Kitz Forum
Broadband Related => Broadband Technology => Topic started by: Bowdon on February 06, 2016, 11:23:52 AM
-
I'm sure this question must have been asked before. But are there any ISP's that offer FTTC Line Bonding?
I'm not even sure its possible. I'm thinking it might be as I know ADSL used to be able to do it. But it was dependent on the ISP.
Are there any ISP's that offer this service?
-
I believe AAISP can do this http://support.aa.net.uk/Bonding_for_resilience
-
AA do this. I use it. It's independent of the lower layer protocols, it just uses IP. (So not like ML-PPP for example where support for it has to be built into the lower layer systems.)
I have three ADSL2 lines bonded together. If one line fails, the load spreading picks it up within a few seconds and the other lines alone are scheduled, so everything carries on as normal.
-
It's just the same with FTTC, from what I've read. So you could bind four FTTC pipes together to get 320 Mbps if you wanted to and were lucky enough to get a whole 80 Mbps per pipe.
-
It's just the same with FTTC, from what I've read. So you could bind four FTTC pipes together to get 320 Mbps if you wanted to and were lucky enough to get a whole 80 Mbps per pipe.
*Rant on*
That would be a good idea. Hmm imagine doing that with 4 G.fast enabled lines.. I'd be zooming around the net ;D
I had a look at the AA website. I don't understand why it is so expensive. Also why do they still have usage caps on their products?
The FTTC seems to be mixed in the Home: 1 package, though it mainly talks about ADSL with a cap of 100gb's a month for £25. You can buy 50gb slots for another £10 (or increase speeds to FTTC). In this day and age people can easily go over 100gb's, especially with fibre.
The Office: 1 apparently as fibre lines. But look at this price list;
Service One-off cost Ongoing cost
Two line Office::1 (200GB) £500+VAT £200+VAT pcm
Three line Office::1 (200GB) £600+VAT £225+VAT pcm
Two line Office::1 (300GB) £500+VAT £275+VAT pcm
Three line Office::1 (300GB) £600+VAT £300+VAT pcm
Office::1 10TB with ADSL backup £500+VAT £300+VAT pcm
Top-up 50GB when over allowance £50+VAT
Why is there an ongoing cost? It just seems to me these companies are pricing themselves out of the market. If they lowered their rates and got more people onboard then they would sell more of their products.
I like it that AA is trying to break away from the current trends in the industry. But its a real shame they have such high prices.
I know the guy is well liked on here. If I'm missing something about line maintenance that puts the price up to big money, over and above what BT would do if a fibre line went down then can I hear the justification for it?
As I mentioned in another thread, I like the outlaws promoting new technology and pushing innovation through. That's why I supported Be internet, and thats also why I like Hyperoptic, Gigaclear and B4RN for breaking new boundaries. AA could also be in this group if they wanted to be.
I'd say 2 unlimited download fibre lines would be worth between £50 and £60 per month. If they offered that package I'm sure quite a few of the heavy usage home users would pay it.
*Rant off*
-
I suspect that:
1. AA don't want the type of consumer that downloads the entire internet every month.
2. Part of the high price will be to do with the exemplary service I understand you get particularly with faults and generally acting on your behalf fighting battles with BTOR by staff at AA who really know their stuff, so do not come on a minimum wage.. All this has to be paid for.
AA are in the same market sub-sector type as say bespoke hifi-shops - if you want a superb service and some high cost equipment all setup professionally in the house, then you go to them. If you are not bothered you buy something off the web from Amazon and the rest.
Likewise say Aston Martin or Morgan in the car business: low volume, very high prices for people willing to pay for a perceived difference.
These niche sub-sectors survive while there are enough people willing to pay more for such a service. They may continue ad-infinitum or, if the numbers willing to pay dwindle then they may not. Alternatively the market and circumstances may change, develop, move on such that the bespoke service offered is no longer appropriate or needed.
