Print

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

[Black Sheep]

I'm hoping you can help me out here, if you don't mind ? 

Often, I (and my colleagues) will attend DSL faults and won't actually find anything at all of note. The EU always complains of 'Losing connection', and investigations generally show that they actually lose their IP connection, not the actual synchronisation.

Obviously, there's nothing we OR folk can do about that other than point it out to the EU. My question is what would be the best way to describe this situation in laymans terms ?? I find myself just saying that it appears to be an 'IP Addressing issue', but I could really do with knowing a little more on this particular subject.

I know there's 'Static' and 'Dynamic' ..... and the difference between them, but not sure what causes the EU to lose the IP Addy from the ISP on a regular basis (IE- 180 times a day) ?? Or what the right terminology is, that the EU should be using when liasing with their ISP regarding this issue ??

I hope this makes sense, and thanks for any feedback you may have.

[burakkucat]

Assuming that you are attending an ADSL fault (i.e. not VDSL2, a FTTC service), then what the EU is seeing is loss of their PPPoA connection with their ISP.

As the vast majority of EU's do not have a static IP address assigned to them by their ISP, every pair of loss/regain events (of PPPoA) will result in the the EU seeing a different IP address being assigned for their broadband connection.

[waltergmw]

Hi B S,

I suspect some of this requires a "Are you sitting comfortably ?
..... Then I'll begin "

and would probably start with an erudite dissertation such as:-

http://www.kitz.co.uk/adsl/ipaddress.htm

With many such things, just as you find umpteen slightly different faults, it's probably worth swatting up on various aspects.

Kind regards,
Walter

[Black Sheep]

Assuming that you are attending an ADSL fault (i.e. not VDSL2, a FTTC service), then what the EU is seeing is loss of their PPPoA connection with their ISP.

As the vast majority of EU's do not have a static IP address assigned to them by their ISP, every pair of loss/regain events (of PPPoA) will result in the the EU seeing a different IP address being assigned for their broadband connection.

Cheers B*Cat ...... yes, I am talking about ADSL pal. I get the PPP session thingy, but what causes it to 'drop' so often, when everything else is testing OK, even on the REIN front ?? I had an ongoing problem at a business premises, whereby the only way to 'mend' the issue was for the EU to move to a 'Static IP', it's been fine for months since this !!??

I suppose I'm trying to understand what ISP's IP addressing system is, and why, or what causes them to go faulty ??

[Black Sheep]

Hi Walt .......... thanks for the linky, I do need to set aside time to read-up on this. I wondered if our respected community had a 'quick fix' reply, in the interim ?? Thanks guys. 

[burakkucat]

Cheers B*Cat ...... yes, I am talking about ADSL pal. I get the PPP session thingy, but what causes it to 'drop' so often, when everything else is testing OK, even on the REIN front ?? I had an ongoing problem at a business premises, whereby the only way to 'mend' the issue was for the EU to move to a 'Static IP', it's been fine for months since this !!??

I suppose I'm trying to understand what ISP's IP addressing system is, and why, or what causes them to go faulty ??

I wonder if you are witnessing (or are receiving reports of) a symptom but not the actual fault. Just suppose that the PPPoA session still keeps dropping out and immediately recovering. With a dynamically allocated IP address, the EU will quickly be aware that something is not right and complain that "my Internet access keeps dropping" (insert any typical EU complaint between the two quotation marks). If the ISP's system is configured to provide a static IP address, the EU may not even notice some form of service "stuttering".

Ideally one would like to be present to witness exactly what it is, of which the EU complains. If, for example, it is a BT Business Broadband Service customer, I would suggest watching the relevant LEDs on the 2Wire 2700HGV or 2Wire 2701HGV-C. The LEDs are not very helpfully labelled on those devices -- the one marked "Broadband" indicates the modem / DSLAM sync status, whilst the one marked "Internet" indicates the state of the PPPoA session.

So really the question is why does the PPPoA session keep dropping? The quick answer is a fault with the ISP's equipment. Maybe bRAS, maybe elsewhere. As for where? 

