SaaS Tool assists IT specialists in detecting public cloud downtime

Prior to the manufacturer's discovery, there are tools available for IT specialists to detect cloud downtime. An azure user has detected service disruptions 15 hours before Microsoft announced the outage, and an Amazon flexible Computing Cloud (EC2) user can use Software-based, service-enabled application performance management tools to track Amazon's support for unnoticed problems.

This SaaS tool is a product from the boundary company and requires the installation of an agent on every service instance that does not belong to any public cloud, authorizing access to the operating system, such as Amazon EC2, Rackspace Cloud servers, and Microsoft Windows Azure. Once installed, the agent listens to the network traffic through the virtual network interface and sends information to the Boundary data center, which is processed and presented to the user through a customized web interface.

"Before we installed boundary, we were not aware of the network performance of our solution," Fredrik Lindstrom, a European cloud service provider Qbranch System architect, whose company uses Windows The Azure service bus allows clients to access their data centers through qnet from the service portal. On October 30 this year, Qbranch was able to warn its users about network problems with the service bus before Microsoft announced its downtime.

However, boundary does not accurately identify the cause of the outage, but it shows that many packet transmissions have been disrupted or dropped between qbranch locations and Microsoft Europe's virtual data centers. Eventually, two faulty network switches are considered to be the cause of intermittent network connectivity. First, suppose the error is at the end of the Qbranch, Lindstrom said, but once the boundary report is analyzed, it will clearly show the fault at Microsoft's end.

Okta share boundary information for Amazon support

The identification and access Management service provider Okta is based entirely on Amazon Web Services (AWS), with approximately 200 instances deployed to test, develop, execute, and produce workloads. They ran the boundary agent on 70 instances, one for production. Okta is accustomed to the visibility of the Web, boundary provides communication with AWS support when a problem occurs, because some problems are too small to cause the attention of the cloud services giant.

"Amazon does such a big business and they don't have to pay attention to our size," says Adam D ' Amico, who is Okta's technical operations director, "with some particularly big issues that need to be transferred to them." "In particular, boundary allows Okta to detect partitions between usable areas," they are most likely to believe me if I detect something I can say, ' Look, I can't use this. ' The D ' Amico says, "Look, it's not just one or two machines, but there are a lot of two different areas." ”

Two users say boundary has the potential to be a more effective tool. For example, the automation critical point is the function they want to automatically expand the detection critical point, but also automatically when the flow of abnormal warning users. In addition, should also accumulate some common sense, when normal, when abnormal.

D ' Amico says boundary has the potential to be a security-monitoring vendor, primarily a network performance security issue that warns users when the server is talking to a port that has never been used.

Boundary's main competitors are the application performance monitoring tool providers, such as AppDynamics, CA Inc., Circonus, ExtraHop, Librato, and Newrelic. There are other cloud infrastructure monitoring tools for large data analysis that help users control cloud management through SaaS delivery. including Cloudphysics, Sumo Logic, Splunk, Appfirst and Scalextreme.

The boundary tool is free for sending 2GB detection data to its data center. Qbranch is 400 USD per month for 5GB data.

(Responsible editor: The good of the Legacy)