Frequently Asked Questions
What is Sosivio’s core offering?
Sosivio is your “Kubernetes Life Jacket.” Sosivio is the first and only non-intrusive autonomous cloud native observability tool that goes far beyond conventional monitoring. In addition to monitoring your Kubernetes cluster, Sosivio provides root cause determination of failure events, and solutions for those failures. Most importantly, we offer predictive maintenance. By correlating events in real time, Sosivio prevents failures before they ever happen!
What do you mean by “Autonomous Cloud Native”?
Autonomous Cloud Native refers to the ability of a system to predict, detect and resolve critical failures in your cloud native environment. The objective of which is to minimize the need for human intervention and eliminate the need for expert assistance. A truly Autonomous Cloud Native solution requires:
- Eventless Failure Detection: The ability of the system to recognize the potential for failures without intervention and even in disconnected environments
- Automated Failure Resolution: The ability to recommend solutions for failures or imminent failures to restore your system back to optimal performance
- Automated Cluster Optimization: The ability to track node resource utilization & waste, and automatically reallocate resources based on system requirements and subject to customer approval
- Unknown failures resolution: The ability to learn, detect and resolve issues based on behavior rather than a rigid and limited set of rules
What benefits does Sosivio provide over a Kubernetes monitoring tool?
Sosivio provides more than just monitoring and observability. Sosivio is a platform for action. We predict failures before they occur and help you prevent them from ever occurring.
Is Sosivio ISO/SOC 2 compliant?
Sosivio is currently in the process of getting SOC 2 certification. Sosivio is built specifically for security minded organizations with compliance as a high priority. None of your data ever leaves your environment and we do not interfere with any of your applications.
Sosivio Platform: Non-Intrusive & Proactive
When Sosivio says it’s non-intrusive what does that mean?
- No broken artifacts: Sosivio is deployed as a set of independent pods. Your artifacts stay sealed and immutable, which ensures stability of your platform allowing applications to run as intended by developers
- No code injection: Sosivio does not introduce any external code to your environment which ensures full security, stability, and does not annul other vendor warranties
- No agents: Sosivio does not install any permanent agents on your Kubernetes Nodes, eliminating many of the security risks and performance impacts introduced by agent-based solutions
- No sidecars: Sosivio does not deploy sidecar Containers to your applications, which eliminates the performance impact on the network, CPU, and memory from sidecar Containers and keeps Pod behavior as intended by the developers
- No instrumentations: Sosivio does not require any instrumentations, which eliminates vendor lock-in, compatibility issues, and any expertise and excessive development that would be required in house to deploy an observability platform
- No data offloading: Sosivio collects, processes, and analyzes data on-the-fly, which means there is no need to send data external to the Kubernetes cluster, dramatically improving the security and integrity of your cluster and customer data
Why is agentless data collection important?
Agents introduce security risks and performance issues. When agents are used, you are locked into that platform, you incur additional unnecessary consumption of your resources, and you introduce another attack surface to your Kubernetes cluster. Agents may also invalidate other vendor warranties or SLAs you may have. For example, Google Cloud Platform has SLA exclusions for any errors that are a result of customers software.
Does Sosivio require external traffic to be sent outside my Kubernetes cluster?
No, Sosivio collects, processes, and analyzes data in-flight, so there is no need to send data external to the Kubernetes cluster.
Does Sosivio use code injection to gather metrics?
No, Sosivio does not touch your application code, which keeps your artifacts sealed and immutable. Despite this, Sosivio can still optimize application performance and make predictions about application failures while being completely non-intrusive.
Can Sosivio automatically adjust resource limits and requests for my Kubernetes deployments to reduce wasted resources?
Sosivio will recommend specific actions such as reallocating your cluster resources, resource limits, and resource request, to fully optimize your resource utilization. Once determined as a recommended action, the user will have the ability to approve the recommended action prior to Sosivio executing it.
Does Sosivio provide the root cause of an issue when there is a failure event inside my Kubernetes Cluster?
Yes, Sosivio utilizes state of the art, lean AI, built for Kubernetes, to provide instant root cause determination of any failure event and even future unknown failure events as they develop. Sosivio performs a wide variety of proactive checks to augment the existing data - much like a consultant troubleshooting a problem would.
How does Sosivio predict a failure in Kubernetes before it happens?
Sosivio’s AI engine is designed to gather data from your full stack, analyze it, and correlate events to provide an early warning of a future failure event. Failures in Kubernetes are not singular events but rather a sequence of events leading to a failure. In almost all cases, failures are caused by a series of seemingly innocuous Kubernetes events. Sosivio is always watching, identifying these events and keeping an eye out for when a series of events may lead the environment into dangerous territory. Using our AI engines, Sosivio can recognize both previously identified and even novel failures and notify the user before they materialize into a critical failure and offer remediation/mitigation. This is the core of our innovation and what truly sets us apart from other Observability products. For more information on just how we do this, please reach out to us at [email protected] and we’d be happy to show you a demo.