Probable Synopsis

Is it an EU equipment fault? No.
Is it an Openreach cabling fault? No.
Is it an ISP equipment fault? Yes.

[bbnovice]

Hi BS,

Maybe a red herring but I did (rarely) have a problem with my old ADSL service provided by Virgin which sounds sort of similar similar. In my case what seemed to happen was that the connection dropped for just a fraction of a second but restarted immediately. The router always showed a green light for an active DSL connection, but the red light then showed for no internet connection. Restarting the router then failed to clear this condition.
 
When I reported a suspected line fault to Technical Support and they could see no fault. However by changing my logon password at their end (and making the appropriate change on the router) I could then get an internet connection. When the password was changed by them (and me) back to what had been originally the fault disappeared.

The technical guy was not 100% sure what was causing this, but suspected (as he had seen something similar before) that their authentification server was getting its knickers in a twist and declining to accept my login. The server was certainly rejecting a valid password.  I think this happened about 3 times in all.

As I'm not an expert and don't know at what stage in the login process the IP lease is issued, but perhaps what you are seeing may be connected to a more general authentification issue?

Stands back and waits to be corrected..........
         

   

[burakkucat]

BBN -- That reads like a perfectly reasonable explanation of a similar but different issue! 

Similar in respect of the authentication dropping, different because there is no problem re-authenticating.

[smurfuk]

Surely isn't the point that as an OR engineer you can only deal with faults as far as the telephone exchange. Their problem is at or beyond the telephone exchange and is the responsibility of their ISP or whoever provides the backhaul link (which may be another part of BT but who has to be contacted through the ISP because of regulatory rules).

One thing I've learned from being on VDSL is just how unreliable the BTw network is. The local loop may be decrepid here but it works 99.9...%+ of the time, BTw have several hours downtime per month (on several occasions) and are responsible for all my loss of internet connectivity for more than a few seconds.

I've sometimes felt that BT should provide a single "network fault report" service for all residential broadband, liaising with the ISP rather than the other way around, rather as operates for the electricity and gas utilities.

[burakkucat]

As to you first paragraph, yes, I agree.

In the case of Black Sheep, I know that he attempts to be as helpful as he can -- which must be applauded. 

The average EU often fails to appreciate that Openreach is a company in its own right. More often that not Openreach, BT Retail and BT Wholesale are regarded as one . . . and all tarred with the same brush. There are Openreach staff, B*Sheep being one, who take an interest in the overall big picture and like to be able to answer EU queries -- thus redressing some of the negative opinions of "Beatie" that abound.

[Black Sheep]

Hi BS,

Maybe a red herring but I did (rarely) have a problem with my old ADSL service provided by Virgin which sounds sort of similar similar. In my case what seemed to happen was that the connection dropped for just a fraction of a second but restarted immediately. The router always showed a green light for an active DSL connection, but the red light then showed for no internet connection. Restarting the router then failed to clear this condition.
 
When I reported a suspected line fault to Technical Support and they could see no fault. However by changing my logon password at their end (and making the appropriate change on the router) I could then get an internet connection. When the password was changed by them (and me) back to what had been originally the fault disappeared.

The technical guy was not 100% sure what was causing this, but suspected (as he had seen something similar before) that their authentification server was getting its knickers in a twist and declining to accept my login. The server was certainly rejecting a valid password.  I think this happened about 3 times in all.

As I'm not an expert and don't know at what stage in the login process the IP lease is issued, but perhaps what you are seeing may be connected to a more general authentification issue?

Stands back and waits to be corrected..........
         

   

BBN ........ that's the stuff I'm pertaining to. I think that you could have hit the nail on the head there with the 'Authentication issue' being the problem ? Our testers bypass this facility and log directly onto the RAS. So no passwords etc are required. I think (hope) thats the answer I'm seeking ?? Cheers for the help here one and all.

[Black Sheep]

As to you first paragraph, yes, I agree.

