Afrihost Uncapped ADSL Feedback

Status
Not open for further replies.
Thinking of getting the 1mb uncapped with the adsl circuit. Are the speeds decent and what's the average you download per month?

(I'm movIng from Cybersmart, the speeds suck)
 
AfriMan, is it safe to assume that since you guys haven't yet been able to figure out what is wrong with Diablo 3 latencies, we are stuck with it like this indefinitely?

Tonight was absolutely abysmal, ranging from 600ms to 1100ms, and at times it was a slide-show. Switched to FNB connect and it was 100% as usual.

Hell, I will donate my FNB Connect account to your techies if it will help them trouble-shoot the problem! In any event, it would be nice to know if the official word on this is "Sorry, we tried, we gave up. Don't play Diablo on Afrihost."
 
al4bb said:
Not Odd. You're in JHB.
Try CT or Durban and see how it is to be routed around the country.
Click to expand...

I am in Cape Town on a Axxess 10Mbps uncapped account:
2273718995.png

2273721296.png

While latencies aint great back to Cape Town, the speeds are fine. For now I will be gaming on my Web Africa account.


Still, I am unhappy that I have yet to see anything about this problem: (routing from SAIX to this new ip range goes international)
Code: 
tpr-route-server>traceroute 105.237.19.173

Type escape sequence to abort.
Tracing the route to 105.237.19.173

  1 tpr-ip-esr-1-fe-1-0-3.telkom-ipnet.co.za (196.25.9.45) 0 msec 0 msec 0 msec
  2 ams-ip-hsll-1-gig-0-1-0.telkom-ipnet.co.za (196.43.9.82) 208 msec 208 msec 208 msec
  3 213.19.192.1 [AS 3356] 204 msec 200 msec 204 msec
  4 ae-51-51.csw1.Amsterdam1.Level3.net (4.69.139.153) [AS 3356] [MPLS: Label 1449 Exp 0] 216 msec 208 msec 216 msec
  5 4.69.162.130 [AS 3356] 204 msec
    4.69.162.142 [AS 3356] 204 msec
    4.69.162.134 [AS 3356] 204 msec
  6 MTN-NIGERIA.edge5.Amsterdam1.Level3.net (212.72.41.90) [AS 3356] 232 msec 212 msec 212 msec
  7  *  *  *
  8  *  *  *
  9 tb-cr-2.za--ls-cr-1.uk.mtnns.net (196.44.31.112) [AS 16637] [MPLS: Label 715678 Exp 0] 204 msec 200 msec 200 msec
 10 196.44.31.106 [AS 16637] [MPLS: Label 414703 Exp 0] 208 msec
    tb-cr-2.za--tb-cr-1.za-a.mtnns.net (196.44.31.5) [AS 16637] [MPLS: Label 493264 Exp 0] 200 msec
    196.44.31.106 [AS 16637] [MPLS: Label 347311 Exp 0] 200 msec
 11 ct-cr-2.za--ct-cr-1.za-a.mtnns.net (196.44.31.108) [AS 16637] [MPLS: Label 777444 Exp 1] 200 msec 200 msec 204 msec
 12 rb-cr-1.za--ct-cr-2.za-b.mtnns.net (196.44.31.250) [AS 16637] [MPLS: Label 405605 Exp 1] 400 msec
    ct-cr-2.za--rb-cr-2.za-a.mtnns.net (196.44.31.30) [AS 16637] [MPLS: Label 336630 Exp 0] 400 msec
    rb-cr-1.za--ct-cr-2.za-b.mtnns.net (196.44.31.250) [AS 16637] [MPLS: Label 405637 Exp 1] 404 msec
 13 jh-cr-2.za--rb-cr-1.za-a.mtnns.net (196.44.31.95) [AS 16637] [MPLS: Label 666256 Exp 1] 400 msec 212 msec 204 msec
 14 jh-dca-2.za--jh-cr-2.za-a.mtnns.net (196.44.0.225) [AS 16637] 204 msec 200 msec
    41.181.180.11 [AS 16637] 248 msec
 15  *  *  *
 16 ipc-send-jh-2a.za.mtnbusiness.net (41.181.178.21) [AS 16637] 248 msec *  *
 17  *
    rb-cr-1.za--jh-cr-2.za-a.mtnns.net (196.44.31.94) [AS 16637] [MPLS: Labels 308704/507 Exp 1] 232 msec 284 msec
 18 41.181.178.69 [AS 16637] 252 msec *  *
 19  *  *  *
 20  *  *  *
 21  *  *  *
 
