There were two notable outages during the week.
On Jan. 3, Oracle experienced an outage on its network affecting Oracle Cloud services in the US. The outage was divided into two occurrences over a 25-minute period. The first occurrence was observed around 5:20 a.m. EST centered on Oracle nodes in Austin, Texas, and lasted about a minute. Five minutes after appearing to clear, they began exhibiting outage conditions again that lasted 12 minutes. The outage was cleared around 5:35 a.m. EST.
On Jan. 5, Microsoft experienced an outage on its network that affected access to services running on Microsoft environments. The outage, first observed around 11:20 p.m. EST, lasted nine minutes and appeared to be centered on Microsoft nodes in Chicago, Illinois. Five minutes later, a number of the Chicago nodes appeared to recover, reducing the number of affected partners. The outage was cleared around 11:30 p.m. EST. Given the duration, timing, and location of the nodes the cause is likely to have been a maintenance exercise.
Update Dec. 13
Global outages across all three categories last week increased from 287 to 356, up 24% from the week prior. In the US, outages increased from 103 to 130, up 26%.
Globally, the number of ISP increased from 209 to 261, up 25%, while in the US, they increased from 85 to 108, up 27%.
Cloud-provider network outages worldwide dropped from 28 to 27. In the US, they increased from two to five.
Globally, collaboration-app network outages jumped from five to 13 outages, while in the US they increased from three to seven.
On December 7, AWS experienced an outage disrupting users and customers accessing its services in regions across the globe. The outage was first observed around 10:40 a.m. EST and appeared to be centered on infrastructure in the AWS US-EAST-1 region, located in Northern Virginia. It initially affected services relied upon by non-Amazon apps and services, across regions including in the US, Europe, and APJC. The impact varied depending on the user's IP address. At around 12:37 p.m. EST, Amazon announced it had identified issues related to an application programming interface (API) and were working on recovering services. At 5:43 p.m. EST, Amazon announced it had mitigated the underlying issue, and services began to return to normal. Around 6:03 p.m. EST, most services had been restored, with some disruption still being experienced on the AWS API gateway service. The disruption can be divided into two occurrences, with the first appearing to be the most prominent and lasting around an hour and four minutes; many of the services were restored around 11:44 a.m. EST. The second, lasting around 8 hours, was cleared around 8 p.m. EST.
On December 9, NTT America experienced an outage that affected some customers and downstream partners across the US and Japan. The 54-minute outage was observed around 2:00 a.m. EST and appeared to center on NTT nodes in Ashburn, Virginia. Fifteen minutes into the outage, some of the nodes appeared to clear, leaving just downstream customers and partners located in the US affected. The outage was cleared around 2:55 a.m. EST.
Update Dec. 6
Global outages across all three categories last week decreased from 290 to 287. In the US they increased from 93 to 103.
Globally, ISP outages decreased from 234 to 209, down 11%, while in the US they increased from 79 to 85, up 8%.
Cloud-provider network outages increased from 17 to 28 worldlwide, and in the US they dropped from three to two.
Globally, collaboration-app network outages decreased from seven to five while in the US they remained at three.
There were two notable outages during the week. On November 29, Hurricane Electric experienced an outage that affected customers and downstream partners including in the US, Hong Kong, Republic of Korea, Singapore, Philippines, Malaysia, Switzerland, and Canada. The 12-minute outage was observed around 4:20 p.m. EST centered on Hurricane Electric nodes in Hong Kong. Five minutes later nodes in San Jose, California, also exhibited outage conditions, increasing the number of partners and countries affected. Around 4:30 p.m. EST, the Hong Kong nodes appeared to clear, leaving nodes in San Jose, Minneapolis, Minnesota, and Seattle, Washington, with outage conditions until the outage cleared around 4:35 p.m. EST.
At 5:55 p.m. EST on December 2, GTT Communications experienced an outage centered on GTT nodes in San Jose, California, that affected services in the US and the UK. The outage was cleared around 6:10 PM EST.
