Afrihost - Pure Fibre Feedback Thread

Status
Not open for further replies.
Afrihost specific clients:

The issues with the new aggregation switches are causing intermittent drops on the NNI between Afrihost and FrogFoot. We are actively working with FrogFoot and their vendor on finding a solution ASAP.

We know these issues have caused extreme inconvenience and we do apologise for the issues experienced. We are working around the clock to mitigate the issue.
 
As it stands currently.

The Majority of Cape Town, Gqeberha and KZN is not experiencing issues.

The rest of the areas are experiencing intermittent connection.
 
Morning Please note the update from FrogFoot below.


We are aware of the ongoing connectivity outages that have affected our network in several parts of the country on Friday and over the weekend and are working toward resolving all issues as a matter of priority.
We are undergoing a major network upgrade, which included migrating a portion of our network from a previous equipment vendor, Extreme Networks, to our current equipment vendor Juniper Networks, and this has resulted in some teething issues that are continuing. These problems have only affected a small portion of our overall network.
One of the ongoing issues has to do with network equipment that is used to aggregate hundreds or thousands of subscribers, known as optical line terminals (OLTs), of which we have hundreds around the country. Following the network migration, these OLTs have started rebooting randomly, making it difficult to identify the root cause. This affects the network nationally and all customers connected to the specific OLTs will be affected while the device reboots.
Just adding the rest of the notice...

We have taken several actions to mitigate the impact, including firmware downgrades, while the underlying problem is being investigated and resolved. We are working closely with Juniper Networks as well as Calix, the manufacturers of the OLT equipment, to capture data on the events in order to identify and resolve the issues.

Separately, there is a problem with the Johannesburg interconnection point for users connected to one of our major ISPs. This has caused connectivity issues for all their customers connecting to the Gauteng region, which includes Limpopo, Mpumalanga, Central (Bloemfontein and Kimberley) and East London. A temporary workaround has been implemented while our equipment vendor and system integrator is identifying and resolving the underlying issue. We are working closely with their engineers to identify and permanently resolve the problems and apologise for this inconvenience.
 
Waiting for MFN 300/300

tenor.gif
 
Just adding the rest of the notice...

We have taken several actions to mitigate the impact, including firmware downgrades, while the underlying problem is being investigated and resolved. We are working closely with Juniper Networks as well as Calix, the manufacturers of the OLT equipment, to capture data on the events in order to identify and resolve the issues.

Separately, there is a problem with the Johannesburg interconnection point for users connected to one of our major ISPs. This has caused connectivity issues for all their customers connecting to the Gauteng region, which includes Limpopo, Mpumalanga, Central (Bloemfontein and Kimberley) and East London. A temporary workaround has been implemented while our equipment vendor and system integrator is identifying and resolving the underlying issue. We are working closely with their engineers to identify and permanently resolve the problems and apologise for this inconvenience.
I actually appreciate the transparency. I wish more companies did this.
 
@AfriNatic - came home now and still experiencing some issues with internet on the new MFN fibre line I have - was wondering if I should log with Afrihost support or if you can assist?

Did some checks when I lose connectivity:
- Provided Huawei router light still green.
- Calix ONT has both broadband and service light on and solid green.

Noticed that some sites are working and others not

Decided to do some tracerts - and it seems like addresses that are resolving to IPv6 addresses are working, but ones which are resolving to IP4 addresses are not:

Screenshot - 25 May 2021 , 6_11_41 PM.png


Screenshot - 25 May 2021 , 6_13_45 PM.png

Then as soon as I click the 'Reconnect' button on the huawei router page and rerun www.pnet.co.za which failed - it works again:
Screenshot - 25 May 2021 , 6_22_08 PM.png

Screenshot - 25 May 2021 , 6_15_16 PM.png

Please advise and let me know if you have any ideas.
 
@AfriNatic - came home now and still experiencing some issues with internet on the new MFN fibre line I have - was wondering if I should log with Afrihost support or if you can assist?

Did some checks when I lose connectivity:
- Provided Huawei router light still green.
- Calix ONT has both broadband and service light on and solid green.

