Assitance Needed re Current Service Levels

[Unchained]

Blackout in Menlo Park (CSIR) for at least 3 hours this afternoon. Anyone with precise start/end times?

No incident report.

I once queried this, the response was that the incident report is for things that have a greater impact than just a tower being down (can't remember the exact wording, but that was the obvious interpretation)

I still don't know what this means and what the criteria are for determining if an incident report should be provided.

Obviously a tower being down is of no consequence
It just happens to be a serious problem in Pretoria because large areas are only covered by one tower.

 

[regardtv]

Simple ... if they remember they'll post ...

I think the problem is fundemantal ... the incident posts are handled by 3rd level .. or the actual radio/ip engineers on duty ... they tend not to get to it ...

It should actually be a function of the 2nd level support person .. ie make 1 person per shift responsible for the maintenance of incident reports....CONTROL ppl control ...

Oh what the heck .. it's not like WBS give a **** anyway...

 

[Roman4604]

Guys... our traceroutes would be basically the same since we all get terminated at an LNS server inside the Gallo Manor PoP.

If this were correct, why are some ppl getting 196.2.120.1 as their 1st hop on other towers (e.g. Wonderboom), while I'm getting 196.2.100.1 off Kyalami?

Cant see why they would make the same LNS router display different IPs ... possibly they have more than one LNS?

BTW both IPs respond to telnet with a Cisco login prompt, but cant tell if they are the same or differing routers ... you would think they would know how to set console access lists in IOS ???

The link errors happen "out of band" between towers.

Agreed, the errors/degredation are happening on the 'hidden' L2TP forwarding network (probably 192.168.X.X range).

Funny how their main gateway (bw mgr ???) sits on a private addr -> 192.168.0.97 ... security possibly? This guy doesnt respond to telnet, but is running SSH (allows key exchange) ... me thinks possibly an open source box?

 

[guest2013-1]

Simple ... if they remember they'll post ...

I think the problem is fundemantal ... the incident posts are handled by 3rd level .. or the actual radio/ip engineers on duty ... they tend not to get to it ...

It should actually be a function of the 2nd level support person .. ie make 1 person per shift responsible for the maintenance of incident reports....CONTROL ppl control ...

Oh what the heck .. it's not like WBS give a **** anyway...

Hey regard, you probably forgot about the whole Sentech thing? I got out as soon as they started billing me incorrectly and having shat speeds. (which I should have done with Sentech from the beginning as well)

Never been happier.

 

[Roman4604]

Speeds are coming back to near normal (on Kyalami) ....

Download time: 5.197 seconds
Size of file: 500 KiloBytes
Estimated line speed: 785.1 (kilobits/second)
Estimated line speed: 96.2 (kiloBytes/second)

Maybe just cause it late & congestion is low?

 

[nocilah]

Speeds are coming back to near normal (on Kyalami) ....

Download time: 5.197 seconds
Size of file: 500 KiloBytes
Estimated line speed: 785.1 (kilobits/second)
Estimated line speed: 96.2 (kiloBytes/second)

Maybe just cause it late & congestion is low?

i am on kyalami and have noticed an improvement with local, but international is still degraded.

locally i get about 70-80KB/s and internationally I get about 50-60KB/s...

usually i get 100KB/s either way...

will phone them tommorrow and request a ticket number to get this sorted as the problem is still not 100% resolved.

is it all the rain?!?

 

[8321]

I once queried this, the response was that the incident report is for things that have a greater impact than just a tower being down (can't remember the exact wording, but that was the obvious interpretation)

I still don't know what this means and what the criteria are for determining if an incident report should be provided.

Obviously a tower being down is of no consequence
It just happens to be a serious problem in Pretoria because large areas are only covered by one tower.

Going back to the reason for this thread, if everyone took the trouble to report outages that are not 'incidents' here, we could build a pattern of service levels ourselves that could be useful later.

 

[regardtv]

They obviously have a few LNS servers ... what I was getting at is that AFAIK there are only 2 locations with lns ... JHB(Gallo Manor) and CPT(?) ...

Unless someone knows otherwise?