Afrihost Fibre Feedback

Status
Not open for further replies.
Thanks - I do appreciate the efforts the social team is making to keep the communications open, but here's the thing:

An example of the incompetence on the AH support side:
Your so-called fibre supervisor/team-lead repeatedly asked me to test the speed when connected directly to the ONT via LAN cable and send speed tests - this I did REPEATEDLY; the speeds remained slow.
She then concluded that the AH network team has to 'troubleshoot my router'... :wtf: !!!!!
If the test via LAN cable directly to the ONT rendered the slow speeds with the router switched off and unplugged, what on earth does my router have to do with this??
That was the sum total of the 'Afrihost fibre team troubleshooting with me'.

The requests logged to Vuma was because AH believed there was something wrong on the line - this was not the case and was made abundantly clear right from the start, by myself and by Vuma and the Britelink technicians after round 1 testing. Yet AH mindlessly logged more tickets and Vuma/Britelink dispatched more technicians who once again confirmed that there was no issue on the line or equipment.

The 3rd round ticket to re-provision was after AH tried the same tactic for weeks on end always giving the same result and believe it or not, they were about to do exactly the same: dispatch yet another technician to test the line.
I contacted Vumatel directly and someone on their team explained some alternatives to me which I had to take back to Afrihost because they had to initiate the request. From the start, all my hints and suggestions that this was either a Vuma or Afrihost account profiling issue and NOT a line/equipment issue were blatantly ignored.

My suggestion to Afrihost would be that in stead of always defending your position, take a long, hard look into how situations like these are handled and use the opportunity to improve and up-skill your support teams to have them actually listen to customers as opposed to arguing with them and mindlessly following standard steps that often make absolutely no sense.

In addition, after 11 months, still no way to accurately track capped fibre data. :(

I totally feel your pain here. We've been chatting for a while and I can promise we're doing absolutely everything to get this sorted out for you.

Perhaps what we can also do is connect you with our QA department so you can give them your details and share your experiences. We're always looking for ways to improve.
 
We are definitely in support of moving to IPv6, and we've moved to 6 for our internal network. As I understand it (and people smarter than me should feel free to chip in) most websites are not accepting IPv6 addresses and it requires complex NAT to switch between 4 and 6 to ensure smooth connectivity. So we as an ISP are not switching our public IP's to IPv6 or offering that until we believe clients will have a good experience.

It is a chicken and egg situation. Websites are not going to convert to IPV6 if there is no demand for it. So we need to push from both ends. And I don't believe 6 to 4 NAT is a complex issue. You will have to run a dual stack into the future at some point in any case.
 
I totally feel your pain here. We've been chatting for a while and I can promise we're doing absolutely everything to get this sorted out for you.

Perhaps what we can also do is connect you with our QA department so you can give them your details and share your experiences. We're always looking for ways to improve.

Happy to help, but as per post #903, my experience is summarised in post #899. Once QA has made sense of that and there are any misunderstandings, they're welcome to contact me.
There are many examples like this across MyBB forums and I'm sure you guys in the social team have plenty to point the QA team towards.

There still are some exceptionally helpful and dedicated team members at Afrihost, but they seem to be spread out too thin while increasingly more rotten/useless bad apples are threatening to spoil the bunch.

May the pudding be the proof that it is being taken seriously and addressed appropriately or before you know it, all the 'safe hands' at Afrihost might be gone and the core rotten.
 
Happy to help, but as per post #903, my experience is summarised in post #899. Once QA has made sense of that and there are any misunderstandings, they're welcome to contact me.
There are many examples like this across MyBB forums and I'm sure you guys in the social team have plenty to point the QA team towards.

There still are some exceptionally helpful and dedicated team members at Afrihost, but they seem to be spread out too thin while increasingly more rotten/useless bad apples are threatening to spoil the bunch.

May the pudding be the proof that it is being taken seriously and addressed appropriately or before you know it, all the 'safe hands' at Afrihost might be gone and the core rotten.

