Afrihost - Pure Fibre Feedback Thread Part 2

William1102 said:
Is anyone having issues with YouTube or Google traffic in general it just dies for 5 minutes and comes back ?
Click to expand...
Yes i just opened Steam and its not loading thought to check reddit quick and guess what its not loading either.

Afrihost you've had a string of issues these past few weeks get your **** together please this is starting to annoy now.
 
Rayne said:
Yes i just opened Steam and its not loading thought to check reddit quick and guess what its not loading either.

Afrihost you've had a string of issues these past few weeks get your **** together please this is starting to annoy now.
Click to expand...
Checkers Sixty60 is also just loading, Netflix was working okay not the greatest.

Pinging google.com [142.251.47.110] with 32 bytes of data:
Request timed out.
Reply from 142.251.47.110: bytes=32 time=79ms TTL=117
Request timed out.
Reply from 142.251.47.110: bytes=32 time=82ms TTL=117

Ping statistics for 142.251.47.110:
Packets: Sent = 4, Received = 2, Lost = 2 (50% loss),
Approximate round trip times in milli-seconds:
Minimum = 79ms, Maximum = 82ms, Average = 80ms
 
AfriNatic said:
Primary NLD between DBN and JHb is seeing 90% packetloss. Our protection is not kicking in because its not hard down. We going to drop the affected link quickly so protection can kick in.

Apologies for the issues.
Click to expand...

Hi Afrinatic

Is it back to Normal cannot update apps on iPhone just never downloads
 
wingnut771 said:
Still the same, had to change to warp to watch f1tv:
View attachment 1861850
Click to expand...
Still ? Damn not all my devices can run warp so TVs have issues, Xbox etc and so many services run in Jhb only like Google and Apple. Even at work I’m struggling with VPN access as Vodacom 5G routes via Jhb

Not sure when they going to sort out the issue or do we just wait it out. I see Cool Ideas is also having complaints about the Dur->JHB NLD
 
AfriNatic said:
Primary NLD between DBN and JHb is seeing 90% packetloss. Our protection is not kicking in because its not hard down. We going to drop the affected link quickly so protection can kick in.

Apologies for the issues.
Click to expand...
Is this sorted out as i see i am still getting high pings to EU, I am on openserve
 
Last edited:
What I can understand is the delays in switching NLDs both Afrihost and Cool ideas customers are having the same issues and neither have got this resolved
 
William1102 said:
What I can understand is the delays in switching NLDs both Afrihost and Cool ideas customers are having the same issues and neither have got this resolved
Click to expand...

Hi

Apologies for the issues experienced last night. I wanted to give a quick breakdown of events.

At 17:05pm we lost 2 x NLD between DBN <> JHB at the same time. These links are protected which means when the primary link goes down our vendor automatically re-routes our traffic on a backup path. This happened as expected however the protected paths were instantly congested with the traffic as it was evening peak with high traffic levels.

Due to that our own backup protection didn't kick in. We forced traffic over onto our own backup path which did not improve at all as it runs on the same NLD as the primary protected paths and the congestion remained. We logged an urgent ticket with the vendors to attend and only saw services improve at 21:00 when one of the primary paths restored which eased the congestion.

Due to this NLD outage our transit to international from Durban was also partially impacted and we switched the forward path to our secondary transit provider. Its currently still on Openserve.

The same happened in JHB. Our primary transit provider Liquid telecoms has a direct JHB <> Marseille France route which was affected by the JHB <> DBN outage. This caused intermittent route changes where traffic would go via CPT to London and JHB to Marseille over the congested NLD via Durban. We also switch primary path for JHB to our secondary provider Openserve and its still on this for now.

Liquid has provided us with a detailed report on the outages however we are discussing internally ways to further reduce the risk of this happening again.
 
moms said:
Is this sorted out as i see i am still getting high pings to EU, I am on openserve
Click to expand...

Hi
You might still see slight increase in latency but there should be no loss. We are currently still on our secondary transit to Europe as a precaution just in case there is another event.
 
MainMeat said:
No much better - comparative results below.
Afrihost (500 / 250)
(London - BT)
View attachment 1860709

(Canada - Quebec - Rogers)
View attachment 1860710

(Canada - Bell)
View attachment 1860719

(Local - Afrihost)
View attachment 1860713


Telkom (300 / 150)
(London - BT)
View attachment 1860715

(Quebec - Rogers)
View attachment 1860716

(Canada - Bell)
View attachment 1860720

Local - Afrihost
View attachment 1860718
Click to expand...
Test speed now that we using openerve and not liquid?
 
T
AfriNatic said:
Hi

Apologies for the issues experienced last night. I wanted to give a quick breakdown of events.

At 17:05pm we lost 2 x NLD between DBN <> JHB at the same time. These links are protected which means when the primary link goes down our vendor automatically re-routes our traffic on a backup path. This happened as expected however the protected paths were instantly congested with the traffic as it was evening peak with high traffic levels.

Due to that our own backup protection didn't kick in. We forced traffic over onto our own backup path which did not improve at all as it runs on the same NLD as the primary protected paths and the congestion remained. We logged an urgent ticket with the vendors to attend and only saw services improve at 21:00 when one of the primary paths restored which eased the congestion.

Due to this NLD outage our transit to international from Durban was also partially impacted and we switched the forward path to our secondary transit provider. Its currently still on Openserve.

The same happened in JHB. Our primary transit provider Liquid telecoms has a direct JHB <> Marseille France route which was affected by the JHB <> DBN outage. This caused intermittent route changes where traffic would go via CPT to London and JHB to Marseille over the congested NLD via Durban. We also switch primary path for JHB to our secondary provider Openserve and its still on this for now.

Liquid has provided us with a detailed report on the outages however we are discussing internally ways to further reduce the risk of this happening again.
Click to expand...
Thanks for the update , recently afrihost and cool ideas has been having alot of issues as of late during weekends especially Saturday nights.

We not use to high latency on afrihost. 316ms to France on the secondary provider hope afrihost can sort the issues out and we can be back to normal soon
 
Local traffic on a 500 / 250 package from Afrihost - this is not great news:

1762761718553.png
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X