GWISA=SCREAMER

rebel998 said:
There was a problem in PTA yesterday up till this morning.

Could you enlighten us as to what caused this Howzit?:D
Click to expand...

The Hillensberg tower was destroyed by lightning but everything was replaced by this am.
 
wow4life said:
Here with good and bad trace:

(A) Good trace

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms xxxxxxxxx.home [192.168.2.1]
2 60 ms 50 ms 49 ms 10.0.64.253
3 61 ms 49 ms 49 ms 10.254.253.254

Trace complete.

(B) DC 1 trace

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 4 ms <1 ms <1 ms xxxxxxxxxx.home [192.168.2.1]
2 60 ms 45 ms 54 ms 10.0.64.253
3 10.0.64.253 reports: Destination host unreachable.

Trace complete.

(C) DC 2 trace

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms xxxxxxxx.home [192.168.2.1]
2 * * * Request timed out.
3 * * * Request timed out.
4 * * 61 ms 10.254.253.254

Trace complete.

(D) DC 3 trace

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 <1 ms <1 ms <1 ms xxxxxxxxxxxx.home [192.168.2.1]
2 * * * Request timed out.
3 * * * Request timed out.
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 76 ms 67 ms 60 ms 10.254.253.254

Trace complete.
Click to expand...


Hi

Just run the same traces and got these results:

A.

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 3 ms 1 ms 1 ms 192.168.1.1
2 54 ms 55 ms 52 ms 10.0.64.253
3 53 ms 54 ms 65 ms 10.254.253.254

Trace complete.

B.

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 2 ms 1 ms 1 ms 192.168.1.1
2 57 ms 58 ms 67 ms 10.0.64.253
3 65 ms 54 ms 58 ms 10.254.253.254

Trace complete.


I will try and establish why you are getting these strange results.
 
I get those results when all is working well ...

The traces that show failure are the ones that I run when I notice that I am disconnecting. I was hoping that it would show that there was one hop failing all the time, which would hopefully point to the solution.
 
Just happened whilst surfing ....

Tracing route to 10.254.253.254 over a maximum of 30 hops

1 4 ms <1 ms <1 ms xxxxxxxxxxx.home [192.168.2.1]
2 65 ms 51 ms 50 ms 10.0.64.253
3 10.0.64.253 reports: Destination host unreachable.

Trace complete.
 
Connection down as of midnight tonight (using 3G atm) and still down. Seems to be the link into the Telkom backbone as I can get to the edge of your network - and can also open www.screamer.co.za - but no further.

Support line is unavailable btw - so don't know what is happening.
 
wow4life said:
Connection down as of midnight tonight (using 3G atm) and still down. Seems to be the link into the Telkom backbone as I can get to the edge of your network - and can also open www.screamer.co.za - but no further.

Support line is unavailable btw - so don't know what is happening.
Click to expand...

Same here. My connection died at around 00:30 and only came back on this morning at about 7.:D
 
Can we please get a support number for the team that know what is happening with the WiMax network? The staff on the advertised number tell me they support WiFi?.

Thanks
 
Oh .... and lots of disconnects today. Approx every 5-10 mins :/

The usual .....

1 <1 ms 4 ms <1 ms xxxxx.home [192.168.2.1]
2 51 ms 39 ms 39 ms 10.0.64.253
3 10.0.64.253 reports: Destination host unreachable.
 
Local improvement - and some on the international side

howzit said:
Hi there Murray 654

Please PM me your account details so I can check why you are having a problem.

Regards
Click to expand...

We have been visited by the Harrismith agent. He has upgraded us from 256k to 512k

This has improved the connection to Harrismith. We now get 5-10% packet loss. Previously this was 10-25%.

There is a smaller improvement to International connections:

Pinging google.fr I get these results:
66 of 100 packets received 34% packet loss min 607ms Avg 654ms max 977ms
81 of 100 packets received 19% packet loss min 682ms Avg 718ms max 984ms

Hopefully this improves further when the Seacom cable comes online... then Screamer can get better access to international bandwidth...

Thanks for the improvement. I look forward to further improvements...
 
Last edited:
murray654 said:
We have been visited by the Harrismith agent. He has upgraded us from 256k to 512k

This has improved the connection to Harrismith. We now get 5-10% packet loss. Previously this was 10-25%.

There is a smaller improvement to International connections:

Pinging google.fr I get these results:
66 of 100 packets received 34% packet loss min 607ms Avg 654ms max 977ms
81 of 100 packets received 19% packet loss min 682ms Avg 718ms max 984ms

Hopefully this improves further when the Seacom cable comes online... then Screamer can get better access to international bandwidth...

Thanks for the improvement. I look forward to further improvements...
Click to expand...

Packetloss of any kind is not good. I don't understand how people can just shrug off 5% packetloss.
You should not accept this level of "service".
 
There are only 2 products that offer < 5% packet loss in our area

1. Diginet
2. ISDN

Last time I checked, Diginet would cost R100 per km per month line rental - we are around 100km from any service provider, so it would cost R10000 per month in line rental alone.