They can definitely review, but they always prefer to deal directly with you so that they can get more details and specifics (to pull recordings and tickets) and also ensure that all aspects of the query are resolve before they close the matter.
 
It is a chicken and egg situation. Websites are not going to convert to IPV6 if there is no demand for it. So we need to push from both ends. And I don't believe 6 to 4 NAT is a complex issue. You will have to run a dual stack into the future at some point in any case.

We're keeping a close eye on it. We'll definitely be jumping onto it when the time is right.
 
Afrihost Fibre pings have been 30ms higher than they normally are ? Any reason for this ? Something broken ?
 
Afrihost Fibre pings have been 30ms higher than they normally are ? Any reason for this ? Something broken ?

I'm not aware of any widespread issues at the moment. To which site or service are you seeing this increased latency?
 
I'm not aware of any widespread issues at the moment. To which site or service are you seeing this increased latency?

Pings to most of the EU.

eu.battle.net use to ping at 173ms. Since yesterday, its now at 196ms.
 
Could you maybe post the full traceroute? :)

Code:
Tracing route to eu.battle.net [185.60.115.40]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  dlinkap.local [192.168.0.1]
  2     2 ms     2 ms     2 ms  165-73-34-65.ip.afrihost.co.za [165.73.34.65]
  3     3 ms     2 ms     2 ms  169-1-21-101.ip.afrihost.co.za [169.1.21.101]
  4     3 ms     3 ms     2 ms  41.169.57.218
  5     4 ms     2 ms     3 ms  ix-ae-7-0.tcore1.JSO-Johannesburg.as6453.net [41.206.178.29]
  6   190 ms   190 ms   190 ms  if-ae-4-2.tcore1.KLT-Cape-Town.as6453.net [41.206.178.2]
  7   197 ms   197 ms   196 ms  if-ae-5-2.tcore2.PV9-Lisbon.as6453.net [80.231.159.93]
  8   197 ms   197 ms   197 ms  if-ae-16-2.tcore2.WV6-Madrid.as6453.net [80.231.159.55]
  9   197 ms   197 ms   197 ms  if-ae-11-2.tcore1.WV6-Madrid.as6453.net [80.231.91.65]
 10   197 ms   197 ms   197 ms  if-ae-5-2.tcore2.WYN-Marseille.as6453.net [80.231.200.5]
 11   198 ms   198 ms   197 ms  if-ae-2-2.tcore1.WYN-Marseille.as6453.net [80.231.217.1]
 12   197 ms   197 ms   199 ms  if-ae-8-1600.tcore1.PYE-Paris.as6453.net [80.231.217.6]
 13   197 ms   197 ms   197 ms  if-ae-11-2.tcore1.PVU-Paris.as6453.net [80.231.153.49]
 14     *        *        *     Request timed out.
 15     *        *        *     Request timed out.
 16   197 ms   197 ms   198 ms  BLIZZARD-EN.ear2.Paris1.Level3.net [212.73.205.158]
 17   198 ms   197 ms   197 ms  ae1-br02-eqpa4.blizzardonline.net [137.221.77.35]
 18   198 ms   198 ms   198 ms  be2-pe01-eqpa4.blizzardonline.net [137.221.77.71]
 19   198 ms   198 ms   198 ms  137.221.66.35
 20     *        *        *     Request timed out.
 21   197 ms   197 ms   197 ms  185.60.115.40

Looks like its the first international hop. That first high latency hop, use to be 170-180ms. And as you can see now, its pretty much 200ms.
 
