a different (Gloucester) server this time! I just let it 'select best server based on ping'.
What determines whether one server is better for me than another?
-
That ping is ridiculous. I find the best servers for me to be Maidenhead Xilo or Milton Keynes Uno.
If that ping is with an idle connection then that is the problem.
-
Hi z1ts
Dray is very knowledgable
Pathping - open a command box in windows with administrative rights, then type pathping bbc.co.uk
You could post the result which would help
Gateway hopping - if not experienced, turn router of for 30 mins usually does this, as it should allow the session to time out fully and not induce dlm
You could try speedguide.net and choose Helweb as test server
As I say though, my gut feeling is congestion on one of daisy pipes but apologies in advance as I could be wrong fully
Many thanks
John
-
Thanks
I've done a pathping and below is the results:-
Throw any light?!
C:\WINDOWS\system32>pathping bbc.co.uk
Tracing route to bbc.co.uk [212.58.246.79]
over a maximum of 30 hops:
0 MY_PC_NAME.company_name.local [192.168.1.35]
1 192.168.1.1
2 lo0.ar87.ts1.bb.daisyplc.net [62.72.136.165]
3 62.72.133.85
4 62.72.133.85
5 be24.cr18.tn5.bb.daisygroup.net [62.72.134.82]
6 be24.cr18.tn5.bb.daisygroup.net [62.72.134.82]
7 daisy-te4-3-503.thdo.bbc.co.uk [62.72.139.66]
8 * * *
Computing statistics for 175 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 MY_PC_NAME.company_name.local [192.168.1.35]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.1.1
0/ 100 = 0% |
2 312ms 0/ 100 = 0% 0/ 100 = 0% lo0.ar87.ts1.bb.daisyplc.net [62.72.136.165]
0/ 100 = 0% |
3 307ms 0/ 100 = 0% 0/ 100 = 0% 62.72.133.85
0/ 100 = 0% |
4 311ms 0/ 100 = 0% 0/ 100 = 0% 62.72.133.85
0/ 100 = 0% |
5 308ms 0/ 100 = 0% 0/ 100 = 0% be24.cr18.tn5.bb.daisygroup.net [62.72.134.82]
0/ 100 = 0% |
6 307ms 0/ 100 = 0% 0/ 100 = 0% be24.cr18.tn5.bb.daisygroup.net [62.72.134.82]
0/ 100 = 0% |
7 309ms 0/ 100 = 0% 0/ 100 = 0% daisy-te4-3-503.thdo.bbc.co.uk [62.72.139.66]
Trace complete.
-
Hi z1ts
Hop 2 shows 312ms, so I would turn router off for 30 minutes and see if that improves your issue
I ran a test to hop 2, which over 50 pings showed 28ms, so it does not look like congestion from here, and no pipes were red. However, that does not mean it is not congested, just means my test to hop 2 showed clear
The pathping shows your issue first starting between your router and gateway, which is why I suggest turn router off
One more thing, has anyone changed your router settings recently
Many thanks
John
-
Thanks @d2d4j
I'll press on with a router power down for 1/2 hour or so and report back.
No one has tinkered with the router settings.
Stand by!
-
The upstream test speeds look low relative to the upstream sync speed, and the sync speed looks low relative to the attainable speed. Low upstream speeds might be adding to the effect of the ping times, making downstream appear more congested.
Something to bear in mind once the issue over ping times gets clearer.
-
Well that sorted the ping out:-
www.speedtest.net/my-result/6267676750 (http://www.speedtest.net/my-result/6267676750)
Thanks!!!
Thanks for that - now upstream is still too low - normally in the region of 1.5 to 2.0 is what I'd expect.
Any suggestions?
-
Hi z1fts
Glad your ping is resolved
Please can you post your new router speeds as you originally did in your post, showing the new speed the router is now working on
Also, please could do another pathping
If your new router connected speeds are similar to your original posted speed, and your hardwired for the testing, with no other internet connections been active, then it's a little more involved as it will be something on the router which is lowering your upload speed
Wwwombat is very knowledgeable
@wwwombat - sorry I neglected to mention your point, but given the ping times, even on .5mbps, the ping times should have been lower
Many thanks
John
-
@d2d4j - as requested:-
Tracing route to bbc.co.uk [212.58.244.23]
over a maximum of 30 hops:
0 MY_PC_NAME.company_name.local [192.168.1.35]
1 192.168.1.1
2 lo0.ar88.ts1.bb.daisyplc.net [62.72.136.179]
3 62.72.133.89
4 62.72.133.89
5 be25.cr18.tn5.bb.daisygroup.net [62.72.134.86]
6 be25.cr18.tn5.bb.daisygroup.net [62.72.134.86]
7 daisy-te4-3-503.thdo.bbc.co.uk [62.72.139.66]
8 * * *
Computing statistics for 175 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 MY_PC_NAME.company_name.local [192.168.1.35]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.1.1
0/ 100 = 0% |
2 292ms 0/ 100 = 0% 0/ 100 = 0% lo0.ar88.ts1.bb.daisyplc.net [62.72.136.179]
0/ 100 = 0% |
3 302ms 2/ 100 = 2% 2/ 100 = 2% 62.72.133.89
0/ 100 = 0% |
4 299ms 2/ 100 = 2% 2/ 100 = 2% 62.72.133.89
0/ 100 = 0% |
5 302ms 5/ 100 = 5% 5/ 100 = 5% be25.cr18.tn5.bb.daisygroup.net [62.72.134.86]
0/ 100 = 0% |
6 305ms 1/ 100 = 1% 1/ 100 = 1% be25.cr18.tn5.bb.daisygroup.net [62.72.134.86]
0/ 100 = 0% |
7 298ms 0/ 100 = 0% 0/ 100 = 0% daisy-te4-3-503.thdo.bbc.co.uk [62.72.139.66]
Trace complete.
