Afrihost Uncapped ADSL Feedback

Status
Not open for further replies.
Dude its got nothing to do with monitoring, or do you not understand the concept of TCP ack/syn ? The wow client and my computer are constantly talking that is how a TCP connection works. The issue is afrihosts network nothing WoW as WoW runs perfectly fine when i am using my other FNB account.

My tcp level is about 50/375 :o

What I'm asking is:

Client connects to server as 80.90.10.1:123 and your surfing to 123.123.123.0:80 is not going to keep alive the client. *If* the 80.xxx could have been closed due to being seen as inactive, this AFAIK can be picked up by sniffing the packets.
 
My tcp level is about 50/375 :o

What I'm asking is:

Client connects to server as 80.90.10.1:123 and your surfing to 123.123.123.0:80 is not going to keep alive the client. *If* the 80.xxx could have been closed due to being seen as inactive, this AFAIK can be picked up by sniffing the packets.

You are failing to understand how TCP works, if i am connected to X then my connection is not inactive. If X is sending packets through it the end point to kinda means there is an active and live connection is there not? The simple fact is MTN is the worst backhaul provider in SA and there is a lot of packet loss on their network. I lost tons of packets this weekend.
 
You are failing to understand how TCP works, if i am connected to X then my connection is not inactive. If X is sending packets through it the end point to kinda means there is an active and live connection is there not? The simple fact is MTN is the worst backhaul provider in SA and there is a lot of packet loss on their network. I lost tons of packets this weekend.

Ok,
Yesterday I quite a few of those (see pathping)
http://mybroadband.co.za/vb/showthr...DSL-Feedback?p=9292796&viewfull=1#post9292796
 
Right, got my monitoring up and running again, let's see what it does when the connection dies for a couple of seconds.

Ok I've got one of the D3 server's IP's on the monitor now, any other IP's anyone wants me to monitor?
 
Last edited:
Right, got my monitoring up and running again, let's see what it does when the connection dies for a couple of seconds.

Ok I've got one of the D3 server's IP's on the monitor now, any other IP's anyone wants me to monitor?

Yes...the two MTN DNS servers...I am seeing these go offline and would like someone to confirm.
 
Just had a DC from D3

Strangely I only lost packets to the D3 server

Code:
	Ping 6	Ping 6	Ping 6
Date Time	Ping Time	Downtime	Coverage
2012-11-12 12:03:18 PM	200 msec	0 %	100 %
2012-11-12 12:02:18 PM	217 msec	0 %	100 %
2012-11-12 12:01:18 PM	201 msec	0 %	100 %
2012-11-12 12:00:27 PM	Error	Error	100 %
2012-11-12 12:00:22 PM	Error	Error	100 %
2012-11-12 11:59:18 AM	201 msec	0 %	100 %
2012-11-12 11:58:18 AM	201 msec	0 %	100 %
2012-11-12 11:57:18 AM	201 msec	0 %	100 %
2012-11-12 11:56:18 AM	201 msec	0 %	100 %
2012-11-12 11:55:18 AM	201 msec	0 %	100 %
2012-11-12 11:54:18 AM	202 msec	0 %	100 %
2012-11-12 11:53:18 AM	201 msec	0 %	100 %
2012-11-12 11:52:18 AM	203 msec	0 %	100 %
2012-11-12 11:51:18 AM	204 msec	0 %	100 %
2012-11-12 11:50:18 AM	200 msec	0 %	100 %
2012-11-12 11:49:18 AM	227 msec	0 %	100 %
2012-11-12 11:48:56 AM	201 msec	0 %	100 %
2012-11-12 11:47:56 AM	201 msec	0 %	100 %
2012-11-12 11:46:56 AM	200 msec	0 %	100 %
2012-11-12 11:46:22 AM	201 msec	0 %	100 %
Sums			
Averages	204 msec	2174 %	19 %
 
Had an increase in RDP and SMTP response times just before the DC.