Routing from Europe to the new IP ranges is very broken:


TeliaSonera Looking Glass - traceroute inet mylocalmachine.dyndns.org as-number-lookup

Router: Amsterdam
Command: traceroute inet mylocalmachine.dyndns.org as-number-lookup


traceroute to mylocalmachine.dyndns.org (105.236.196.x), 30 hops max, 40 byte packets
1 te4-1-10G.ar2.AMS1.gblx.net (64.215.195.173) [AS 3549] 8.085 ms te8-3-10G.ar2.AMS1.gblx.net (64.215.195.17) [AS 3549] 0.503 ms 0.456 ms
2 MTN-DUBAI-LIMITED.TenGigabitEthernet1-4.ar8.AMS2.gblx.net (208.48.1.110) [AS 3549] 0.964 ms 1.284 ms 0.951 ms
3 am-tpr-1.nl--am-cr-2.nl-a.mtn.net (209.212.111.219) [AS 16637] 1.023 ms 1.050 ms 1.095 ms
MPLS Label=570730 CoS=0 TTL=1 S=1
4 ct-nt-1--rb-nt-1-b.mtnns.net (196.44.31.154) [AS 16637] 8.894 ms 9.042 ms 8.919 ms
MPLS Label=437777 CoS=0 TTL=1 S=1
5 tb-cr-2.za--ls-cr-1.uk.mtnns.net (196.44.31.240) [AS 16637] 157.069 ms 160.310 ms tb-cr-2.za--ls-cr-1.uk.mtnns.net (196.44.31.112) [AS 16637] 156.819 ms
MPLS Label=724686 CoS=0 TTL=1 S=1
6 tb-cr-2.za--tb-cr-1.za-a.mtnns.net (196.44.31.5) [AS 16637] 157.347 ms 157.008 ms 196.44.31.106 (196.44.31.106) [AS 16637] 157.463 ms
MPLS Label=419839 CoS=0 TTL=1 S=1
7 196.44.31.106 (196.44.31.106) [AS 16637] 157.321 ms ct-cr-2.za--ct-cr-1.za-a.mtnns.net (196.44.31.108) [AS 16637] 157.385 ms 157.298 ms
MPLS Label=783396 CoS=1 TTL=1 S=1
8 rb-cr-1.za--ct-cr-1.za.mtnns.net (196.44.31.128) [AS 16637] 189.569 ms rb-cr-1.za--ct-cr-2.za-b.mtnns.net (196.44.31.250) [AS 16637] 209.500 ms 191.281 ms
MPLS Label=653477 CoS=1 TTL=1 S=1
9 jh-cr-2.za--rb-cr-1.za-a.mtnns.net (196.44.31.95) [AS 16637] 184.740 ms jh-cr-1.za--rb-cr-2.za.mtnns.net (196.44.0.42) [AS 16637] 184.939 ms 210.173 ms
MPLS Label=389666 CoS=0 TTL=1 S=1
10 41.181.180.11 (41.181.180.11) [AS 16637] 191.059 ms jh-dca-2.za--jh-cr-2.za-a.mtnns.net (196.44.0.225) [AS 16637] 191.056 ms 41.181.180.11 (41.181.180.11) [AS 16637] 191.134 ms
11 jh-cr-2.za--jh-dca-2.za-a.mtnns.net (196.44.0.224) [AS 16637] 189.904 ms * *
MPLS Label=562736 CoS=1 TTL=1 S=0
MPLS Label=308704 CoS=1 TTL=1 S=0
MPLS Label=507 CoS=1 TTL=1 S=1
12 * 41.181.178.33 (41.181.178.33) [AS 16637] 225.152 ms jh-cr-2.za--jh-dca-2.za-a.mtnns.net (196.44.0.224) [AS 16637] 205.666 ms
MPLS Label=562736 CoS=1 TTL=1 S=0
MPLS Label=308704 CoS=1 TTL=1 S=0
MPLS Label=507 CoS=1 TTL=1 S=1
13 * rb-cr-1.za--jh-cr-2.za-a.mtnns.net (196.44.31.94) [AS 16637] 219.818 ms 209.503 ms
MPLS Label=308704 CoS=1 TTL=1 S=0
MPLS Label=508 CoS=1 TTL=2 S=1
14 * ipc-send-rb-3a.za.mtnbusiness.net (41.181.178.81) [AS 16637] 207.006 ms *