I half think RevK rants about the snoopers charter bill is less about a concern over the deterioration in all our 'rights', 'privacy' etc than a worry that as his costs start to escalate and as BB generally gets more reliable then a future for niche BB suppliers like AA may not exist and AA may go to the wall.
-
I think that the tariff choices and general (but not complete!) lack of all-you-can eat tariffs is to do with preserving the zero-congestion network (as far as the transit and wholesale backhaul suppliers will allow). AA doesn't have to do traffic shaping or anything horrendous of that ilk.
I wanted line bonding and IPv6, when I was out shopping for an ISP. I wasn't interested in any kind of posh customer service, in fact didn't know about it. I saw AA was very well loved on ISPReview and had consistently good user reviews. I was also Impressed by the "we'll fix your line or your money back" no-nonsense approach. In the end, Zen was out because of the endless wait for IPv6, although I did trial Zen. But in the end Zen failed on both IPv6 and line bonding fronts and were out of the race.
The Firebrick and the fact that AA are co-designers of it are real powerful reasons to use them, because the FB is superb and AA can really support it, that I've found very useful.
But the reality for this user is that posh customer service was not a factor, but now I'm a customer, I can't face the thought of ever having to talk to idiots on a telephone and not being able to view, control and change parameters myself.
There seems to be a lot of misunderstanding about AA. You have to ask real customers to find out what really matters. Most of the customers I've talked to are very experienced to ultra-knowledgeable, are Linux geeks, and have zero tolerance for idiot customer service scripts
-
In my 'rant' post I was typing it more out of frustration. Because AA is so close to getting more customers from the non-specialist customer base they usually get. I appreciate that they give the personal touch when dealing with customers and connection problems. I know when I was at Be it was good talking to people who really cared about the issues I was having.
It just saddens me when I was reading about the new Home: 1 Terebyte package. Though kind of expensive it would be within reason. Then I read that if a person goes over the package limit then they will cap the speed to 3Mbps! There is no option to buy a top-up.
I'm wondering if there are any other ISP's that would do FTTC bonding. Does the ISP have to do anything on their side? or is it all done at the EU side with the modem / router ?
Is it a complicated procedure to setup line bonding? I think there would be a big take up for line bonding, even ADSL style too.
I'm not an excessive downloader, though I'd say I might at one stage been above average. I'm abit too old these days to be downloading anything I can grab ;D But I still would be interested in fibre line bonding. I'd be prepared to pay at least double. As I joked the other day, imagine the speed of 2 G.fast connections. I'd be high on speed! :lol:
-
Proper line bonding will require configuration at the CP/ISP's end as well as at the user's end. As Weaver will attest . . .
-
Some of AA's packages give you a choice of what you want to happen when the paid-for amount of GB is exceeded. This is to protect the person paying the bill, and is an extremely good thing.
-
Line bonding requires a tin box of some kind at both ends. I have three lines going through a Firebrick 2500 router at my end. At AA's end there will be - I suspect - a Firebrick FB 6000 series which splits a stream of IP packets three ways heading for my different lines. AA get live information from BT or TalkTalk Wholesale as to how fast each of the individual lines' downstream pipes is running at any particular instant and this information is used to split the IP stream in the correct ratio when it's scheduled by the FB6000 to go into the three pipes. If one of the pipes goes down, then AA load splitting takes action within about ten seconds and knocks that pipe out of the scheduling algorithm, so having load spreading helps reliability too, the service just goes slower if I lose one or more of the pipes.
In the upstream direction, my Firebrick 2500 splits traffic across the three lines in a similar way. Unfortunately I don't have a near real-time feed of info telling me how fast each upstream pipe is going, so I have to put the relevant numbers into the Firebrick config in order to get the right load splitting ratios. Which is a nuisance.
The system works very well indeed and takes action really quickly when a line goes down so the rescheduling is seamless. The lines are probed every few seconds using PPP packets so the system knows if a line is really working as opposed to just being supposedly working.
