Frequently Asked Questions
What is Kausal?
Kausal is a cloud-native observability service for software engineers. By combining monitoring, log aggregation and distributed tracing, we allow users to seamlessly switch between different data sources, saving precious minutes during incident response. Kausal helps you better understand the behaviour of your cloud-native application.
Is Kausal right for me?
Our ideal customer has already adopted Kubernetes for their infrastructure and Prometheus as a monitoring tool and is looking for long term storage, scale up and Prometheus-inspired log aggregation.
How can I sign up?
We are just warming up at Kausal, so before signing you up we’d like to get to know you a little better. This helps you getting the most out of Kausal, and us shape the product more towards your needs. Drop us an email at [email protected] and we’ll take it from there.
What is Prometheus?
Prometheus is a monitoring system containing a sophisticated timeseries database and a powerful query language. Kausal users run a Prometheus instance on their infrastructure to collect metrics and forward them to Kausal.
Can I use my Grafana dashboards with Kausal?
Yes. If you already have dashboards for a local Prometheus datasource, you can simply edit the datasource URL to point to your Kausal account (and add an authentication token).
How is Kausal different from CoreOS Prometheus operator?
CoreOS’s Prometheus operator helps you set up and run Prometheus on Kubernetes. Kausal on the other hand does not run a Prometheus for you, but rather acts as a scalable remote storage and query endpoint. The two play well together: the Prometheus operator can help you run your local agent that forwards metrics to Kausal’s storage. You can then consume the metrics by querying via Kausal’s UI, or by configuring your local Grafana to use Kausal as a datasource.
Does Kausal support a service meshes?
Yes. Kausal uses Prometheus as it log collection agent, and most service meshed expose Prometheus-compatible metrics.
How can I send logs to Kausal?
Promtail is our log collection agent which is designed to be used as a Daemon Set under Kubernetes, running on every host in your infrastructure. The log streams are labelled with metadata from the Prometheus’ service discovery so that they can be queried using PromQL-style selectors from our UI.
How is Kausal’s logging service different from Elasticsearch etc?
Kausal’s log aggregation focuses on a cost-effective, scalable log storage for microservices and cloud native applications. We index log streams using the metadata provided by Prometheus’ service discovery, allow us to easily find relevant logs related to a given Prometheus query. We do not offer full-text indexing of logs, enabling us to store orders-of-magnitude more logs for less cost. We never want you to have to ask your fellow developers to log less.
Does Kausal support other log shippers like fluentd, logstash or oklog?
Whilst we have basic support for fluentd, service discovery driven metadata is a central part to Kausal’s log offering. Users get the most value out of our logging solution when they use Promtail. We’re very much interested in supporting alternative logging agents, or indeed adapting Promtail to send logs to other log aggregation system, so feel free to propose ways to integrate that with your favorite log shipper. PRs are welcome!
How does Kausal’s tracing relate to OpenTracing?
OpenTracing is a standard set of APIs that defines how your service need to be instrumented to expose traces and spans. OpenTracing client libraries implement these interfaces and transport the spans to tracing systems, like Kausal.
When will Kausal’s tracing be available to customers?
We will focus on our tracing service after the log aggregation is out of beta.
Does Kausal’s on-premise version support cloud-based datastores like DynamoDB?
Yes, you can run the system on your machines backed by cloud-based storage. The cloud-based datastores we currently support are Google’s BigTable and AWS’s DynamoDB.
How do I run Kausal in my own datacenter?
We want to add support for storing our data in Apache Cassandra for non-cloud deployments. If you interested in sponsoring this work, let us know at [email protected].
Please contact us if you have more questions.