Noticed that some sites are working and others not

Decided to do some tracerts - and it seems like addresses that are resolving to IPv6 addresses are working, but ones which are resolving to IP4 addresses are not:

View attachment 1077221


View attachment 1077225

Then as soon as I click the 'Reconnect' button on the huawei router page and rerun www.pnet.co.za which failed - it works again:
View attachment 1077229

View attachment 1077231

Please advise and let me know if you have any ideas.

Can you switch IPv6 off and see if the issue goes away?
 
Can you switch IPv6 off and see if the issue goes away?
Shot - tried it before on the router but didn't restart - so it wasn't working as expected.

Tried switching it off again now and restarting, and seems to be better. Will monitor and see.

When plugging in the Huawei, straight off the bat it was not configured as per SSID/PW in AH application process. So maybe it got reset somehow and IPv6 was switched off in original config but switched back on when reset. Any other AH specific router configs that might need to be put back into place?
 
Shot - tried it before on the router but didn't restart - so it wasn't working as expected.

Tried switching it off again now and restarting, and seems to be better. Will monitor and see.

When plugging in the Huawei, straight off the bat it was not configured as per SSID/PW in AH application process. So maybe it got reset somehow and IPv6 was switched off in original config but switched back on when reset. Any other AH specific router configs that might need to be put back into place?

Nothing really.

We use DHCP on Metro.

Do you still have IPv6 switched on? I want to see if the issue is with the device not falling back to IPv4 or whether it's an issue with IPv6 on the specific vlan that you are on.
 
Nothing really.

We use DHCP on Metro.

Do you still have IPv6 switched on? I want to see if the issue is with the device not falling back to IPv4 or whether it's an issue with IPv6 on the specific vlan that you are on.

Hey, no I switched it off(with reboot), but that seems to have resolved the inconsistent site issues.

However still noticing a total connection drop where the No Internet access globe icon shows up.

Happened twice since I posted at 9:44PM. Also resolving this by 'reconnecting' WAN.

However I believe you forwarded/logged this for me yesterday and there is an outstanding call with MFN now regarding that, as per AH support. Will update further once I get some clarity on that..
 
Hey, no I switched it off(with reboot), but that seems to have resolved the inconsistent site issues.

However still noticing a total connection drop where the No Internet access globe icon shows up.

Happened twice since I posted at 9:44PM. Also resolving this by 'reconnecting' WAN.

However I believe you forwarded/logged this for me yesterday and there is an outstanding call with MFN now regarding that, as per AH support. Will update further once I get some clarity on that..

I see a lot of /62 IPv6 allocation allocated to you.

The fault is still open with MetroFibre.

I'm thinking it might be router or cable related. Do you have another router to test maybe or able to do testing with PC connected to a laptop or desktop for a while to see if you get any drops.
 
I see a lot of /62 IPv6 allocation allocated to you.

The fault is still open with MetroFibre.

I'm thinking it might be router or cable related. Do you have another router to test maybe or able to do testing with PC connected to a laptop or desktop for a while to see if you get any drops.

Ok - do have a my old DGN2200 - I believe the one ethernet can be used as a WAN port - will test that when I am home.

I did move the router closer to the ONT with the ethernet cable that came with the router. Still had the issue overnight, dropped and had to reboot router :(

When you say cable, are you meaning the fibre cable from ONT or the ETH from ONT to router.

Thanks.
 
Ok - do have a my old DGN2200 - I believe the one ethernet can be used as a WAN port - will test that when I am home.

I did move the router closer to the ONT with the ethernet cable that came with the router. Still had the issue overnight, dropped and had to reboot router :(

When you say cable, are you meaning the fibre cable from ONT or the ETH from ONT to router.

Thanks.

I was suspecting either the Ethernet cable between the ONT and Router or the router itself.

If you test with the DGN and it works we need to log a return request to have the router replaced.
 
I am done with Vumatel's pricing. I cancelled today. The message that came back didn't specify when I will be cancelled. Can you check for me @AfriNatic ? I need to order an alternative service in the mean time (LTE). If I order LTE today, how long does it take to get the sim?
 
Status
Not open for further replies.
Top
Sign up to the MyBroadband newsletter
X