Generally we have had a really positive experience with Rollbar. It intergrates well with other tools and really helps focus on key issues rather that occasional issues

The deduplication works well for the most part, and the ability to capture lots of additional information (both automatically and manually) is great. Makes identifying what an issue could be really easy.

Doing custom searches look like they could be really useful but are is really slow compared with more general purpose logging tools like ELK/CloudWatch/Splunk and I wish it was possible to expand and change the time periods of the graphs.

We used Rollbar for last few years and many projects and it always work great! Rollbar go fast forward for the last few years with a new features and improvements.

Rollbar is very easy to configure and use in our stack (Ruby + .NET + React). Rollbar gather error information with many details about situations it occurs. Interface is very intuitive and have option to set up many notification destinations (we use Slack + daily reports on mail).

We didn't experience any cons about Rollbar

It's great that it logs errors. And the grouping of errors is sometimes helpful.

That searching for anything other than the unique identifier is almost impossible. Found a work-around using RQL, but that is sometimes hours later by the time the RQL database is populated with recent errors... so basically, we have to reproduce and error and then hurry and click the errors at the top and hope we can maybe find it. It's worse than logging errors to a text file honestly... cause at least I could search that in a reasonable amount of time.

Brilliant, literally set it and capture you errors.

I've used a lot of error tracking tools in the past but I was really pleased with Rollbar. For node applications it's very well laid out, keys are provided for you server and your client setup. Errors are tracked and logged for your application, broken down by which side the error occured. And if an error has been recurring you can track it and log it easily, plus the emails give a good breakdown of the error from the code side.

None at present, I may like the UI navigation to be improved so you can administer the setup quicker but once it's in place it kind of runs itself.

The setup is very simple and the instant notifications for any unhandled issues make this product very attractive to use. It is very developer friendly.

I do not have anything to add which I find less useful.

I have enjoyed how easy Rollbar is to set up with Ruby, Javascript and Ruby on Rails. Their libraries made it painless to set up and already start reporting errors. I don't know that half of the errors caught by Rollbar would have been fixed in the short turn around time as if we didn't have Rollbar. Rollbar is also user friendly and they are constantly putting up more features. My favorite feature is with a given error they include the stack-trace in my code and who wrote the code that is erroring out. This is very useful for debugging.

In short the downtime. Albeit very minimal, there have been some days I can remember that Rollbar was down for more than an hour and if I recall correctly a whole work day. These times it made it hard not to worry if anything is breaking or throwing errors. For this reason I have enjoyed having another performance monitoring product.

Use this tool on one big project for all platforms (backend, web, mobile) and it is good enough to cover your needs in errors or debug info reporting. At the first glance it looks like it lacks features compared with other platforms, but after some time you understand that the product has everything you would need for successful project state monitoring.

Love the minimalistic UI, which keeps everything clear and understandable. Developers constantly add new features, like automatic events grouping, etc. Love to have all platforms on one Dashboard to look at reported error from different platforms point of view.

The quality of Android SDK isn't the top tier with not enough options for event reporting. Also had to so some adjustments for test suits, since in some cases the SDK wouldn't initialize correctly.

The live view and simplistic, but informative, interface

Nothing comes to mind with the web interface, but the mobile app is very limited - I cannot even expand the view of the log

It's an excellent tool for monitoring and tracing back production issues. I would definitely recommend.

I get the notifications that matter to me about broken stuff in production, saving precious time and allowing me to be ahead of users complaints.

Rollbar has much more than what I actually use at this point, but I'd be using even more if I found the setup process to be smoother.

We’ve implemented a database timeout so we receive a rollbar error whenever we have a slow query, and we can track the number of occurrences we have. It was very helpful

I’ve added rollbar to many projects and it was very simple and easy to integrate. Rollbar help and getting started was helpful. The tool is useful to understand what are the weakest parts of the applications and what are the most frequent source or errors.