Too many timeouts, and the routes look very unstable as well.
 
Compare to routing to another MTNNS connection:

traceroute to www.fnb.co.za (196.11.125.149), 30 hops max, 40 byte packets
1 abovenet-ic-139683-adm-b5.c.telia.net (80.239.135.10) 0.397 ms 0.393 ms 0.316 ms
2 ge-3-3-0.mpr1.ams1.nl.above.net (64.125.25.13) [AS 6461] 0.671 ms 42.163 ms 0.745 ms
MPLS Label=768435 CoS=0 TTL=1 S=1
3 ge-6-1-0.mpr1.lhr3.uk.above.net (64.125.22.138) [AS 6461] 7.247 ms 7.303 ms 7.237 ms
MPLS Label=671329 CoS=0 TTL=1 S=1
4 xe-2-0-0.mpr2.lhr3.uk.above.net (64.125.28.93) [AS 6461] 7.127 ms 7.160 ms 7.118 ms
5 213.152.234.229.available.above.net (213.152.234.229) [AS 6461] 7.374 ms 7.261 ms 7.289 ms
6 ls-cr-2.uk--ls-pr-1.uk-a.mtn.net (209.212.111.200) [AS 16637] 7.380 ms 12.638 ms 31.190 ms
MPLS Label=602029 CoS=1 TTL=1 S=1
7 rb-cr-1.za--ls-cr-2.uk-c.mtnns.net (196.44.31.84) [AS 16637] 240.745 ms 217.721 ms 197.533 ms
MPLS Label=308656 CoS=1 TTL=1 S=1
8 rb-mse-1.za--rb-cr-1.za-a.mtnns.net (196.44.0.181) [AS 16637] 197.097 ms 197.393 ms 199.483 ms
9 196.31.44.186 (196.31.44.186) [AS 16637] 197.343 ms 199.057 ms 201.950 ms


Why does FNB traffic go straight to above.net, while Afrihost traffic bounces around to Dubai?
 
StNick said:
AfriMan, is it safe to assume that since you guys haven't yet been able to figure out what is wrong with Diablo 3 latencies, we are stuck with it like this indefinitely?

Tonight was absolutely abysmal, ranging from 600ms to 1100ms, and at times it was a slide-show. Switched to FNB connect and it was 100% as usual.

Hell, I will donate my FNB Connect account to your techies if it will help them trouble-shoot the problem! In any event, it would be nice to know if the official word on this is "Sorry, we tried, we gave up. Don't play Diablo on Afrihost."
Click to expand...

No, we haven't given up, and we don't expect you to either. We'll keep working on this, I know MTN's support team are working on this 24/7. We do take this seriously and this is a high priority for us.
 
Slow Speeds

My speeds are bad at the moment, i cant even load the speedtest.net page.
Im on a 4MB package. AfriMan are there network issues or is it just my account?
 
craigg123 said:
My speeds are bad at the moment, i cant even load the speedtest.net page.
Im on a 4MB package. AfriMan are there network issues or is it just my account?
Click to expand...

Best to send me a PM so we can run through the various possibilities.

Network is running fine for me on this side, but we'd need to verify everything from your specific setup.
 
Tinuva said:
Afriman,

When will the routing from SAIX to the new IP range be sorted out?
Click to expand...

I've sent the results through to our Dev team and they are analyzing this to make suggestions to the MTN network guys. We are slowly chipping way at all the last few issues until everyone is getting great performance :)
 
al4bb said:
Not Odd. You're in JHB.
Try CT or Durban and see how it is to be routed around the country.
Click to expand...