Update Nov. 29
Globally, outages across all three categories last week decreased from 388 to 290, down 25% and dropped from 153 to 93 in the US, down 39%.
Globally, the number of ISP outages last week decreased from 287 to 234, down, 18%, and from119 to 79 in the US, a 34% drop.
Worldwide, cloud-provider network outages decreased from 22 to 17, and remaind at three in the US.
Globally, collaboration-app network outages decreased from eight to seven, while in the US they dropped from six to three.
On Nov. 22, Telia Carrier experienced an outage that impacted customers and partners across countries including the US, Panama, Costa Rica, Brazil, Argentina, and Australia. First observed around 11:20 p.m. EST, the outage appeared to center on nodes in Atlanta, Georgia. Nodes in San Francisco, California, exhibited outage conditions 10 minutes and appeared to recover around 11:35 p.m. EST. The Atlanta outage continued and was cleared around 11:50 p.m. EST. Click here for an interactive view of the outage.
On Nov. 23, Cogent Communications, experienced an outage affecting providers and customers in countries, including the US, Spain, New Zealand, Mexico, Canada, and Australia. The outage played out in three occurrences over an hour and five minutes starting around 1:20 a.m. EST centered on Cogent nodes in San Francisco, Oakland, and Sacramento, California, and Washington, DC. The first occurrence lasted four minutes before all nodes appeared to recover. The second occurrence started 25 minutes later around 1:50 a.m. EST and lasted around eight minutes. It appeared to center on nodes in Sacramento. Five minutes into the second occurrence nodes in San Francisco, Oakland, and Salt Lake City, Utah exhibited outage conditions. The third occurrence started 10 minutes after the second appeared to clear and affected nodes in Atlanta, Georgia. The outage lasted a total of 21 minutes and was cleared at around 2:25 a.m. EST. Click here for an interactive view of the outage.
Update Nov. 15
Global outages across all three categories increased from 307 to 360, a 17% increase compared to the week prior. In the U.S., outages increased from 141 to 163, a 16% increase.
Globally ISP outages increased from 221 to 259, up 17%, and in the US they increased from 111 to 135, up 22%.
Globally, cloud-provider network outages increased from 16 to 27, a 69% increase, and in the US, they rose from one to two.
Globally, collaboration-app network outages rose from six to 11, and from six to nine in the US.
On Nov. 9, Comcast Cable Communications experienced two outages that affected downstream partners and customers across the US. The first outage, lasting over an hour, was observed at approximately 12:45 a.m. EST and appeared to center on Comcast nodes in Sunnyvale, California, and primarily affecting customers on the West Coast. After appearing to clear around 1:25 AM EST, they again exhibited outage conditions, before clearing 1:35 a.m. EST. The second outage observed about 8:05 a.m. lasted over an hour and intermittently impacted routes across regions including Chicago, Illinois; Pittsburgh, Pennsylvania; and Ashburn, Virginia. The outage was cleared around 9:20 a.m. EST.
On Nov. 11, NTT America experienced an outage that impacted some of its customers and downstream partners across the US, Brazil, China, Japan, and Canada. Observed around 1 p.m. EST, the 19-minute outage appeared to center on NTT nodes in Osaka, Japan. Five minutes later, nodes in San Jose, California, also began to exhibit outage conditions. Around 1:10 p.m. EST the San Jose nodes appeared to recover, and the outage was cleared around 1:20 p.m. EST
Update Nov. 8
Global outages in all three categories increased from 278 to 307, a 10% increase compared to the week prior. In the US, however, they decreased from 147 to 141, a 4% dip.
Globally, ISP increased from 197 to 221, up 12%, and also increased in the US from 109 to 111, up 2%.
Cloud-provider network outages globally decreased from 22 to 16 outages, and in the US, dropped from five to one.
Globally, collaboration-app network outages, all of them in the US, decreased from eight to six.