Code:
	RDP (Remote Desktop) 1	RDP (Remote Desktop) 1	RDP (Remote Desktop) 1
Date Time	Response time	Downtime	Coverage
2012-11-12 12:08:27 PM	159 msec	0 %	100 %
2012-11-12 12:07:27 PM	161 msec	0 %	100 %
2012-11-12 12:06:27 PM	173 msec	0 %	100 %
2012-11-12 12:05:27 PM	158 msec	0 %	100 %
2012-11-12 12:04:27 PM	173 msec	0 %	100 %
2012-11-12 12:03:27 PM	168 msec	0 %	100 %
2012-11-12 12:02:27 PM	158 msec	0 %	100 %
2012-11-12 12:02:08 PM	357 msec	0 %	100 %
2012-11-12 12:01:08 PM	360 msec	0 %	100 %
2012-11-12 12:00:08 PM	368 msec	0 %	100 %
2012-11-12 11:59:08 AM	364 msec	0 %	100 %
2012-11-12 11:58:08 AM	385 msec	0 %	100 %
2012-11-12 11:57:08 AM	381 msec	0 %	100 %
2012-11-12 11:56:08 AM	367 msec	0 %	100 %
2012-11-12 11:55:08 AM	364 msec	0 %	100 %
2012-11-12 11:54:08 AM	360 msec	0 %	100 %
2012-11-12 11:53:08 AM	375 msec	0 %	100 %
2012-11-12 11:52:08 AM	375 msec	0 %	100 %
2012-11-12 11:51:08 AM	355 msec	0 %	100 %
2012-11-12 11:50:08 AM	357 msec	0 %	100 %
2012-11-12 11:49:08 AM	353 msec	0 %	100 %
2012-11-12 11:48:08 AM	371 msec	0 %	100 %
2012-11-12 11:47:08 AM	155 msec	0 %	100 %
2012-11-12 11:46:14 AM	156 msec	0 %	100 %
2012-11-12 11:45:14 AM	171 msec	0 %	100 %
2012-11-12 11:44:14 AM	157 msec	0 %	100 %
2012-11-12 11:43:14 AM	156 msec	0 %	100 %
2012-11-12 11:42:14 AM	156 msec	0 %	100 %
Sums			
Averages	201 msec	82 %	100 %

Code:
	SMTP 1	SMTP 1	SMTP 1
Date Time	Response Time	Downtime	Coverage
2012-11-12 12:09:32 PM	100 msec	0 %	100 %
2012-11-12 12:08:32 PM	95 msec	0 %	100 %
2012-11-12 12:07:32 PM	101 msec	0 %	100 %
2012-11-12 12:06:32 PM	129 msec	0 %	100 %
2012-11-12 12:05:32 PM	106 msec	0 %	100 %
2012-11-12 12:04:32 PM	132 msec	0 %	100 %
2012-11-12 12:03:32 PM	112 msec	0 %	100 %
2012-11-12 12:02:32 PM	101 msec	0 %	100 %
2012-11-12 12:02:13 PM	337 msec	0 %	100 %
2012-11-12 12:01:13 PM	459 msec	0 %	100 %
2012-11-12 12:00:13 PM	418 msec	0 %	100 %
2012-11-12 11:59:13 AM	318 msec	0 %	100 %
2012-11-12 11:58:13 AM	342 msec	0 %	100 %
2012-11-12 11:57:13 AM	304 msec	0 %	100 %
2012-11-12 11:56:13 AM	296 msec	0 %	100 %
2012-11-12 11:55:13 AM	301 msec	0 %	100 %
2012-11-12 11:54:13 AM	304 msec	0 %	100 %
2012-11-12 11:53:13 AM	306 msec	0 %	100 %
2012-11-12 11:52:13 AM	406 msec	0 %	100 %
2012-11-12 11:51:13 AM	293 msec	0 %	100 %
2012-11-12 11:50:13 AM	298 msec	0 %	100 %
2012-11-12 11:49:13 AM	294 msec	0 %	100 %
2012-11-12 11:48:13 AM	289 msec	0 %	100 %
2012-11-12 11:47:13 AM	98 msec	0 %	100 %
2012-11-12 11:46:19 AM	118 msec	0 %	100 %
2012-11-12 11:45:19 AM	102 msec	0 %	100 %
2012-11-12 11:44:19 AM	101 msec	0 %	100 %
2012-11-12 11:43:19 AM	98 msec	0 %	100 %
2012-11-12 11:42:19 AM	94 msec	0 %	100 %
2012-11-12 11:41:19 AM	96 msec	0 %	100 %
2012-11-12 11:40:19 AM	97 msec	0 %	100 %
2012-11-12 11:39:19 AM	97 msec	0 %	100 %
2012-11-12 11:38:19 AM	96 msec	0 %	100 %
Sums			
Averages	249 msec	0 %	100 %
 