-
Unfortunately I don't have a near real-time feed of info telling me how fast each upstream pipe is going, so I have to put the relevant numbers into the Firebrick config in order to get the right load splitting ratios. Which is a nuisance.
That's interesting, it sounds like you're carrying out a manual process. Is there no way to script this?
-
> is there no way to script this?
No, the Firebrick 2500 has no way of altering the ratios dynamically, and would need the addition of an API to fix this.
Getting hold of the numbers is messy. I check the latest upstream bandwidth figures given in the AA "clueless" control panel server's system logs. An API to query those values is needed too, otherwise it would mean trying to interrogate the web UI (hopefully valid XHTML but you never know) which would be somewhere between a bit hacky and not so bad.
-
Currently I just check the system logs every once in a while to see if anything dramatic about the line's characteristics has made the lines' upstream bandwidth change. I keep a little note of the latest values in a little iPad spreadsheet and edit the Firebrick's XML config file if necessary (and reboot it, to make changes take effect).
-
I thought XML was an API :)
-
I'm wondering if there are any other ISP's that would do FTTC bonding. Does the ISP have to do anything on their side? or is it all done at the EU side with the modem / router ?
Is it a complicated procedure to setup line bonding? I think there would be a big take up for line bonding, even ADSL style too.
Line bonding is available from several ISPs but it's targeted towards the business market.
You might be more interested in a load balancing (http://www.broadbandbuyer.co.uk/store/broadband-routers/load-balancing-routers/) router (software solutions also available)
Demand for bonding tends to be low for residential customers as it means paying for 2 or more broadband connections (inc. line rental) and I don't think it will take off unless Openreach decided bonding 2 lines from the DP was the favoured solution for providing higher speeds which isn't on the agenda any time soon.
-
> thought XML was an API
No. It's like saying that a CSV file or a .DOC file is an API.
XML could be part of one.
-
> XML could be part of one.
Yeah. I know :)
-
@Dray apol ;D ;D
-
Nothing to apologise for ;)
I just through it sounded an interesting thing which could be useful for you and others who are using bonding too :)
In fact I wouldn't be surprised if someone hadn't done it already.
-
The killer is that you would need to inject the change of config into the FB, and making any change take effect currently involves a reboot. You can't just go around rebooting your firewall-router randomly, so some internal proper design mods would be needed to have a protocol for updating the firebrick upload traffic scheduling ratios dynamically. And this isn't something a third-party would be able to do, it's a feature request for RevK.
-
We all have this problem with the Internet, we often don't know how much your correspondent knows, how much expertise she has. :-[
-
> I'd be prepared to pay at least double.
You don't have to pay that much. With the traditional 'units-based' tariff, you pay a small amount per line and pay once only for the total amount of downstream traffic you need. This is the scheme I use, and it suits me very well. (It's the oldest tariff scheme.)
-
This got me thinking, does PPPoE have flow control? It seems to me that it must otherwise even in a single router/modem combination the router wouldn't know how to pace the outgoing packets.
-
> As I joked the other day, imagine the speed of 2 G.fast connections. I'd be high on speed! :lol:
You'd need one of the new (rumoured) forthcoming Firebrick's instead to be fast enough to route it all. ;D
My FB2500 is the slowest of the FB2000 range, but it would be quite happy routing a heap of low speed lines. (More than four lines means the performance starts to degrade slightly depending on how good the TCP stacks in the devices are). It can even handle more lines than you have room for on free physical ports, by using a VLAN switch.
-
This got me thinking, does PPPoE have flow control? It seems to me that it must otherwise even in a single router/modem combination the router wouldn't know how to pace the outgoing packets.
It certainly isn't part of the standard PPPoE spec. I would expect flow control to appear in the standard place - at the TCP endpoints - and for intermediate routers and concentrators to cope with congestion in the standard way - by employing packet drops judiciously, using whatever queueing mechanism works best.