ISDN we use sometimes when all else fails. However it would cost us R2500 per month in call charges if we used it as our primary internet connection. And that is 64k - if we used 128k it would cost us double.

So for R570 per month I think I will live with 5% packet loss.

I think if you live in an area with 3G and ADSL you can afford to be fussy about packet loss. We don't have the luxury out here in the sticks.
 
murray654 said:
There are only 2 products that offer < 5% packet loss in our area

1. Diginet
2. ISDN

Last time I checked, Diginet would cost R100 per km per month line rental - we are around 100km from any service provider, so it would cost R10000 per month in line rental alone.

ISDN we use sometimes when all else fails. However it would cost us R2500 per month in call charges if we used it as our primary internet connection. And that is 64k - if we used 128k it would cost us double.

So for R570 per month I think I will live with 5% packet loss.

I think if you live in an area with 3G and ADSL you can afford to be fussy about packet loss. We don't have the luxury out here in the sticks.
Click to expand...

Or, you could stop being shafted, and complain like hell to your "Service" provider.
 
Spoke too soon

murray654 said:
We have been visited by the Harrismith agent. He has upgraded us from 256k to 512k

This has improved the connection to Harrismith. We now get 5-10% packet loss. Previously this was 10-25%....

... I look forward to further improvements...
Click to expand...

They put us back on 256k and now we have 18-20% packet loss when we ping the gateway - This sucks. :(
 
:( Pta North towers 1 to 3 down again. Does anyone know when I can surf the net again?
 
18-20% packet loss?
Thats not a useable internet connection!
When I have setup anything wireless I dont accept any packet loss. a slightly higher ping with no loss is better than a lower ping with loss.
Daffy might be able to correct me on this but im pretty sure the WUG links dont drop packets like that.
 
Packet Loss explained

Rocket-Boy said:
18-20% packet loss?
Thats not a useable internet connection!
When I have setup anything wireless I dont accept any packet loss. a slightly higher ping with no loss is better than a lower ping with loss.
Daffy might be able to correct me on this but im pretty sure the WUG links dont drop packets like that.
Click to expand...

the local agent explained how a routerboard got damaged and they are running 4 different high sites through a rb400 series - aparently too many customers connecting through the board and it is working too hard... Its been like this since November when I started measuring.

He says they are waiting for the 600 series routerboard - I am sure it does not take 3 months to arrive. Maybe if they visit uniterm direct they can pick one up - I see they are on special. But who am I to give advice. I am jsut the customer.

The agent put me back to 512k, but I still get 19% packet loss.
 
Earlier this month Screamer disconnected me for non payment. After phoning around and being told it's a signal problem(what a laugh!) I eventually spoke to the accounts people and after nicely apologising they reconnected me.

Then a couple of days ago I get an sms saying I've reached my cap.(even bigger laugh, as I had no connection for two out of the eight days or so the month was old!) After a call or two I'm told to ignore the sms.

Now again today, my account gets cut off! This after I in fact paid my account a month in advance!? Once again I had to make a few phone calls and only after calling Freeg the local agent on his cell did I manage to get re-connected again.

The beauty of Screamer's system is that they cut off your account late in the afternoon then no-one at support can help you. I wonder why its done this way?:D
 
Rocket-Boy said:
18-20% packet loss?
Thats not a useable internet connection!
When I have setup anything wireless I dont accept any packet loss. a slightly higher ping with no loss is better than a lower ping with loss.
Daffy might be able to correct me on this but im pretty sure the WUG links dont drop packets like that.
Click to expand...

Keep in mind that these high pings are caused by the fact that I'm pinging over a VPN (From Ireland to South Africa IS DSL). In reality the hops are about 3ms each, and there are 4 wireless hops in this.

traceroute to 172.16.250.100 (172.16.250.100), 30 hops max, 40 byte packets
1 172.16.0.193 (172.16.0.193) 0.578 ms 0.856 ms 1.152 ms
2 192.168.253.1 (192.168.253.1) 283.985 ms 286.710 ms 289.956 ms
3 172.16.15.30 (172.16.15.30) 292.680 ms 296.753 ms 300.196 ms
4 172.16.254.145 (172.16.254.145) 308.296 ms 311.633 ms 315.349 ms
5 172.16.255.138 (172.16.255.138) 540.625 ms 544.050 ms 547.939 ms
6 172.16.250.100 (172.16.250.100) 551.004 ms 537.022 ms 540.408 ms

Hop 2 is my VPN.
3-6 are over the Wug.
Its also peak period on the Wug, so the link between 4 and 5 is heavily saturated, hence the slighty elevated ping.
Keep in mind that the hop between 4 and 5 is possibly the busiest link on the Wug, and the one with the worst performance due bad to noise at both endpoints (its a popular route with the WISPs, from one end of JHB to the other)

And after all of that.

60 packets transmitted, 60 received, 0% packet loss, time 59216ms
rtt min/avg/max/mdev = 277.242/324.163/678.003/59.947 ms

No packetloss!

If we can do it, with 0 budget or funding, so can a WISP.
But they don't... <Insert reasons here>
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X