From everyone who have cancelled i wonder how mutch Afrihost now loose p/m from all the clients.
 
very frustrated this morning, different kind today, as ping is stable around 240ms and D/L 1.9 internationally. the bane is mini. DC's
times logged with everyone - 0605, 0630, 0655, 0720, 0750, 0810, 0830, 0850, 0920, 0955, 1025, 1040, 1105, 1125, 1140, 1155, 1215... atleast can plan my dailies around this but limits me to doing anything else
 
From everyone who have cancelled i wonder how mutch Afrihost now loose p/m from all the clients.

I wonder how many of these clients were "high usage" clients. It might actually not be such a bad thing for them if they are losing a few bandwidth hogs.

I was a fairly high usage user (did about 250GB in October), don't think they'll miss me too much now that I've left. I think however they really don't want to piss off the guys that pay for 10Mbit uncapped lines and then use like 20Gig for the month.

Not saying that losing any customers is good, but it is less of a problem if it's a low margin customer.
 
I wonder how many of these clients were "high usage" clients. It might actually not be such a bad thing for them if they are losing a few bandwidth hogs.

I was a fairly high usage user (did about 250GB in October), don't think they'll miss me too much now that I've left. I think however they really don't want to piss off the guys that pay for 10Mbit uncapped lines and then use like 20Gig for the month.

Not saying that losing any customers is good, but it is less of a problem if it's a low margin customer.

250GB isn't much on a 10Mb line. You should see what some people do.
 
Mine was on a 4MB line, which I think is not that high, but probably in the top 20-30% of users.

250gb is nothing these days, i was pulling 600gb when i was with mweb just because i wanted to burn it out. Last few months ive been doing around 80-100. Previous months i didnt even pass 40.
 
2302341352.png


pretty bad! international ping to google.com is between 470 and 500ms...

Afrihosts website says everything is ok again :/
 
Last edited:
Just had a DC from D3

Strangely I only lost packets to the D3 server

Code:
	Ping 6	Ping 6	Ping 6
Date Time	Ping Time	Downtime	Coverage
2012-11-12 12:03:18 PM	200 msec	0 %	100 %
2012-11-12 12:02:18 PM	217 msec	0 %	100 %
2012-11-12 12:01:18 PM	201 msec	0 %	100 %
2012-11-12 12:00:27 PM	Error	Error	100 %
2012-11-12 12:00:22 PM	Error	Error	100 %
2012-11-12 11:59:18 AM	201 msec	0 %	100 %
2012-11-12 11:58:18 AM	201 msec	0 %	100 %
2012-11-12 11:57:18 AM	201 msec	0 %	100 %
2012-11-12 11:56:18 AM	201 msec	0 %	100 %
2012-11-12 11:55:18 AM	201 msec	0 %	100 %
2012-11-12 11:54:18 AM	202 msec	0 %	100 %
2012-11-12 11:53:18 AM	201 msec	0 %	100 %
2012-11-12 11:52:18 AM	203 msec	0 %	100 %
2012-11-12 11:51:18 AM	204 msec	0 %	100 %
2012-11-12 11:50:18 AM	200 msec	0 %	100 %
2012-11-12 11:49:18 AM	227 msec	0 %	100 %
2012-11-12 11:48:56 AM	201 msec	0 %	100 %
2012-11-12 11:47:56 AM	201 msec	0 %	100 %
2012-11-12 11:46:56 AM	200 msec	0 %	100 %
2012-11-12 11:46:22 AM	201 msec	0 %	100 %
Sums			
Averages	204 msec	2174 %	19 %

Ping 6 Ping 6 Ping 6

That might be it :eek:
 
I wonder how many of these clients were "high usage" clients. It might actually not be such a bad thing for them if they are losing a few bandwidth hogs.

I was a fairly high usage user (did about 250GB in October), don't think they'll miss me too much now that I've left. I think however they really don't want to piss off the guys that pay for 10Mbit uncapped lines and then use like 20Gig for the month.

Not saying that losing any customers is good, but it is less of a problem if it's a low margin customer.

My client used around 40GB / month - they just wanted no suprises like being capped at the end of the month or not having to worry if a big update came along...
 
Status
Not open for further replies.
Top
Sign up to the MyBroadband newsletter
X