Two notable outages:
On Nov. 4, Telia Carrier experienced an outage affecting customers and downstream partners across the US. The disruption lasted a total of 43 minutes, divided into two occurrences over a 55-minute period. First observed around 7:40 a.m. EDT, the first occurrence lasted 39 minutes cetnered on nodes in Phoenix, Arizona, and Los Angeles, California. Five minutes later, the Phoenix nodes appeared to recover, before appearing to exhibit outages again 10 minutes later. The Los Angenes nodes appeared to recover 35 minutes after first being observed, leaving the Phoenix nodes the only ones exhibiting outage conditions. The outage was cleared around 8:35 a.m. EDT.
On Nov. 4, Cogent Communications experienced an outage affecting some downstream providers and customers in the US, Mexico, Hong Kong, Canada, the UK, Germany, and Singapore. The 33-minute outage was observed around 10:15 p.m. EDT centered on nodes in Kansas City, Missouri, and Chicago, Illinois. After 15 minutes, nodes in Cincinnati and Cleveland, Ohio, San Francisco, California, and Denver, Colorado, also showed outage conditions. Five minutes later the Cleveland and Cincinnati nodes appeared to recover, leaving the US as the only affected country. Thirty minutes after the initial occurrance, only the Chicago nodes showed outage conditions. The outage was cleared around 10:50 p.m. EDT.
Update Nov. 1
Outages globally across all three categories decreased from 354 to 278, a 21% drop compared to the week before. In the US they decreased from 154 to 147, down 5%.
ISP outages globally decreased from 254 to 197, down 22%, and from 112 to 109 in the US, down 3%
Both globally (22) and in the US (5), cloud-provider network outages remained the same.
Collaboration app network outages increased from one to eight, all of them in the US.
There were two notable outages during the week.
On Oct. 27, Hurricane Electric experienced an outage that affected customers and partners across countries including the US, the UK, Russia, Brazil, France, Hong Kong, New Zealand, Finland, Japan, Germany, the Netherlands, and Canada. The outage was divided into eight occurrences over two hours and 10 minutes. The first, lasting around 5 minutes, was observed at around 9:29 p.m. EDT, centering on Hurricane Electric nodes located in New York, New York, and Chicago, Illinois. Around 15 minutes after it appeared to clear, the second one started. Five minutes into it, nodes in Ashburn, Virginia exhibited outages. Around 10 p.m. EDT, all the nodes appeared to recover, but 10 minutes later the nodes in Chicago began exhibiting outage conditions again for about two minutes. Fifteen minutes later, nodes in New York and Chicago again exhibited outage conditions. Nodes in those locations repeatedly appeared to recover then exhibited outage conditions for the next five occurrences. The total outage lasted around 40 minutes and was cleared at around 11:40 p.m. EDT
On Oct. 25, Rackspace Technology experienced an outage affecting some customers and partners across countries including the US, France, the UK, Spain, Singapore, Hong Kong, Vietnam, Mexico, Chile, Switzerland, Canada, and the Netherlands. The outage, lasting around 28 minutes, was observed around 10:46 a.m. EDT and appeared to center on Rackspace nodes in Chicago, Illinois. Twenty minutes later, some of the nodes apparently began to recover before clearing at around 11:15 a.m. EDT.
Update Oct. 25
Global outages across all three categories dropped from 387 to 354, a 9% decrease compared to the week prior. In the US, outages decreased from 185 to 154, down 17%.
The number of ISP outages globally dropped from 281 to 254, down 10%, and US outages dropped from 150 to 112, down 25%.
Globally, cloud-provider outages decreased from 30 to 22, down 27%, while in the US, they were cut in half, from 10 to five.
There was just one collaboration-app network outage and that was in the US. The previous week there were two outages globally.
There were two notable outages during the week.
On Oct. 19, Hurricane Electric, experienced an outage affecting customers and downstream partners across the US, Australia, the United Arab Emirates, Thailand, Malaysia, and Canada. The outage was divided into three occurrences over 34 minutes, starting at 1:51 a.m. EDT, centering on Hurricane Electric nodes in Chicago, Illinois. Around five minutes after the first occurrence appeared to clear, outage conditions appeared at nodes located in Los Angeles, California that lasted around eight minutes. The outage was cleared around 2:25 a.m. EDT.