Print

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

[Chrysalis]

regarding congestion since last week or 2 there has been an increase which is either BTw or plusnet side.

I had yet another ppp drop last night so am on a different gateway and will find out tonight if I am congested again.

[HighBeta]

Plusnet have major issues still with the load balancing  (amongst other things) of the new bgn's.

Its that much of an issue BT's network people are now  lead.   

[Chrysalis]

Plusnet have major issues still with the load balancing  (amongst other things) of the new bgn's.

Its that much of an issue BT's network people are now  lead.   

yeah and this came in earlier

Service: Broadband
Posted: Wed, Oct 01 2014 at 16:38:21
Subject: Emergency Broadband Network Maintenance - Wed 1st October 4.30pm - 6.00pm

[Chrysalis]

some info here backing up the DLM using ES and it does match exactly to the 2880/day theory.

http://support.zen.co.uk/kb/KnowledgebaseArticle.aspx?articleid=11511

[roseway]

That's a very helpful one.

[NewtronStar]

I don't get this part as it does not specify what type of error they are talking about is it CRC's ES or SES

The additional thresholds are:
 
Standard – allowing no more than 10 drops in a 24 hour period and no more than 1 error every minute.
 
Stable – allowing no more than 5 drops in a 24 hour period and no more than 1 error every 2 minutes.

[roseway]

You're right, it isn't clear. "One error" seems to imply one CRC, but I think it probably means one ES.

[Chrysalis]

on a SES the ES also tallies, it will be talking about ES.

[les-70]

  I find some aspects of this very odd, the standard ZEN connection is

 "Zen Internet provides our Fibre Optic Broadband (FTTC) services on the “Speed” option. A connection should drop no more than 20 times in a 24 hour period, and should error no more than 2 times per minute."

  This would mean that a resync would in effect only be noticed via the error count.  Unless you have 20+ a day of course! However a line error caused resync event is what most would regard as the one event that the DLM really needs to avoid happening.

   If they are intended to spot resyncs caused by errors, then I am not sure of the errors.   120 ses or es an hour can easily conceal a resync event whilst 2 per min is common.  120 CRC/hour (2/min) is possible but it does seem too strict.  On the other hand I think it really is a big CRC event that is more associated with a resync than a big es and ses.

   I have to say I feel as though were only being given a general idea and not the much needed detail that you need to make some value judgments.

[Dray]

I think resyncs are counted separately. The modem issues a "dying gasp" so DLM can tell the difference between a loss of power and a loss of sync.

[les-70]

  "A connection should drop no more than 20 times in a 24 hour period"

   If so which drops are you allowed 20 of?

[Dray]

I think that is 20 resyncs.

[Chrysalis]

les you have lost me.

Of course they can detect resyncs

LOS UAS or they must log every new sync event and count that as a resync.

[les-70]

  Sorry to loose you.  When I said "This would mean that a resync would in effect only be noticed via the error count.  Unless you have 20+ a day of course! " I am assuming that the resyncs are counted but that your allowed 20 a day.  To me 20 day is like not really counting them.  If I lost conection 5 times a day i would regard that as dreadful.  I would go nuts before 20 a day!

[Chrysalis]

yeah its leniant but remember it is the speed(standard) profile not stable.

yet some still get caught out by it which is odd, power cut, swapping modem then hit by DLM.