A quick google tells me that someone has written PPPoE extensions for flow control, but with a specific intention of use in some funny wireless setups. The RFC adds this note at the front:
The PPP Extensions Working Group (PPPEXT) has reservations about the
desirability of the feature described in this document. In
particular, it solves a general problem at an inappropriate layer and
it may have unpredictable interactions with higher and lower level
protocols.
-
It certainly isn't part of the standard PPPoE spec. I would expect flow control to appear in the standard place - at the TCP endpoints - and for intermediate routers and concentrators to cope with congestion in the standard way - by employing packet drops judiciously, using whatever queueing mechanism works best.
How would that work in a typical DSL scenario with PPPoE router connected by Ethernet (at least 10meg) to a DSL modem with let's say an 800K upload speed? If the router just sends the packets as fast as they arrive and the modem drops them if the uplink is still busy, then that means no possibility of upstream QoS on the router. Or to be more precise no possibility of prioritisation by the router.
-
Hi
It has always been my understanding that the Internet does not have QoS, so from your modem outbound, no QoS applies
Many thanks
John
-
I thought it was well known that if you turn off QOS on a Huawei HG612, your upstream sync increases by 1 Mbps.
-
Hi
Yes, it might increase in the short term, depending upon the traffic congestion, but if your congestion becomes high, you may slow down as there is no priory for traffic going out.
I just checked our hg612, and QoS is on, set to fair queue priority
Also, unless you have control over the DSL at both ends, your QoS is not fully working, only on the outgoing/incoming from your modem only
I hope that makes sense as the more I read my reply, the more I think I have not explained well sorry
Many thanks
John
-
How would that work in a typical DSL scenario with PPPoE router connected by Ethernet (at least 10meg) to a DSL modem with let's say an 800K upload speed? If the router just sends the packets as fast as they arrive and the modem drops them if the uplink is still busy, then that means no possibility of upstream QoS on the router. Or to be more precise no possibility of prioritisation by the router.
Wouldn't you have that anyway? Where your PC, attached with a "gigabit ethernet" connection to the switch, sends too much for the switch to forward to the router, because it is only connected with a 100Mbps "fast ethernet" connection? Now your worry has moved into your switch, not your modem.
The truth is that the TCP protocol starts slow, and ramps up ... until packets start to be dropped somewhere (anywhere) along the route. It then falls back to a speed where everything was happy.
If the bottleneck is at the link between router and modem, as your example, then the router would never have been receiving stuff to send on at 10Mbps; once the modem even started to dump packets (at 800k if it is dumb; earlier if smart), TCP will have backed off at source, and the router will only end up receiving packets at (or below) 800k. A sustainable level for passing onward into the modem.
If a second PC starts trying to send, the two independent TCP stacks will fall back to around 400k.
TCP doesn't work by dumping a huge file as fast as it can into a router, and then waiting for the queue to empty. It attempts to maintain a steady-ish speed that can be sustained end-to-end.
If you want QoS to work, then all your network nodes that have any congestion control within them need to support the same QoS system; this gets used by every place that chooses what packets to dump, and what to favour. It builds on top of standard TCP congestion mechanisms.
It has always been my understanding that the Internet does not have QoS, so from your modem outbound, no QoS applies
"The internet" doesn't. But QoS exists on many, more private, networks. BT's network - which your packets traverse before arriving at your ISP and "the internet" - does employ QoS. As we go further forward, and your voice calls become a service on top of IP, you can be sure that BT will have QoS mechanisms that keep voice prioritised.
I thought it was well known that if you turn off QOS on a Huawei HG612, your upstream sync increases by 1 Mbps.
Yup. Even if you don't send any QoS information with your upstream data (and almost noone does), the HG612 would spend time looking for those QoS markers, to decide how to queue your packets, and to decide what order to drop things in. This check was enough to slow down upstream throughput. Wasted work... but work, nonetheless.