Code:
Tracing route to eu.battle.net [185.60.115.40]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  dlinkap.local [192.168.0.1]
  2     2 ms     2 ms     2 ms  165-73-34-65.ip.afrihost.co.za [165.73.34.65]
  3     3 ms     2 ms     2 ms  169-1-21-101.ip.afrihost.co.za [169.1.21.101]
  4     3 ms     3 ms     2 ms  41.169.57.218
  5     4 ms     2 ms     3 ms  ix-ae-7-0.tcore1.JSO-Johannesburg.as6453.net [41.206.178.29]
  6   190 ms   190 ms   190 ms  if-ae-4-2.tcore1.KLT-Cape-Town.as6453.net [41.206.178.2]
  7   197 ms   197 ms   196 ms  if-ae-5-2.tcore2.PV9-Lisbon.as6453.net [80.231.159.93]
  8   197 ms   197 ms   197 ms  if-ae-16-2.tcore2.WV6-Madrid.as6453.net [80.231.159.55]
  9   197 ms   197 ms   197 ms  if-ae-11-2.tcore1.WV6-Madrid.as6453.net [80.231.91.65]
 10   197 ms   197 ms   197 ms  if-ae-5-2.tcore2.WYN-Marseille.as6453.net [80.231.200.5]
 11   198 ms   198 ms   197 ms  if-ae-2-2.tcore1.WYN-Marseille.as6453.net [80.231.217.1]
 12   197 ms   197 ms   199 ms  if-ae-8-1600.tcore1.PYE-Paris.as6453.net [80.231.217.6]
 13   197 ms   197 ms   197 ms  if-ae-11-2.tcore1.PVU-Paris.as6453.net [80.231.153.49]
 14     *        *        *     Request timed out.
 15     *        *        *     Request timed out.
 16   197 ms   197 ms   198 ms  BLIZZARD-EN.ear2.Paris1.Level3.net [212.73.205.158]
 17   198 ms   197 ms   197 ms  ae1-br02-eqpa4.blizzardonline.net [137.221.77.35]
 18   198 ms   198 ms   198 ms  be2-pe01-eqpa4.blizzardonline.net [137.221.77.71]
 19   198 ms   198 ms   198 ms  137.221.66.35
 20     *        *        *     Request timed out.
 21   197 ms   197 ms   197 ms  185.60.115.40

Looks like its the first international hop. That first high latency hop, use to be 170-180ms. And as you can see now, its pretty much 200ms.

Thanks! Which DNS servers are you using at the moment? Our default DNS?
 
Thanks! Which DNS servers are you using at the moment? Our default DNS?

Address resolves, stop blaming DNS. Typical Afrihost.
Game server the address will be resolved once, past that point it's cached, DNS is out of the picture.
 
Thanks! Has there been any change this evening in the latency?



Nope, still the same...

More strange, is its not ALL of EU. There is still some amazon nodes i can ping at 170, so its clearing a routing issue.

Reply from 35.176.0.252: bytes=32 time=174ms TTL=231 ( Amazon EC2 London node)
 
Nope, still the same...

More strange, is its not ALL of EU. There is still some amazon nodes i can ping at 170, so its clearing a routing issue.

Reply from 35.176.0.252: bytes=32 time=174ms TTL=231 ( Amazon EC2 London node)

Pinging London at 174ms is not bad though. Anything under 200ms should still deliver a good experience.

We might need you to give us exact steps and settings so we can try to replicate your experience and analyse further.
 
I think that's the good node he is talking about, afriman.... Conveniently it's also blizzard/battle.net that he's having issues with.
 
I think that's the good node he is talking about, afriman.... Conveniently it's also blizzard/battle.net that he's having issues with.

We need to see the whole trace to see which hop is out of place. It could be a specific piece of kit on our network or a network we peer with overseas that we'd need to log a ticket with.
 
We need to see the whole trace to see which hop is out of place. It could be a specific piece of kit on our network or a network we peer with overseas that we'd need to log a ticket with.

If i wanted 200ms, i would of kept with my adsl, it has less issues.

I gave you the tracert, and i even gave you the hop that is giving the issues and is clearly have issues. Hell, you can google the ip of that hop, and see COUNTLESS REPORTS of issues with it, ranging back YEARS.

I gave you the ping to A DIFFERENT EU location, an AMAZON ECS SERVER, LOCATED IN LONDON. That pings correctly, hence its getting routed correctly. This is to show, it isnt my first mile fibre.

Get it fixed, and stop making excuses, thanks. Im not some idiot who doesnt understand these things and you can give the round about.
 
Last edited:
Status
Not open for further replies.
Top
Sign up to the MyBroadband newsletter
X