We run Elasticsearch in production and learned the hard way that even though it usually does a great job, it can often become very difficult to manage over time. As operations scale, ES can behave in unexpected ways, leading to problems that can be extremely challenging to solve.
Generic monitoring tools that handle various systems that have different architectures just don’t do the trick. Neither does relying on managed services that have significant limitations and disadvantages when running mission-critical Elasticsearch.
We are operation experts who spent years mastering how to keep Elasticsearch running smoothly.