On July 20, Hurricane Electric, a network transit provider headquartered in Fremont, CA, experienced an outage that impacted customers and downstream partners across multiple regions, including the U.S., Canada, United Arab Emirates (UAE), Colombia, Germany, South Africa, Brazil, Malaysia, Japan, and the U.K. The outage, first observed at around 2:10 PM EDT, lasting a total of 13 minutes, appeared to center on Hurricane Electric nodes located in San Jose, CA. Around five minutes into the outage, the number of nodes exhibiting outage conditions located in San Jose, CA, appeared to increase. This increase in the exhibiting outage conditions appeared to coincide with the increase in the numbers of partners and regions impacted. The outage was cleared at around 2:25 PM EDT.
On July 20, Microsoft experienced an issue that affected access to Microsoft Teams globally. First observed around 9:15 PM EDT, the outage, lasting around 3 hours, appeared to impact users' ability to access the service. However, network connectivity to the service did not appear to experience any significant issues throughout the outage. Around 11:00 PM EDT Microsoft announced that they had determined that a recent deployment had resulted in a connectivity issue to an internal storage system and began rerouting traffic to an alternate region in an effort to restore functionality to the service. Microsoft Teams availability appeared to be recovered for most global users around 12:15 AM EDT.
Updated July 17
Global outages across all three categories last week decreased from 281 to 272 compared to the week prior, and in the US decreased from 120 to 88, down 27%.
Global ISP network outages decreased from 217 to 210, and in the US they dropped 34% from 97 to 64.
Global cloud-provider network outages decreased from 19 to 9 – a drop of 53% – and in the US they dropped from five to three.
Global collaboration-app network outages doubled from five to 10, while in the US, they jumped from three to eight, a spike of 167% compared to the week prior.
Two notable outages:
On July 14, Arelion (formerly known as Telia Carrier), a global Tier 1 ISP headquartered in Stockholm, Sweden, experienced an outage that impacted customers and downstream partners across multiple countries including, the U.S., the Czech Republic, Hungary, Mexico, Colombia, Brazil, Chile, Peru, Singapore, Canada, France, and Germany. The disruption lasted a total of 23 minutes, divided into two occurrences over a thirty-five-minute period. First observed around 8:45 PM EDT, the first occurrence and the longest, lasting 14 minutes, appeared to center on nodes located in Atlanta, GA. Ten minutes after appearing to recover, the nodes located in Atlanta, GA once again began exhibiting outage conditions. The outage was cleared around 9:20 PM EDT.
On July 14, Twitter experienced a service disruption that impacted users globally. First observed around 8:05 AM EDT, users were unable to access the service although the application remained reachable from a network perspective. Errors seen during the incident were indicative of a back-end application issue. The disruption lasted 40 minutes, with service access restored to a number of users around 8:50 AM EDT. With full resolution confirmed by Twitter around 12:37 PM.
Updated July 11
Global outages across all three categories last week decreased from 283 to 281 compared to the week prior, and in the US decreased from 148 to 120, down 19%.
Global ISP network outages increased from 208 to 217, and in the US they dropped from 109 to 97.
Global cloud-provider network outages increased from 18 to 19, and in the US they dropped from 11 to five.
Global collaboration-app network outages dropped from 12 to five, while in the US, they decreased from nine to three.
Two notable outages:
On July 7, AT&T experienced an outage that impacted AT&T customers and partners in the US, Canada, China, Austria, Spain, Bulgaria, Japan, Australia, South Africa, Ireland, India, and the Netherlands. The 4:20 a.m. ET outage lasted around 19 minutes, and appeared initially to center on AT&T nodes in San Jose, California, and Seattle, Washington. Ten minutes later nodes in Chicago, Illinois, and Ashburn, Virginia, were also affected. The outage was cleared around 4:40 a.m. ET.
On July 5, Cogent Communications experienced a series of outages over a period of two hours and 10 minutes affecting multiple downstream providers in the US, China, Ireland, Hong Kong, the Netherlands, New Zealand, Australia, Singapore, and Japan. The outage, lasting a total of 26 minutes, was first observed around 12:15 a.m. ET centered on Cogent nodes in Oakland and San Francisco, California. After three minutes, the outage cleared, and the Cogent environment was stable for 16 minutes. It then experienced a nine-minute outage affecting nodes in Boston, Massachusetts, and Oakland and Los Angeles, California. An hour and twenty-five minutes after the initial outage, a four-minute outage was observed on Los Angeles nodes in Los Angeles Oakland, and San Francisco, California. Thirty minutes after that appeared to clear, the Los Angeles nodes exhibited outage conditions again. That final outage was cleared around 2:25 a.m. ET.