-
Hi z1ts
Your pings look high
Could you post another speedtest to show if your pings have increased
Many thanks
John
-
Yes :(
www.speedtest.net/my-result/6267984759 (http://www.speedtest.net/my-result/6267984759)
-
Am I correct in surmising that the problem is looking as though there's someone(thing) on our network uploading a large amount of data (which I am unaware of)?
We did have a similar issue a bit back with an online backup that begins at 18:00 every night but had over-run and was still processing when we arrived at work the following morning. I have checked and it's NOT this, but I am just wondering if there's something that is on our network that's chewing through a lot of data....
Is there any way telling traffic through certain IP's on a TP-LINK?
-
Hi z1ts
I think your overthinking
The best way to test if your right, is to unplug the router from your network, turn off wifi at router level and then with 1 pc connected hard wired to router, repeat the speed test 3 or 4 times, then restart the router and repeat speedtest 3 or 4 times. Then use the average of the tests
If same ping and slow speed, do you have a different router to test with, to exclude your current router
Personally, my gut feeling is it's a daisy issue, but you need to exclude your router and network ideally before you contact your broadband provider.
Also, are you using a plug in filter or a nte faceplate with inbuilt filter
Many thanks
John
-
I think that something is uploading sounds quite likely.
The TP-Link does have System Tools -> Statistics.
Generally I wouldn't recommend anyone use a TP-Link TD-W9980 because it does not support vectoring, and is now "End of Life" in terms of TP-Link support for it.
-
Is it worth using a ThinkBroadband "ping monitor" 24x7, to see how consistent the high ping time is?
http://www.thinkbroadband.com/ping
-
Well - I've yet to try the "ping monitor" but I think our SBS server was doing something I've rebooted it and now:-
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fwww.speedtest.net%2Fresult%2F6269903530.png&hash=4b508f0ab6e01e000e5ff04977b7df29a8a195af) (http://www.speedtest.net/my-result/6269903530)
All seems a bit happier today! :)
-
Hi z1ts
Glad you resolved your issue
My experience with daisy has not been good in the past, usually because of congestion on their pipes, so I guess I'm a little biased
The quickest way to determine if it's provider or your systems is to fully disconnect your network/systems from the router and test using a pc by hardwire and if good result, it's your systems/network, if bad, its provider - usually and only use as guide to where the issues lies
Many thanks
John
-
In that case, it looks like an over-used upstream was causing all the latency problems, and so likely to impact on all other downstream activities too.
The thing to decide now is whether to chase that "max attainable" upstream speed of 2.3Mbps instead of the 1.2Mbps you currently have.
-
All - thanks so much :)
If I have to move heaven & earth to chase the 2.3Mbps then I'll not bother, but if there's a few 'tweaks' I could make to improve it then happy days!
I'm actually looking to improve the situation wildly here with a few creative solutions! Bonding 2 or 3 FTTC lines is an option - using someone like Evolving Networks (https://evolving.net.uk/) (don't know if anyone has any experience??)
Unfortunately our budget doesn't stretch to a Fibre Leased Line (FTTP) at £350 + VAT per month with 'excess' charges of >£15k..... :no: :o :o (Would be nice though :) )
-
I have no experience of bonding FTTC, but some do. @Ignitionnet on here has done so in the past.
Also on here, @weaver is using AAISP to bond 3 ADSL2+ lines together, using one of their own routers. Firebrick?
Chasing the 2.3Mbps will entail getting deeper statistics from the modem, to find why it isn't using 2.3Mbps.
It looks like you need to move a little bit of heaven in the pursuit of this...
http://forum.kitz.co.uk/index.php/topic,14377.msg268677.html#msg268677
The root cause could well be the error rate, which DLM may have adapted. That would mean trying to monitor with one of the graphing programs such as DSLstats, but I'm not sure one works with that modem.
-
Also on here, @weaver is using AAISP to bond 3 ADSL2+ lines together, using one of their own routers. Firebrick?
On Weaver's behalf, I'll say "Yes, a FireBrick FB2700 (http://www.firebrick.co.uk/products_2700.php), fully loaded".
-
Is it worth using a ThinkBroadband "ping monitor" 24x7, to see how consistent the high ping time is?
http://www.thinkbroadband.com/ping
Have you checked that web page ?. :(
-
Uhoh. Looks like someone got the decorators in...
-
Tommy began a separate thread (http://forum.kitz.co.uk/index.php/topic,19748) about it . . .
-
Is it worth using a ThinkBroadband "ping monitor" 24x7, to see how consistent the high ping time is?
http://www.thinkbroadband.com/ping
It looks like TBB are back up with a new website, and the BQM's have had a small update - they can now work to a hostname, not just an IP address. If your IP address changes, then BQM will work with a "Dynamic DNS" name that has been updated with the new IP address.