-
I thought it was well known that if you turn off QOS on a Huawei HG612, your upstream sync increases by 1 Mbps.
Do you have any links to experimental results, please? (I am well aware of the BatBoy claims in the TBB forum but would appreciate sight of something more substantial.)
-
No, I haven't tried. I just meant it was a "well known fact" :)
But as far as I'm aware, you can only control QOS on the upstream. It's up to the ISP to control it on the downstream, like Plusnet do for example.
-
How would that work in a typical DSL scenario with PPPoE router connected by Ethernet (at least 10meg) to a DSL modem with let's say an 800K upload speed? If the router just sends the packets as fast as they arrive and the modem drops them if the uplink is still busy, then that means no possibility of upstream QoS on the router. Or to be more precise no possibility of prioritisation by the router.
Wouldn't you have that anyway? Where your PC, attached with a "gigabit ethernet" connection to the switch, sends too much for the switch to forward to the router, because it is only connected with a 100Mbps "fast ethernet" connection? Now your worry has moved into your switch, not your modem.
When I say QoS I'm referring as I said specifically to queuing and prioritisation, congestion management techniques. It is a first principle that these need to be applied at the point of congestion, where the decision about which packets to send and which to drop is being made. For example I might configure the router to reserve bandwidth for voice or voice signalling traffic. Then when the link becomes fully utilised, those voice packets will be forwarded in preference to others. However in the PPPoE router/modem scenario the router will never see it's Ethernet becoming fully utilised, so that selective treatment won't come into play and the forward or drop decision will be made by the modem.
If you want QoS to work, then all your network nodes that have any congestion control within them need to support the same QoS system; this gets used by every place that chooses what packets to dump, and what to favour. It builds on top of standard TCP congestion mechanisms.
In an ideal world yes, however if the drop probability is significantly less further along the path then the sort of techniques I described above are effective.
-
By the way, my reason for asking in the first place was to do with traffic allocation between links that are load sharing on a packet by packet basis. If PPPoE had it's own flow control then each output queue on the FB would only empty at the speed that it's modem was forwarding. Hence load sharing wouldn't need the relative bandwidths hard coded.
-
Hi
If dray is willing to reverse his QoS setting, I will reverse my QoS, which should answer the theory of QoS turned off/on give any benefit
I have only just joined online logging with tony, so if dray is willing, I just need a time/date to complete this test and leave running until for say 2 days, and you can see the results
Many thanks
John
-
Sorry, from my somewhat faulty memory, I think it only has an effect if the HG612 is being used as a router. In bridge mode, the QOS settings are ignored.
-
Do you have any links to experimental results, please? (I am well aware of the BatBoy claims in the TBB forum but would appreciate sight of something more substantial.)
Turning QoS off did have a beneficial impact on upstream speeds for me.
I've found a graph from my SamKnows box when I was on a 40/10 package with Plusnet.
I had an equally good result on the 80/20 package, but I can't find a graph for that ... and my SamKnows account got reset when they sent me a new box, so I lost all the old results.
-
Hi
If it helps, and as wombat is using samknows as the guide, I might fully turn off QoS on the hg612, and let it run for a few days then post the result.
On samknows, our upload has been fairly stable at just under 17mb, so any difference should show
Would this be of any use to anyone
Myself, it does not matter either way, but please be aware we are a heavy user, well I think so, last month with uploaded over 50gb and downloaded over 350gb, and that is with some days missing from dslstats
Dray, sorry I never thought of that, but wombat post appears to show differently unless wombat uses it as a router and not bridge mode. If router, then I suspect it cannot handle both, whereas our draytek can
I did consider if you were meaning an external speedtest as apposed to syncd speed, but if so, we use helweb from speedguide.net, as this is our server and we know it's setup/network etc
Many thanks
John
-
As I said, faulty memory so I'd appreciate a refresh from someone else as I couldn't test it at the moment.