Updated July 4
Global outages across all three categories last week increased from 247 to 283, up 15%. In the US, they increased from 92 to 148, up 61%.
Globally, ISP outages increased from 173 to 208, up 20%, and in the US they jumped from 63 to 109, up 73%.
Globally, cloud-provider network outages increased from six to 18, and in the US increased from six to 11.
Globally, collaboration-app network outages increased from 10 to 12, and from seven to nine in the US.
Two notable outages:
On June 28, Hurricane Electric experienced an outage that impacted customers and downstream partners across countries including the US, Malaysia, Turkey, Argentina, Germany, Slovenia, Switzerland, Hong Kong, Australia, Japan, and the UK. The outage was first observed around 10:20 p.m. EDT and lasted a total of 24 minutes. It appeared initially centered on Hurricane Electric nodes in Marseille, France, and around five minutes later nodes in New York, New York, and Vienna, Austria showed outage conditions. The number of affected parties appeared to peak 10 minutes after the outage was observed, and it was cleared around 10:45 p.m. EDT. Click here for an interactive view.
On June 30, Cogent Communications experienced a 31-minute outage affecting multiple downstream providers and customers in countries including the US, Brazil, Singapore, Australia, Hong Kong, New Zealand, Mexico, the UK, Germany, France, Spain, India, and Austria. The outage was first observed around 4:45 a.m. EDT centered on Cogent nodes in Los Angeles and San Jose, California; Phoenix, Arizona; and Houston, Texas. Five minutes later, nodes in San Francisco and Oakland, California; Miami, Florida; El Paso, Texas, and Hong Kong also showed outage conditions. The outage was cleared around 5:20 a.m. EDT. Click here for an interactive view.
Updated June 27
Global outages across all three categories last week decreased from 281 to 247, down 12% compared to the week prior. In the US, outages increased from 89 to 92, up 3%.
Globally the number of ISP outages decreased from 215 to 173, down 20% and in the US increased from 62 to 63.
Globally cloud-provider network outages decreased from seven to six, and in the US increased from four to six.
Globally collaboration-app network outages jumped from one to 10 and in the US increased from one to seven.
Two notable outages:
On June 21, Cloudflare suffered an interruption that impacted its customers globally. First observed around 2:30 a.m. EDT, the disruption lasted around 1 hour and 10 minutes, and saw a Cloudflare nodes exhibiting outage conditions in London, England; Vancouver, Canada; Dallas, Texas; Singapore; Tokyo, Japan; Cadiz, Spain; Frankfurt, Germany; and Sydney, Australia. The interruption appeared to prevent some customer traffic from flowing properly to websites and services that rely on Cloudflare. The company announced that the outage was a result of a network-configuration change that prevented traffic from flowing to the Cloudflare infrastructure. Around 3:00 a.m. EDT, Cloudflare announced it had identified the cause and began rolling back the change. Around 3:10 a.m. EDT, connectivity appeared to be restored, and Cloudflare declared the outage cleared around 3:40 a.m. EDT. Click here for an interactive view.
On June 23, UUNET Verizon experienced an outage that impacted customers and partners across countries including the US, UK, Canada, Italy, Japan, the Netherlands, Portugal, China, Germany, and India. The outage was first observed around 12:40 AM EDT and appeared to be centered on Verizon Business nodes located in New York, New York; Newark, New Jersey; San Jose, California; and Seattle, Washington. The outage was divided into three occurrences spanning 4 hours and 45 minutes. Mainly customers and partners in the U.S. were impacted. The total outage lasted around 3 hours and 22 minutes and was cleared around 5:25 a.m. EDT. Click here for an interactive view.
Updated June 20
Global outages across all three categories last week decreased from 309 to 281, down 9% from the week prior. In the US, outages decreased from 129 to 89, down 31%.
Globally, ISP outages decreased from 228 to 215, down 6%, and in the US, they decreased from 99 to 62, down 37%.
Globally, cloud-provider network outages increased from four to seven, while in the US they increased from two to four.
