Migrating Grafana Agent to Grafana Alloy

Authored by Gunasekhar Allukuru

Introduction

In the ever-evolving landscape of observability, tools like Grafana have been pivotal in monitoring and managing the performance of applications. We actually started with LGMT for our observability stack where Grafana Agent served as our base for metric as well as log collection. So that was working pretty well. Although this setup served us well, the Grafana Alloy offered a promising step forward in terms of performance, scalability, and functionality.?

Why We Migrated?

Grafana Alloy is the new name for the Grafana distribution of the OTel collector. Grafana Agent has been deprecated and is in Long-Term Support (LTS) through October 31, 2025. Grafana Agent will reach an End-of-Life (EOL) on November 1, 2025.

The number of agents has grown over the years. Grafana Agent brought in Promtail for Loki logs and the Grafana Agent Operator for scenarios that are Kubernetes-specific. When users are engaged, providing too many options often makes it more difficult to determine which agents and which deployment methods to use. Further, having different binaries means one has to manage and install different services if they have both types of requirements within their architecture as a whole.

Grafana Alloy is an alternative to Grafana Agent. Alloy is the new, open-source distribution of the OpenTelemetry Collector that is fully compatible with OTLP and natively pipelines both OpenTelemetry as well as Prometheus telemetry format, which includes metrics, logs, traces, and profiles. It does everything anyone would expect from the Grafana Agent and other agents besides offering much more capabilities. If you're currently using Grafana Agent within your observability stack, it is suggested that you're moving to Alloy.

The future of observability is here. Experience it with Grafana Alloy

The Challenges We Face During Migration

While the migration to Alloy held much promise, we did face several challenges in the process of migrating:

1. Configuration Complexity:

Our legacy monitoring system was highly customized, in most cases tailored to a custom context. The complexity had to be replicated on Alloy with an understanding of both systems. Optimization of different parameters and thresholds in Alloy to match our desired sensitivity and alerting levels was time-consuming.?

2. Data Integration Challenges:

Our monitoring environment covered a very diverse set of data sources ranging from databases, applications, network devices, to cloud services. This brought along the necessity to pay much attention to the format of data, protocols and mechanisms for authentication at the point of integrating it all together into Alloy. For timely detection of anomalies and incidents, data must be ingested and processed in real-time. So, the need is always to configure pipelines with maximum optimization of data flow while avoiding latency and data loss.

3. Data Quality and Consistency:?

Maintaining data quality and consistency across different data sources became critical. Techniques such as validation, cleansing, and normalization were required to ensure the accuracy and reliability of monitoring.

4. Resource Optimization:

It was a challenging task of how to size and what configuration our Alloy infrastructure should be. Over-provisioning would result in unnecessary costs while under-provisioning would be destructive for performance and scalability. It involves fine-tuning Alloy's configuration and resource allocation in terms of performance. It includes determining bottlenecks for performance, optimizing query execution, and caching mechanisms. A cost-effective new Alloy infrastructure requires managing all costs related to this infrastructure.

Such costs include cost savings, rightsizing of resources, utilization of spot instances, and optimum usage of resources. Migration to Alloy was thus done in a successful manner with further benefits into a stronger, more efficient, and scalable monitoring system by proactively facing these challenges.

Elevate your monitoring game with Grafana Alloy. Experience the difference

Collect

Alloy uses more than 120 components to collect telemetry data from applications, databases, and OpenTelemetry collectors. Alloy supports collection using multiple ecosystems, including OpenTelemetry and Prometheus. Telemetry data can be either pushed to Alloy, or Alloy can pull it from your data sources.

Transform

Alloy processes data and transforms it for sending. You can use transformations to inject extra metadata into telemetry or filter out unwanted data.

Write

Alloy sends data to OpenTelemetry-compatible databases or collectors, the Grafana stack, or Grafana Cloud. Alloy can also write alerting rules in compatible databases.

Grafana Alloy

Grafana Alloy can be installed on Docker, Kubernetes, Linux, macOS, or Windows. We are running LGMT stack in linux. Some Debian-based cloud Virtual Machines don’t have GPG installed by default. To install GPG in Linux Machine, run the following command in a terminal window.

sudo apt install gpg        

Import the GPG key and add the Grafana package repository.

sudo mkdir -p /etc/apt/keyrings/
wget -q -O - https://apt.grafana.com/gpg.key | gpg --dearmor | sudo tee /etc/apt/keyrings/grafana.gpg > /dev/null
echo "deb [signed-by=/etc/apt/keyrings/grafana.gpg] https://apt.grafana.com stable main" | sudo tee /etc/apt/sources.list.d/grafana.list        

Update the repositories.

sudo apt-get update        

Install Alloy.

sudo apt-get install alloy        

First get the existing grafana agentConfigMap by using the following command. I have a namespace called monitoring and isolated all the monitoring related services in the same namespace.

kubectl get configmap -n monitoring        

Output:

Once this is done,use alloy CLI and use alloy convert command to migrate grafana agent configuration to grafana alloy config.

Don’t worry, we’re just getting started—Part 2 is coming soon!

In the next part of this series, we’ll dive into migrating from Grafana Agent to Grafana Alloy. We’ll cover everything that needs to know about the transition—what to expect, the key steps to follow, and how Grafana Alloy can take your monitoring setup to the next level.