-
This might be worth reading from 2012 (Kitz link):
Re disabling QOS on HG612 (http://forum.kitz.co.uk/index.php/topic,11520.0.html)
-
Hi
Many thanks and not read that before thank you
I am talking about hg612 and HH as we use draytek, but interestingly, I did not post, but the QoS on the hg612 does not appear to have profiles setup, either for LAN or WAN, other then FQP, so I'm not sure if turned on, it actually does any QoS
If I do any tests, please let me know the tests, but we have a range of IP address, which some must remain working with the exception of quick breaks, so can do most tests.
If using samknows as baseline, I can post present upload graphs, with QoS as standard on, then turn it off and let it run for a few days, and you can see the result
If you want me to do this, just let me know and I'll organise it, along with any other test which is feasible to complete without bringing our IPs offline
Many thanks
John
-
This might be worth reading from 2012 (Kitz link):
Re disabling QOS on HG612 (http://forum.kitz.co.uk/index.php/topic,11520.0.html)
Hmm . . . nine posts to that thread and one third of them were mine!
I have absolutely no recollection of that thread. :paperbag:
-
If using samknows as baseline, I can post present upload graphs, with QoS as standard on, then turn it off and let it run for a few days, and you can see the result
If you want me to do this, just let me know and I'll organise it, along with any other test which is feasible to complete without bringing our IPs offline
Thank you. I think that would be a worthwhile experiment and it would be interesting to see if the result is similar to that which WWWombat obtained.
As for other tests . . . perhaps others will make suggestions, please. ;)
-
I thought it was well known that if you turn off QOS on a Huawei HG612, your upstream sync increases by 1 Mbps.
Do you have any links to experimental results, please? (I am well aware of the BatBoy claims in the TBB forum but would appreciate sight of something more substantial.)
I turned my QOS off last month. According to my Samknows box, the upload increased from 8.6/7MB to about 9.3MB
-
Hi
Please see the result when QoS turned off this morning.
I will leave running with QoS turned off until Thursday or Friday then turn QoS back on, where it should be easy to see on the graph from Samknows when off and on.
I hope that helps, but I do wonder why when QoS is on, but no rules shows for either Lan or Wan, why it makes a difference, other then as suggested, it is slowing due to checking
Many thanks
John
-
:hmm: Hmm . . . Very interesting. Thank you. :)
To be honest, I don't really understand what is occurring. Considering the HG612 (configured exactly as distributed by Openreach), it is just a pure bridge, converting from/to VDSL2 to/from Ethernet. Where in that setup could any form of QoS be configured to operate meaningfully? My answer is "nowhere". There is just a frame of data passing through the HG612 . . . it enters in one format and exits in another format.
Hopefully we can reach a sensible understanding as to what is taking place. :-\
-
Hi
If it helps to better understand, please see 2 pics showing the QoS before disabling QoS.
Many thanks
John
-
Hi
Please see last pic of QoS turned back on
Many thanks
John
-
So is that a rise with QOS off and a fall with QOS on or something else?
-
Hi dray
Sorry, the rise is with QoS disabled, the drop is when QoS was enabled.
It would appear that you were correct with your advice
Also, I checked all the other graphs, upstream jitter etc, even speedtests using speedguide.net, helweb as test server and all looked OK with no changes but a rise in speed upload
Many thanks
John
-
Thanks, I'll just fix my question :)
-
I'm using the HG612 as a VDSL router, just turned of QOS and typical upload increased from 10.5 to 12!
Thanks for flagging this.
-
This doesn't specially surprise me, it must be quite processor intensive looking inside those PPPoE packets and managing queueing. I really comes back to what Weaver was saying about knowing the upload rate, shape the output from the router to suit, and you can run the QoS on the router instead.
