Sentry Reviews & Product Details

Sentry gives me the full picture of what's happening in my app at any given time. The UI makes it easy to navigate — error grouping, stack traces, and breadcrumbs are laid out in a way that actually makes sense. When something breaks, I'm not left guessing: it pinpoints exactly where things went wrong and why, down to the specific line of code.

The integration with Next.js and our CI/CD pipeline was straightforward, and performance monitoring has helped us catch bottlenecks we wouldn't have noticed otherwise. The AI-assisted issue triage is a nice touch too — it helps prioritize what actually needs attention. It's the kind of tool that makes you wonder how you ever shipped software without it. Review collected by and hosted on G2.com.

The initial setup and configuration can take some time to get right, especially when fine-tuning alert thresholds to avoid noise. Review collected by and hosted on G2.com.

Sentry has become the industry standard for error tracking because it shifts the focus from "something is wrong" to "here is exactly how to fix it."

The "magic" of Sentry lies in its ability to provide a timeline of events leading up to a crash. It captures:

Breadcrumbs: A trail of user actions (clicks, navigation, API calls) that occurred right before the error.

Session Replay: A video-like reproduction of the user's screen, allowing you to see exactly what they saw without asking them for a screenshot.

Stack Traces: Direct links to the specific line of code in your GitHub or GitLab repository that caused the failure Review collected by and hosted on G2.com.

Duplicate Issues: Sometimes slight variations in a stack trace (like different line numbers in different versions) cause Sentry to treat the same bug as two separate issues, leading to duplicate notifications. Review collected by and hosted on G2.com.

Clear, actionable error tracking

Instead of dumping raw logs, Sentry groups similar errors, shows stack traces, and highlights the root cause—so you’re not digging through noise.

Great developer experience

The UI is clean and intuitive, and the setup for most frameworks (JavaScript, Python, etc.) is straightforward. You can get useful insights with minimal configuration.

Real-time alerts that actually matter

It notifies you when something breaks, but with smart grouping so you don’t get spammed with duplicate alerts.

Powerful context & debugging tools

You get breadcrumbs, user sessions, request data, and environment info—basically a replay of what led to the error.

Performance monitoring + tracing

Beyond crashes, it helps track slow transactions and bottlenecks, which is huge for improving user experience.

Integrations that fit dev workflows

Works well with tools like GitHub, Slack, Jira, etc., so issues flow naturally into your existing process. Review collected by and hosted on G2.com.

Self-hosting is not trivial

While it’s open-source, running your own instance of Sentry reliably at scale requires significant infrastructure and maintenance effort. Review collected by and hosted on G2.com.

I really enjoy Sentry's ease of integration, i've worked on a lot different codebases, and Sentry is always very easy to integrate regardless of the architecture.

Recently, i integrated Sentry on a pretty distributed app architecture, and we were able to get traces from React, Node and Rust, and using trace continuation across Typescript, TSX and Rust.

I also use Sentry's notifications on Slack, so i can follow specific labels, for errors on the modules that i'm the owner of. Review collected by and hosted on G2.com.

I think this is user error, but it's very easy to get a noisy Sentry if not configured correctly.

When integration you need mainly uncaught exceptions are logged into sentry, and any important errors, such a specific integration thats key to your app.

I try to avoid having captureException calls around, and having a good architecture that makes it so actual exceptions, DB errors, Library Errors, etc, do make it to the top level. Review collected by and hosted on G2.com.

I like that Sentry's UI is fast and presents a lot of the information clearly and efficiently. The filtering is good and allows us to have logs across all three environments with easy switching between them. The screen layout is efficient, which helps individuals, whether they're a developer or engineer, find the information they need to resolve issues. Additionally, I found the setup process to be really good. Review collected by and hosted on G2.com.

I would ideally like the Lambda logging to include individual requests. At the moment, you can only capture top-level boto3 invocations, where we would move all of our current observability over if we could have more granular control over the request response payloads. Review collected by and hosted on G2.com.

