- The reliability of services delivered by Internet Service Providers (ISPs), cloud providers, and conferencing services indicates how well-served businesses are through the internet.
- Thousand Eyes is tracking how these providers are handling the performance challenges they face.
- It will offer Network World a roundup of exciting events of the week in delivering these services.
Global outages across all three categories last week increased from 374 to 381, up 2% compared to the prior week. In the US, outages increased from 94 to 130, up 38%.
ISP outages increased from 293 to 298 worldwide, up 2%, while in the US, they increased from 72 to 107, up 49%.
Globally, cloud-provider network outages reduced from 10 to nine; in the US, they remained the same at four.
Globally collaboration-app network outages decreased from seven to five, and they increased from four to five in the US.
Three Notable Outages
On 25th October, Zscaler experienced an outage that impacted customers using Zscaler Internet Access (ZIA) services on the Zscaler Cloud network 2. First observed around 7:46 a.m. EDT, the outage appeared to affect customers’ network connectivity.
Around 7:46 a.m. EDT, Zscaler announced it had identified the cause of the issue and begun mitigation. It appeared that most customer connectivity had been restored by 11:34 a.m. EDT, with Zscaler announcing the issue was resolved around 4:22 p.m. EDT.
Around 1:30 a.m. EDT on 27th October, Salesforce experienced an outage that affected customers globally that appeared to last about an hour and 24 minutes. It manifested as a series of server errors and timeouts, consistent with a backend service issue.
Around 2:14 a.m. EDT, Salesforce announced it was taking steps to alleviate the issue. Around 2:35 a.m. EDT, services appeared to return, with a significant portion of the issue clearing around 3:15 a.m. EDT. Around 7:23 a.m. EDT, the outage was officially cleared.
On 28th October, Facebook experienced a service disruption that rendered the application inaccessible to some users worldwide. First observed around 3:33 p.m. EDT, the disruption appeared to prevent some users from accessing content and manifested as a combination of HTTP server errors and packet loss at the network edge of Facebook. The incident appeared to clear around 4:45 p.m.EDT.