The error grouping sometimes is not perfect and impacts in how much time I should dedicate to group errors properly in order to identify the real amount of errors and occurrences of the same problem in different parts of the application.

Overall, I like Rollbar and I hope to see some improvements in its functionality in the future.

Rollbar's integration in our architecture is one of the easiest I have seen so far in terms of any small third-party service. A quick setup with token and environment, and you're done. I like how it can map the errors with timelines and usage. The most amazing thing is its seamless integration with Slack, to which we send all of our critical alerts.

I wrote a Go application for which Rollbar recently announced a client. The integration was easy as usual. However, I couldn't write unit tests around it. I wanted to mock the function calls using the widely accepted method of converting the function to a struct and call via an interface. However, neither the root `rollbar` nor the `rollbar.Client` could be restructured in that way. I finally had to remove the token from the testing environment to control the flow of error messages. Secondly, in our organisation people routinely shift from tribe to tribe to work on different features. And in some scenarios, certain individuals require temporary access to Rollbar platform to review the magnitude of error messages. I'd like if Rollbar adds an X days pass like feature, where I don't have to revoke their access manually.

Motivating time saver April 01, 2017 12:38 I'm a Rollbar user since the very beginning of the project. Before Rollbar I've switched a few ways of dealing with logs. First, I was just reading logs when needed with tail and getting a weekly batch of errors by email. That was OK for local development and tiny projects but for bigger deployments that were useless because the amount of data. Then I've found some offline log aggregation tools and these were OK but still required exporting logs, running these tools and checking results. Rollbar automates all that. You're getting a realtime picture of your errors. You can sort messages by the number of occurrences thus getting a good list to work on. Another fun part is how it charges you. The more errors you have the more you pay. Rollbar motivates you to fix things to pay less. At first, I've tried it in personal projects then in a huge commercial project that is already running for more than 7 years. Rollbar saved me hundreds of hours and I'm glad I've stumbled upon it back then.

Saves time. Usability. Features.

One in half a year there's a lag between sending an error and seeing it analyzed. Not a big issue though.

We're Rollbar customers for a about two years, moved to Rollbar after trying few other log aggregation services but none of them had features we needed or didn't deliver on their promise. We're using Ruby on Rails, integration was a breeze, started receiving error messages in less than 5 minutes. The interface is easy to use, live feed works great showing errors as they come up, resolving issues and tracking deploys flows works perfectly. Also I can't recall when Rollbar had downtime or a bug, yet we see new gem releases regularly. I wish other services we use had such a robust infrastructure as Rollbar do. Also I'd like to mention their exceptional support, with very quick responses from highly qualified agents that solve your problems spot on. With Rollbar tools we manage to resolve majority of the issues as they happen, before customers contact our support teams.

Uptime, easy to use api, live feed, rql, people tracking, exceptional support

Unified dashboard between projects would be great, but can be implemented with their api

I love the immediate notifications for errors as users see them. Shooting for "Rollbar Zero" on a team can be the best way to reap value from the product. It gives immense visibility into what's failing.

Why it's failing is not often immediately obvious looking at the rollbar; we have to use multiple observability tools to figure that out. I also am not a huge fan of the UI, it's slow and cluttered and deters me from using it all the time. - Grouping is not great - If Rollbar goes down or stops sending messages, a lot of visibility is taken away -- it's reliability could be a reason why we don't continue using it in the future.

The pricing structure. I have a bunch of small projects with low numbers of users.Per project pricing gets too expensive fast.

Having to fill out multiple free text areas in their marketing survey.

I reported a major bug to Rollbar 2 months ago. Rollbar support acknowledged the bug and said they opened a bug report and escalated it within their development team. I have since reached out asking for an update 4 times (1 month ago, 25 days ago, 15 days, and 1 day ago). I have received no replies at all. I'm shocked that a company would recognize a bug, say they would fix it, and then not reply at all.

I would not recommend rollbar at all given the way they have treated our issues and their lack of response.