I really appreciate how easy Sentry is to set up right out of the box. It's been straightforward from the beginning, reducing friction to get started. I also like how active they are in adding features. We were early adopters of the logging features, which was especially helpful in bridging gaps in understanding weird behaviors on our devices. It helps us gain visibility and get alerted immediately to any erroneous activities, and allows us to track the performance of the devices, which helps us compare and contrast different device models. Review collected by and hosted on G2.com.

I think their tracing feature lacks some functionalities. For example, it was having random issues and unpredictable on how it was working, I couldn't set up sub spans properly. Yes, make it more feature rich, allow us to have sub spans. Make it easier to hook onto span lifecycles. Review collected by and hosted on G2.com.

I like how easy Sentry is to use and all the integrations it offers, especially with Slack and MCP, which really make identifying errors and assigning them to the right people effortless. The out-of-the-box deep integration and instrumentation with many langauges is great because I can easily add it in and it works perfectly (no need to manually instrument). I also appreciate the smooth integration with PagerDuty, as it ensures we get alerted for important issues. Additionally, working with our AI models, the MCP helps us quickly locate and address bugs. The initial setup was really easy too; getting into the code base and logging events was a breeze. Review collected by and hosted on G2.com.

I think the tracing is a little bit hard to use sometimes. It's a little bit difficult to use the trace viewer compared to something like New Relic, where it's easier to see where the bottlenecks are (ie. DB and distributed tracing).in the the LLM instrumentation, there was a little bit of memory leaks that we noticed inside the Sentry library. It was open source, so we fixed that up ourselves. Review collected by and hosted on G2.com.

I really like Sentry's Slack integration because it helps me get real-time notifications of new issues without having to constantly check the Sentry website. Also, the logs are good for tracking deep inside, especially when a bug appears on a client device. They come in very handy for checking API calls or pages visited around the time of the issue. The seamless integration and ease of use were reasons my team switched to Sentry from Datadog. The initial setup was also very easy with the documentation provided. Review collected by and hosted on G2.com.

I have one suggestion in logs, as when the response of any API is long, it gets truncated. There could be an option to allow users to download the whole response without truncation in JSON format for in-depth analysis. Review collected by and hosted on G2.com.

I really appreciate Sentry's super detailed traces and the information about errors, which are extremely helpful. The error grouping in the 'issues' tab makes it easy to see how often an issue is occurring across our stack and provides all the possible information needed to understand what's happening. I also love how it's connected to Claude Code, allowing the agent to pull all the relevant information and look through the logs for me. This connection is beneficial because debugging would take me 10-20x longer without Claude's help, as it efficiently searches through all of our traces and logs to find patterns in usage and errors. Review collected by and hosted on G2.com.

The logs don't have very good advanced filtering on the UI based on attribute, also there isn't a good API connection to pull the logs from projects. I sometimes have to use MCP with Claude, but the data received isn't always well-organized, leading to confusion. Also, the pricing model/rate limiting is an issue. My bills were extremely high unexpectedly, and I couldn't see the logs I needed until the quota reset. More documentation and advice on minimizing unexpected costs would be helpful. The initial setup was pretty complex and overwhelming, taking me 40 hours to really grasp it, and I still feel I'm not using it to its full capacity. Review collected by and hosted on G2.com.

I use Sentry for visibility, traceability, and error detection in my web app. It helps me detect errors before my customers report them, which is useful because there are errors that don't create enough friction for customers to report them. I appreciate that it helps me document my errors thanks to its traces and its connection via MCP, which allows me to analyze bugs with the help of AI agents. I like the replays, as they have helped us gain visibility into errors we didn't know existed and that were never reported by customers. Additionally, I find that the initial setup of Sentry was very easy. Review collected by and hosted on G2.com.

I would like it to have a more user-friendly interface for my customer support staff, as they find it very technical. They have said: 'I can't or don't know how to search for a replay related to a bug reported by a user.' I feel it would be useful for them to group replays by user with their logs and traces, or if this is already done, to make it easier for the user. Review collected by and hosted on G2.com.