No, odd. I'm in Cape Town, at least last time I checked anyway. Same city as the poster of the post I commented on.
 
Last edited:
@AfriMan - A friend of mine has spent a good portion of his leave trying to get back to the latency he was on prior to the move. He has now forwarded me a mail from Afrihost explaining that the ping (45ms local / 260ms international) he is getting is acceptable, and nothing more need be done. Even though the ping he now has is nowhere near what he had before the move (16ms local / 165ms international).

I am in the exact same boat, but have less time to spend with support. Is there even a point to being hopeful that we will get our once decent Pings back anytime in the next month, or are we now being officially told that we've been screwed, and we must just suck it up cause things are acceptable, even though we are worse off?

[email protected] said:
The tracert you have sent seems perfect. The time for international will take from 200 to 270.
Click to expand...
 
[s]uNq;9216288 said:
@AfriMan - A friend of mine has spent a good portion of his leave trying to get back to the latency he was on prior to the move. He has now forwarded me a mail from Afrihost explaining that the ping (45ms local / 260ms international) he is getting is acceptable, and nothing more need be done. Even though the ping he now has is nowhere near what he had before the move (16ms local / 165ms international).

I am in the exact same boat, but have less time to spend with support. Is there even a point to being hopeful that we will get our once decent Pings back anytime in the next month, or are we now being officially told that we've been screwed, and we must just suck it up cause things are acceptable, even though we are worse off?
Click to expand...

The pings will improve once they have IPC running in other cities and not only in JHB.
 
[s]uNq;9216288 said:
@AfriMan - A friend of mine has spent a good portion of his leave trying to get back to the latency he was on prior to the move. He has now forwarded me a mail from Afrihost explaining that the ping (45ms local / 260ms international) he is getting is acceptable, and nothing more need be done. Even though the ping he now has is nowhere near what he had before the move (16ms local / 165ms international).

I am in the exact same boat, but have less time to spend with support. Is there even a point to being hopeful that we will get our once decent Pings back anytime in the next month, or are we now being officially told that we've been screwed, and we must just suck it up cause things are acceptable, even though we are worse off?
Click to expand...

We are aware of the problems and we know the current performance is not ideal for gaming. We are doing everything possible to get this improved as soon as possible. I'll be straight though - IPC in Cape Town is not going to happen this year, and we are exploring every possible route to getting a better solution until we do have that in place. What we have planned for that stage is pretty awesome, with upgrades across the board based on current usage and projected growth. We are slowly chipping away at he problem, and quite a few people have reported that gaming is now stable for them. I'm confident that we'll get there. We sometimes forget that this network is only 14 days old, and how far we've come in terms of performance in that short time. In another 14 days, I think we'll be a lot closer to our goal of the premium ADSL network in SA.
 
AfriMan said:
We are aware of the problems and we know the current performance is not ideal for gaming. We are doing everything possible to get this improved as soon as possible. I'll be straight though - IPC in Cape Town is not going to happen this year, and we are exploring every possible route to getting a better solution until we do have that in place. What we have planned for that stage is pretty awesome, with upgrades across the board based on current usage and projected growth. We are slowly chipping away at he problem, and quite a few people have reported that gaming is now stable for them. I'm confident that we'll get there. We sometimes forget that this network is only 14 days old, and how far we've come in terms of performance in that short time. In another 14 days, I think we'll be a lot closer to our goal of the premium ADSL network in SA.
Click to expand...

Sorry to say, but AfriHost won't be a premium ADSL network with no IPC in Cape Town. Like I've said previously, you will have a bunch of unhappy clients because you guys omitted to tell Joe public that their pings and routes are going to be screwed if you live in Western Cape.

Good luck with assuming people are happy with "stable" pings ( Maybe for games like Peggle )

Not bashing AfriHost, but you guys need to sort out the IPC issue. There is no point in saying we have this massive amount of bandwidth, but we can't even give you guys what you had before moving over to a new network. This just doesn't make sense, and reeks of poor planning.
 
Status
Not open for further replies.
Back
Top
Sign up to the MyBroadband newsletter
X