Telkom had a major outage — what happened

Telkom suffered a significant outage in parts of the Western Cape on Thursday due to a cable break that disrupted data traffic.

Outage tracking platform Downdetector showed that Telkom customers began reporting problems shortly after 08:00 on 29 May 2025.

Reports for fibre network operators Openserve and Vumatel also spiked around the same time. However, all indications were that only Telkom Internet service provider customers were affected.

Fibre network operator Frogfoot told one customer from the Western Cape region that only Telkom users were experiencing problems in the area.

A Telkom spokesperson has confirmed to MyBroadband that it suffered an outage due to a cable break incident.

“Our technical teams responded swiftly and successfully restored the link at 10h50,” Telkom said.

“Traffic has normalised across the affected areas, and our engineers continue monitoring the network closely to ensure stable and consistent connectivity.”

Telkom thanked customers for their patience and understanding during the outage.

“We understand how vital reliable Internet access is for our customers, and we remain committed to delivering seamless connectivity, even in the face of unexpected challenges.”

Telkom’s downtime was the latest of several significant Internet outages in South Africa during May.

Internet service provider Cybersmart suffered an extended outage earlier in the month when several of its core routers failed due to a sudden surge in global routing data.

Cybersmart still had a few old Cisco 6500 routers in its network, which the network equipment vendor had stopped manufacturing in 2015.

On 21 May, Vodacom experienced a network outage due to a power fault in a Midrand data centre. The outage prevented customers from making calls and accessing the Internet for nearly two hours.

Details about what caused the power fault have not been released. The issue could have been caused by many things, including a misconfiguration by Vodacom or a problem with the data centre’s redundant power feeds.

NAPAfrica incident

Also on 21 May, two incidents briefly interrupted data traffic for many Internet service providers (ISPs) at South Africa’s largest Internet exchange point, NAPAfrica.

The downtime happened because of a software bug in a piece of networking equipment that prevented it from responding correctly to a flood of traffic from a remote network.

According to feedback from industry sources, the outage impacted most Internet service providers as they are all heavily reliant on NAPAfrica.

NAPAfrica is an initiative of South Africa’s largest data centre operator, Teraco, and operates three Internet exchange points in Johannesburg, Cape Town, and Durban.

It offers free peering services to network service providers, allowing anyone with a presence at one of the data centres to exchange traffic.

NAPAfrica’s business model and technical excellence made it incredibly popular. Internet service providers also credit it with reducing the cost of high-speed broadband in South Africa.

However, this also resulted in many ISPs peering exclusively at NAPAfrica, making them much less robust than those who also peer at INX-ZA.

NAPAfrica said the particular issue that caused the flood of traffic not to be handled correctly was previously logged with its vendor and has been acknowledged as an official bug.

“The recommendation from the vendor was to use port protection mode to prevent port flapping and to ensure that leaks would not affect the network,” it said.

“In this particular case, protection mode has proven to be ineffective, resulting in the same issues related to the aforementioned bug.”

To mitigate this issue, NAPAfrica said it has begun implementing static access control lists to ensure stability.