Globally, collaboration-app network outages dropped from 10 to one, and in the US dropped from four to one.
Two notable outages:
On June 16, Cogent Communications, experienced an outage impacting downstream providers as well as customers in countries including, the US, Italy, the UK, Canada, Spain, South Africa, Germany, and Japan. The 19-minute outage was first observed around 8:25 a.m. EDT centered on nodes located in London, England. Ten minutes later nodes in York, England were also affected. The outage was cleared around 8:45 a.m. EDT. Click here for an interactive view.
On June 16, NTT America experienced an outage impacting customers and downstream partners across countries including, the US, Argentina, Uruguay, Brazil, Panama, and Japan. The 28-minute outage was first observed around 3:50 p.m. EDT and appeared to center on nodes in Miami, Florida. The outage was cleared around 4:20 p.m. EDT. Click here for an interactive view.
Updated June 13
Global outages across all three categories last week jumped from 183 to 309, up 69%, and in the US, increased from 85 to 129, up 52%.
Globally, the number of ISP outages jumped from 132 to 228, up 73%, and in the US increased from 59 to 99, up 68%.
Globally, cloud-provider network outages decreased from seven to four, and from three to two in the US.
Globally, collaboration-app network outages doubled from five to 10, and remained the same at four in the US.
Two notable outages:
On June 7, Hurricane Electric experienced an outage affecting customers and downstream partners across regions, including the US, Mexico, Peru, Singapore, China, Canada, Argentina, Costa Rica, and Brazil. The outage, first observed around 5:40 p.m. EDT, was cleared at around 6:10 PM EDT. Click here for an interactive view.
On June 8, Cogent Communications, experienced an outage affecting downstream providers as well as Cogent customers in countries including, the US, Australia, Singapore, Republic of Korea, the UK, Germany, New Zealand, Hong Kong, Japan, Italy, Spain, Israel, Bulgaria, and Canada. First observed around 12:25 a.m. EDT, the outage, lasted 36 minutes in total, distributed across six occurrences over a 3 hour and 45-minute period. The outage was cleared around 4:15 a.m. EDT. Click here for an interactive view.
Updated June 6
Global outages across all three categories last week decreased from 228 to 183, down 20% compared to the week before. In the US, outages decreased from 99 to 85, down 14%.
Globally, ISP outages decreased, from 165 to 132, down 20%, and in the US dropped from 79 to 59, down 25%.
Globally, cloud provider outages increased from five to seven, and in the US remained three.
Globally, collaboration-app network outages increased from two to five, and in the US jumped from one to four.
There were two notable outages during the week.
On May 30, a Hurricane Electric outage affected customers and downstream partners across the US, Spain, Canada, Italy, Japan, Thailand, Australia, Sweden, Costa Rica, the UK, Malaysia, Singapore, Switzerland, Belgium, India, Brazil, New Zealand, and Hong Kong. The outage, first observed at around 7 a.m. EDT, lasted a total of 34 minutes and initially appeared to center on Hurricane Electric nodes located in Marseille, France, and Frankfurt, Germany. Ten minutes into the outage, the Marseille and Frankfurt nodes appeared to recover, and nodes in New York, New York, exhibited outage conditions. After 20 minutes nodes in London, England, Paris, France, and Marseille, also exhibited outage conditions, representing the peak in terms of numbers of partners and customer affected. After about five more minutes, London and Paris nodes appeared to clear. The rest of the outage was cleared around 7:35 a.m. EDT. Click here for an interactive view.
On June 1, Amazon experienced an interruption affecting some of its downstream partners and customers in the US, Australia, India, and Sweden. The 14-minute outage was first observed around 7:20 a.m. EDT and appeared centered on Amazon nodes in Ashburn, Virginia. Five minutes into the outage, some of those nodes appeared to recover, then returned to outage conditions until the outage was cleared around 7:35 a.m. EDT. Click here for an interactive view.
Updated May 30
Global outages across all three categories last week decreased from 277 to 228, down18%, and from 102 to 99 in the US, a 3% decreaser.
Globally ISP outages dropped from 212 to 165, down 22%, and rose slightly in the US from 76 to 79, up 4%.
Globally cloud provider-network outages dropped from 17 to five and from 11 to three in the US.
Globally collaboration-app network outages dropped from four to two, and in the US from four to one.
There were two notable outages during the week.