They do not take major bugs seriously, and they will not fix them or even reply. Their error reporting does not point to the right file or line. They rely on an orig_trace to be accurate in order to trace issues back to the correct file and line, but that fundamental orig_trace is incorrect, causing all associated error reports to point to the incorrect file and incorrect line. I raised this bug with them 2 months ago.

Anytime a user of our site hits an error Rollbar catches it and tells us, and we have a great amount of information for debugging the error later.

-Has a great integration with our software stack (python + django)so setup is easy and it captures a great amount of debugging information out of the box -Groups errors nicely, but also moves them up when they occur again -Good alerting rules

It can be spammy if you have a lot of errors, it's not really Rollbar's fault and there are many tools to ignore spammy errors.

Rollbar is a must-have for the debugging toolkit. You get errors reported in one place, with history, tracebacks, all request data, user data and so on. Capturing of locals and args is *incredibly* useful and speeds up debugging greatly. You can easily have many people looking at a bug at the same time. The interface is really clear and easy to use, even for the non-dev teammates. You won't be needing much support, but when you do it'll be both very responsive and very effective. Setup is easy and Rollbar never gets in the way of your code. Just add that integration with Slack and you'll find Rollbar indispensable. Massive opportunity for squashing bugs quickly. Definitely do recommend!

- Tracebacks - Locals and args captured - Request and user data - RQL!

- Not everything can be exported as CSV (yet?) - More out-of-the box integrations would be very useful (i.e. Celery)

At DocEngage we have been using Rollbar for more than a year now. We track real-time exception and respond to the issues quickly, sometimes we reach out to our customers before they find time to report the issue. This yields very impactful customer gratification. It also helps us to pinpoint the issue with real use case and evidence. When we have issues with manful traces we observed there is less tension between developers and quality/support team. We tried many competing real-time issue tracker and settled with Rollbar for its simplicity and integration capabilities. We use Rollbar for tracking Java and Javascript exception and our effort to create integration hook was almost zero. As far as collaboration is concerned, we love the way it's integrated with JIRA and HipChat.

Simplicity, Ability to integrate with tools and our application to track real-time issue.

None

Rollbar is super easy to use and a breeze to integrate. It has a ton of features like sourcemaps and fingerprinting that work smoothly for thousands of errors a day.

Whenever there's a timeout on rollbar it shows a message that the site is undergoing maintenance. This happens to me at least once a day which is frustrating. The couple of times using Rollbar where the service is down they haven't updated their status page until hours into the incident.

We have been using Rollbar for last 3 years and we are not planning on quiting! WE LOVE IT! Easy to integrate into our products. Maintaining crash-free app became a lot easier. Support on HIGHEST level. INTEGRATION with JIRA is also a great add-on.

Integration with JIRA could be more advanced. In our company we use kanban boards and it would be a great feature if we could prepare tickets from Rollbar to automatically have every necessary field filled in.

I like that it is easy to sort errors and which environment they're occurring on as well as the types of errors that are happening

Some of the features of the UI don't make a ton of sense to me... Searching for items is really clunky. Sometimes you can search for keywords in the error titles but sometimes you cannot, seems to depend on the word or title.

Rollbar is an absolutely fantastic service and highly-responsive support if you ever hit an issue, all at a very reasonable price point. I can't recommend it highly enough.

- Easy integrations make setup a breeze - Ability to filter errors by environment and assigned level make management great - Deployment tracking, view errors by user, search errors using SQL-like syntax, deobfuscation of stack traces - more features than you can shake a stick at

- Automatic grouping of errors leaves something to be desired, but this is can be worked around by marking these errors manually

Not happy at all. On the verge to move log rockett which likely has similar pricinging and more useful feature.

The real time error loging and mostly easy setup.

UI needs to lot of works to make it user friendly.Telemetry doesn't give a better picture of the error and often confusing. Suspect deploy always defaults latest deployment which make that feature useless.