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.

I like the UI and the UX of the Sentry web app, as it's really nice and modern. It's also quite easy to work with the SDK, which makes integrating and using Sentry straightforward. I’ve used their Discord channel a number of times to reach out to the community for help with issues, and I always get a quick response from Sentry employees there. So I definitely recommend Discord for support. Review collected by and hosted on G2.com.

There were some issues with the logging side of Sentry when we first set it up with the OTLP log drain. We raised the problems in their Discord channel, and while Sentry team members said they had fixed them, I believe we're still experiencing some issues when trying to filter using the filter bar for certain properties in the log messages. It doesn't find the property, so you have to manually expand a log message and go through additional steps to add it to the filter. I've thought this issue would have been fixed by now as it's been quite a number of months. Also, the auto refresh on the logs seems quite slow and not really as smooth as I’d expect. It sometimes stops, and I've got to refresh the page myself. I've stopped using it as often because of this and tend to manually click refresh. Review collected by and hosted on G2.com.

What I like most about Sentry is how easy it was to integrate into our frontend and how quickly we were able to start getting meaningful insights. As a frontend engineer, the real-time error tracking has been especially helpful for debugging issues fast, since it provides clear stack traces along with useful user context. I also appreciate how seamlessly it connects with tools like Slack, which makes it easier to stay on top of new errors as they come in. Review collected by and hosted on G2.com.

One thing I dislike about Sentry is the amount of duplicate or noisy events it can generate. As a frontend engineer, this becomes frustrating because it clutters the dashboard and makes it harder to quickly identify real issues.

It also has a direct impact on cost—since Sentry’s pricing is largely based on event volume, even duplicate or low-value errors can consume your quota and increase billing if you go beyond limits . So without careful filtering on the frontend, you can end up paying for noise. Review collected by and hosted on G2.com.

I love the attribute tagging and use Sentry a lot for observability and metrics. I'm able to quickly respond to bugs because of the really helpful trace data. Using attributes to strongly type our logs makes searching through our multi-tenant platform very easy. I also find it great how Sentry integrates with other apps, like Slack. Review collected by and hosted on G2.com.

I really want more visualization tools. I have all these logs but limited ways to show a narrative. From memory, I can only group by a certain amount before the unique groups get clumped together. Because security alert data volume is really important, it means I must leverage different tools to get this data over a time series. Review collected by and hosted on G2.com.

I use Sentry for logging, profiling, and error detection. One major advantage is that it allows us to find and resolve errors much faster, often before customers even notice them. Sentry's logging setup is particularly easy compared to other solutions, which tend to be more difficult and confusing. I like how Sentry's documentation is comprehensive and covers everything I need. This made it easy for me to set up logging and tracing across about 10 micro-services in just two days. The platform itself is intuitive to navigate. I love the unique and fun UI of Sentry. The error tracking is valuable because it provides the complete stack trace, which simplifies problem solving. Additionally, the logging is nearly real-time with only a one-minute delay, which I find impressive. The initial setup was smooth, akin to spreading butter, making the entire adoption seamless. Review collected by and hosted on G2.com.

They have a strict un-changeable 30 day retention period for logs. I would very much like for it to be customizable even if it costs extra. Review collected by and hosted on G2.com.

I really find Sentry helpful for identifying code level bugs through the issues section. It’s great that I can add our app and get warnings and error messages which I can report to the dev team to optimize app performance. The insights and dashboards are also really useful, especially for creating weekly reports. I like that the dashboard provides web page load times with different percentile values, which is important for understanding user impact and improving user experience by reducing loading times. After the initial setup, it became very comfortable to use. Review collected by and hosted on G2.com.

Sentry does not provide insights for video content like video start time and buffer time, making it difficult to analyze video performance. Also, although Sentry tracks page loads, it would be beneficial to have an option to view weekly or monthly uptime for different pages. Review collected by and hosted on G2.com.

I really like Sentry's documentation, it is very good and well indexed and structured, which makes it easy to change my observation strategy with agility. Additionally, I think Sentry's price is very good and competitive compared to other more expensive tools. The integration with our application framework was quite simple, as the perfect use case was documented and well updated. Sentry also integrates well with other tools we use, such as our Cloud Code and Slack, for issue resolution and receiving alerts. Review collected by and hosted on G2.com.

Yes, what I have already mentioned. I think that traces, at least in NodeJS and TypeScript, sometimes you lose the trace due to the asynchronous nature of Node, which results in orphan traces. I remember this didn't happen to me with Dynam Trace, as it was injected at the container level, while Sentry seems to rely on Open Telemetry, which has its own small inconveniences in TypeScript, which are a bit cumbersome. Review collected by and hosted on G2.com.