-
Hi
Yes, I believe it is due to processing of QoS, but if you look at my 2 pics showing the QoS, when enabled, there are no rules/policies setup fir WAN or LAN.
Many thanks
John
-
I've just experimented here with an HG612 in bridge mode, and got similar results to the others (using speedof.me):
Upload speed with QoS enabled: 18.05 Mbps
Upload speed with QoS disabled: 19.88 Mbps
-
So I just looked inside my HG612 advanced settings and sure enough QOS was enabled.
Now disabled ;)
-
Yes, I believe it is due to processing of QoS, but if you look at my 2 pics showing the QoS, when enabled, there are no rules/policies setup fir WAN or LAN.
So we now have a definite and (multiple) verified result that a HG612 with QoS disabled shows a greater US throughput than when the same device has QoS enabled.
The other interesting fact (as John recounts, above) is that even without any defined rules the overhead generated by turning on QoS is sufficient to degrade the US throughput.
-
Has anyone tried defining QoS rules? And then sending traffic that uses them?
-
What happens to your latency when you turn QoS off and max the upstream?
I'd suspect there's a little bit of "wiggle-room" being left on US so everything else doesn't get lagged out.
-
Hi rizla
I have not noticed any difference to upload and our upstream can be maxed a few times a day
I did notice this morning though, our upstream interleave went from 0 to 1 then back to zero (I understand 0 and 1 mean no interleave applied) but I am wondering if the small upload reserve is for dlam communication to hg612 perhaps
It's just a thought and is most likely wrong
Many thanks
John
-
What happens to your latency when you turn QoS off and max the upstream?
I wonder if you have seen the observation, a little earlier above, that there are no QoS rules defined?
As distributed, the HG612 has QoS turned on but with null rule(s).
-
Hi burakkucat
I hope you don't mind, but I think rizla is trying to find out if QoS off, and max upload reached, is there are loss or buffering.
I could be wrong so I apologise in advance
Also, I don't think the state of the hg612 is known prior to the wolf firmware been installed, which then gives access to login, to see
If it helps, we have hg612 as standard supplied, so could swap out for BT standard and rely on mydslstats/samknows to see if the upload decreases or increases or remains the same
I hope that makes sense and sorry if I'm wrong
Many thanks
John
-
Spot on d2d4j.
-
I tried this when I first read about it in the TBB forum - possibly as far back as 2011.
As my HG612 had been running with QoS enabled by default when I swapped it to a newer 3B version some time ago (I had forgotten to disable it), I disabled it this morning & also forced a modem resync & a router reset.
For essentially no change in US sync speed (from 4784 Kbps to 4789 Kbps), US throughput speed using the BT tester went from 3.72 Mbps to 4.32 Mbps.
-
OK what I was looking for was whether the latency went into thousands of milliseconds when QoS was off & you max the upload but not when it was on.
I'm pretty certain that in many (all?) of the busybox builds QoS had prioritising ACKs built in by simply reserving a percentage of available bandwidth up to a maximum value.
If that's the case then there ought to be enough bandwidth for ICMP to show a significant difference in latency between QoS on & QoS off while upstream bandwidth is "maxed".
-
I’ve finally managed to get my hands on a second user Firebrick 2700 from AAISP and have successfully bonded my two lines from them. It was a fairly straightforward process following the configuration snippets on the support site. Speeds are impressive as is the Firebrick itself. I’m on the Home::1 product with an extra broadband service sharing the one usage allowance. :)
-
Which options from AAISP did you select and what does it cost ?
thanks
-
Here's what I currently have from them all under the Home::1 product:
Copper pair for broadband service
Broadband 100GB usage allowance
Add-On: VDSL/FTTC
Add-On: 80/20Mb/s cap on VDSL/FTTC/FTTP instead of standard 40/10 cap
Extra broadband service on same login (no extra usage allowance on my BT line)
Add-On: VDSL/FTTC
Add-On: 80/20Mb/s cap on VDSL/FTTC/FTTP instead of standard 40/10 cap
This totals £75 pm :'(. I still have my original line with BT until my LRS runs out when I will move it too + £10. No PSTN calls can be made in their lines, but they can port original BT numbers to their VoIP service I believe.