Does Sosivio provide guidance on how to fix a failure once it has detected one?
Yes, Sosivio provides the root cause determination and a fix for the root cause of the event. Sosivio will never fix an issue without the administrator’s approval.
Can Sosivio generate novel problems by fixing the wrong thing?
Sosivio will NEVER take action unless confidence is 99.9% of a sequence. Every action taken is validated prior to applying it either as pre trained data in the model or, in case of an unknown sequence, in a Sosivio sandbox. Only after validation, the action will be taken on the entity of interest.
How do I install Sosivio? What sort of engineering spend, and effort is required on the customer side to set up?
Sosivio has an automated installer, which can be downloaded from the website after you create an account, and that deploys Sosivio for you in less than 5 minutes. There is no engineering effort involved, no pre-requirements, and no special resources needed (like a dedicated cluster).
Want to use your own repository? See the instructions for disconnected environments here.
What platforms is Sosivio compatible with?
See the compatible Kubernetes platforms here.
What is required prior to the installation?
See the prerequisites here.
What if the installation fails?
See troubleshooting procedures here.
How do I put Sosivio on each of my clusters?
You need to run the Sosivio installer once per Kubernetes cluster. Once installed on each cluster, you can link the clusters together using the Add Cluster button on the Multi-Cluster View page. For more details on how to add a cluster to your Multi-Cluster view see here.
Can Sosivio monitor, predict, and resolve on multiple Kubernetes Clusters?
Yes, Sosivio can be installed on as many clusters as you like. In fact, using our Multi-Cluster View you can see an overview of each of your clusters through any individual cluster. This feature is extremely useful for teams running large environments.
NOTE: HTTP/HTTPS communication must be enabled between clusters in order to allow the Multi-Cluster features.
What is the quality of data Sosivio collects?
Sosivio is the first Kubernetes observability tool that was built from the ground up, specifically built to tackle Kubernetes platforms and the applications running on them. Sosivio does not rely on open-source data collectors but rather uses proprietary data collectors to passively gather data from the Kernel, OS, Kubernetes API Infrastructure Layer, Container, and Applications. Sosivio’s data is wider in scope, more granular, and more accurate than any other tool that exists today. The ability to pull this much and this quality of data, in a real-time, lean and non-intrusive way, is what fuels our AI Engines predictive capabilities.
Do I need disk space for storage of Sosivio data?
No. See here for performance requirements.
How much CPU and Memory does Sosivio consume? Will I need to size up my instances on my Kubernetes cluster?
Sosivio uses less than 5% of cluster resources, so in most cases you will not need to size up on your instances.
Do I need a separate host for a monitor server?
No, Sosivio does not require any additional servers for monitoring.
How does Sosivio help interpret the large amounts of data I would normally find in an Observability tool?
Sosivio takes all of the hard work out of the equation by collecting, interpreting, and displaying only actionable insights to you. Using only the Command Center, you can keep your cluster and applications running smoothly without having to inspect monstrous log files.
Do I need a dedicated staff member to manage Sosivio? Do we need a full-time DevOps engineer from our side dedicated to Sosivio?
No, Sosivio does not require any additional staffing. In fact, by using Sosivio, your teams will become more effective and allow them to focus on more important work. Sosivio is a fully autonomous cloud native tool that takes the burden of Kubernetes troubleshooting out of the equation and allows your teams to focus on their core task.
How does Sosivio charge me?
Sosivio offers a number of options to meet your demands. For basic users, enjoy our Community version completely free of charge. For those looking for advanced functionality, we offer flexible licenses. We charge you based on the size of your environment (number of cores, with a minimum of 200 cores) and not based on number of users. We offer 1, 3 and 7 day licenses along with an annual license. For more custom pricing options, just ask! Please see pricing details here.
Does Sosivio offer a free version?
Yes, the Community version is always free and includes the following features:
- Real-Time Metrics
- Multi-Cluster Monitoring
- Cluster Health Checks
- Failure Alerts
- Application Profiling
- Infrastructure Component Health
What are the different options for buying a License?
There are two different types of licenses for the premium version - Pay-as-you-Go and Annual. Pay-as-you-Go are short term licenses that can be purchased directly on the website. You can buy 1 day, 3 day, or 7 day licenses. Annual licenses are discounted based on the size of your environment.
What happens when my 4-week trial ends?
Have no fear. You can still get plenty of value by using our Community version (your version will automatically revert to the free Community version once the trial ends). No need to sign up for an annual license right away...you can continue to use Sosivio and when we predict any problems, use our Pay-as-you-Go licenses to get temporary access to the Premium features.
I am ready to purchase an annual license, who do I contact?
What support do you offer customers?
Customers with a Premium license have support through various channels including slack, email, ticketing, and phone support. Details will be provided when a license is purchased.