In the case of Black Sheep, I know that he attempts to be as helpful as he can -- which must be applauded. 

The average EU often fails to appreciate that Openreach is a company in its own right. More often that not Openreach, BT Retail and BT Wholesale are regarded as one . . . and all tarred with the same brush. There are Openreach staff, B*Sheep being one, who take an interest in the overall big picture and like to be able to answer EU queries -- thus redressing some of the negative opinions of "Beatie" that abound.

Thank you for the accolade B*Cat. Tis true that I try to make the EU 'BB aware' as they are generally that frustrated when we eventually arrive, they don't tend to believe anything we say (See another post from Adrian H about this).

Alongside this though, we get 'Repeat Reports' (Discipline route) if a job is re-reported within 28 days. I can do my very best and explain the LED's on the router (which I do) regarding synch and IP Address (PPP Session), and demonstrate it's the ISP's issue, but guess what ?? Chances are I'll be back there again within a day or two with the exact same issue !!

I believe BBN has summed it up regarding the A\Layer being possibly corrupt. Its this info that I can pass on to my EU's in the hope they can relay the same message to their ISP. We always put great notes on our jobs when closed, does anybody read them though Have a pot shot !!

[Bald_Eagle1]

Hi BS,

We always put great notes on our jobs when closed, does anybody read them though

Sadly, from my own experiences, that is not always the case.

I suppose that like most jobs, some engineers may treat it just as a job, whereas others (yourself included) treat it also as a service, where it doesn't take long & can be very helpful to service users to feed them with relevant information & provide clear & relevant confirmation notes for reference.

You obviously gain tremendous job satisfaction from being helpful & informative, exhibiting a willingness to learn (above & beyond the call of duty?)

I would imagine that if you were given a case with a similar history to my connection's, you would conduct the home visit with a slightly different approach than some of the engineers that have visited my home to date.

I once ventured into an engineers forum & saw at first hand some of the attitudes some engineers display when dealing with end users.

Paul.

[coolsnakeman]

The way i explain it to my customers over the phone (bare in mind these are BT business customers) is that they are getting a signal ok to the router but they are not getting an IP address to connect to the broadband from the exchange. If you are seeing a large result of 180 login attempts then that usually can be subject to a faulty piece of kit causing this to happen. Another cause of this is an incorrectly configured router (giving they have a static IP address). Customers VPI or VCI could be configured to the incorrect settings but this would also cause problems if they where dynamic aswell. Another thing i find when on a static IP is that if they are only using a single static but have this configured manually into the router and not as automatic from the SP this is not going to work for BT Business broadband as the exchange is set to deliver the static on recognition of the network username and password (not to sure if this is the same with other SP) so this can cause multiple login attempts. Most people usually say that internal wiring can not be the cause of loss of PPP issues but in my experience i have proved that to be rubbish by fixing a good number of drop in PPP connections by going into the test socket but with you being an engineer i would presume all that should have been done by the SP before you where tasked out to the job. Well that is my insight hope this helps 

Gary

[coolsnakeman]

Just on the read the engineers notes thing there i have to say i do read the engineers notes all the time and some of them vary with very little information to more than enough information. Another thing aswell which was mentioned in one of the posts was about the change of network password. That in itself is all standard procedure by the SP so that should of been done aswell as the test id or id's. For BT business the test id's are bt_test_user@hgxx.btclick.com and bt_user@startup_domain. These 2 test id's will test the HGW (Home Gateway) at BT's exchange. If the customer gets the BTW landing page with any 1 of these test id's then that means the problem lies outside of BTW's network so it means it is beyond the HGW. If the customer does not get access to the landing page then that means the fault is inside BTW's network. We have a desk called TOU (technical ownership unit) that deal with all static IP issues aswell as PPP issues outside BTW's network so all SP should have an escalation department to put the customer to for further investigation. I can even remember the last time i booked a boost visit or a FTTC engineer to a drop in PPP cause you simple do not do it unless it becomes the very last resort and you have exhausted all your troubleshooting.

Gary