The Firebrick came in a little bit under half price for the fully loaded version. :)
-
thank you !
-
A (belated) warm welcome to vic0239 into the Firebrick-bonded A & A club!
As you may already know, I run a bonded set of three ADSL2 lines, extremely long. Speed : ~ 2.2 Mbps downstream each, so combined is ~6.6 Mbps, and the combined total upstream is ~0.7 Mbps at the moment (for some reason, it has dropped, it used to be > 1.0 Mbps until recently).
I'm on the "units tariff", I buy a varying number of usage units, but my usage isn't very high. I don't pay "line rental" to anyone else. I buy three copper lines for DSL from A & A, no voice service on them, and I also pay for three lots of BTW priority ("premium" what's it called?) at ~£10 pm per line. I also buy an AA 3G data SIM service for one of my devices plus I have to pay for traffic on that connection. My total bill is typically something like ~£150 per month, I think.
I run a Firebrick FB2700 which I love. I get a very nice IPv4 /26 block, so no NAT at all, and I have an IPv6 /48 of course. I have a single static IPv4 address for my 3G mobile connection (and no IPv6, yet). At some point I'm going to get round to testing VoIP, to see if I've got it set up right.
-
Thank you weaver! :drink:
Yes, I love the Firebrick too, great functionality. I have just recently taken a VoIP number from A & A to enable me to get to grips with it prior to porting my existing land line to them. I have set up the Firebrick as a back-to-back SIP endpoint and register my N300 and SPA3102 to the Firebrick. It all feels (and sounds) very stable. I did consider the data SIM, but mobile reception is patchy and prone to going AWOL for a few days here so not much use as a backup service.
-
I have a Siemens N300 VoIP box as well. I had some trouble getting VoIP to work, possibly because I don't have enough free bandwidth, I don't know. So I got AA support to configure the device properly for me.
I have to get round to testing it thoroughly anyway, both with and without heavy load on the Internet connection in both directions. I should have done this ages ago, but I've been seriously under the weather, and then some, a poor excuse.
Shame it doesn't speak IPv6, AA would probably growl at them for this omission.
I haven't worked out how to get the Firebrick to act as a back-to-back gateway, I've configured the firewall rules to let the AA VoIP servers talk directly to the N300 and vice versa. Since I have no NAT at all, I don't see the need to involve the Firebrick in processing the traffic, one less thing to deal with. I could of course get AA staff to help configure the Firebrick to be a man in the middle, but if it works without, well, we’ll see.
-
I have just a single IPv4 address so decided to go the VoIP gateway route. Once you have your firewall rules in place there are just two further elements to code, one for the carrier (A&A in this instance, but I also have it working with Sipgate) and one or more telephone elements depending on whether you have multiple VoIP phones.
On the N300 point it to my.firebrick.co.uk using the credentials in the telephone element.
<voip area-code="0xxxx"
comment="AAISP">
<carrier name="AASIP+44xxxxxxxxxx"
allow="81.187.30.110-119 2001:8b0:0:30::5060:0/112"
registrar="voiceless.aa.net.uk"
username="+44xxxxxxxxxx"
password=“pwd from control pages"
extn="+44xxxxxxxxx"/>
<telephone name="N300A"
username="gigaset"
password=“a pwd of your choice"
ddi="+44xxxxxxxxxx"
extn="100"
carrier="AASIP+44xxxxxxxxxx"/>
</voip>
+44xxxxxxxxxx is your A&A VoIP number. You can see the call progress on the Firebrick VoIP status page. Such fun! ;D
-
Many thanks for putting that tip together, very good of you.
Apologies for hijacking this thread and drifting seriously off-topic btw.