What is Rollbar?
What is Sentry?
Need advice about which tool to choose?Ask the StackShare community!
Sign up to add, upvote and see more prosMake informed product decisions
What are the cons of using Rollbar?
Sign up to get full access to all the companiesMake informed product decisions
Sign up to get full access to all the tool integrationsMake informed product decisions
I had narrowed it down to two tools LogRocket and Sentry (I also tried Bugsnag but it did not make the final two). Before I get into this I want to say that both of these tools are amazing and whichever you choose will suit your needs well.
I firstly decided to go with LogRocket the fact that they had a recorded screen capture of what the user was doing when the bug happened was amazing... I could go back and rewatch what the user did to replicate that error, this was fantastic. It was also very easy to setup and get going. They had options for React and Redux.js so you can track all your Redux.js actions. I had a fairly large Redux.js store, this was ended up being a issue, it killed the processing power on my machine, Chrome ended up using 2-4gb of ram, so I quickly disabled the Redux.js option.
After using LogRocket for a month or so I decided to switch to Sentry. I noticed that Sentry was openSorce and everyone was talking about Sentry so I thought I may as well give it a test drive. Setting it up was so easy, I had everything up and running within seconds. It also gives you the option to wrap an errorBoundry in React so get more specific errors. The simplicity of Sentry was a breath of fresh air, it allowed me find the bug that was shown to the user and fix that very simply. The UI for Sentry is beautiful and just really clean to look at, and their emails are also just perfect.
I have decided to stick with Sentry for the long run, I tested pretty much all the JS error loggers and I find Sentry the best.
Sentry has been very useful for me and my team. I caught a bug in staging — just as one example — which I wouldn't have caught before deploying to production. That's the sort of thing that happens on a regular basis with Sentry.
I didn't personally make the decision to use Sentry since I'm part of a very large organization that chose it before I joined company. But we've been more than happy enough with Sentry that we use it across most of our teams, regardless of their stack.
Team Rollbar--I LOVE you guys and your wonderful service! This review is far too long overdue.
Let me save you a bunch of time and make the decision for you. If you're not already using an error tracking platform, you must. If you're deciding between which services to use, just go with Rollbar, and stop deliberating.
Rollbar is hands-down, THE BEST full-stack application error and exception monitoring/tracking system.
I was an early user and first started using Rollbar in early 2012 (back when it was still called Ratchet.io). Suffice it to say, it has completely transformed and leveled-up the way I build and write applications.
Rollbar is now a must-have for any application I build. TDD? Yeah, could do that, or you can just be more lean and start building, and Rollbar will catch all of your exceptions for you. Large team? Even more so that you need Rollbar, so that you can detect and fix errors before they inconvenience your users.
Here are the reasons for why I think Rollbar is great:
Rollbar has an exceptional, world-class team. Rollbar is built by engineers, for engineers. I know because I've worked with a few folks at team Rollbar, including the founder/CEO Brian Rue, who has been a mentor and advisor to me at a few startups, and some of my elite former colleagues who were hand-picked to join the Rollbar team. They are extremely talented hackers and engineers.
Rollbar scales, an is extremely reliable. We're not just talking Mickey Mouse pretend scale, but they actually do. Guess what? Unlike most of the rest of the web, they're actually NOT primarily an AWS-based stack (yeah, because AWS outages can cause large chunks of the web to fail). Rollbar is in multiple data centers across the world for improved latency, durability/reliability of data.
Rollbar is extremely easy to integrate and very well documented. There are modules/agents for just about every stack and programming language. A basic setup takes 5-10 minutes.
Rollbar is thoughtful. In the early days of Rollbar, data wasn't scrubbed so potentially you could see sensitive information in the notifications you got. Now, they've significantly improved the reporting agents and UI so that sensitive information can be scrubbed before sending notifications, and additional sensitive/PII fields can be configured in the interface.
Rollbar beats the pants off of their competitors. The primary competitors I'm thinking of are Sentry and New Relic. New Relic is more for infrastructure than application, and often, infrastructure smells and problems are code problems. In terms of budgeting or planning IT spend, I would maximize spending on Rollbar and maybe some more basic infrastructure monitoring like hand-rolled Nagios or even Datadog (which also totally rocks, btw) instead of New Relic. As for comparison with Sentry? See next bullet point.
Rollbar is an adult, whereas Sentry** is a kid. Sentry came out of Disqus, and was built by designer-engineers. Don't get me wrong--they have good engineers, but not as good as Rollbar's. I don't care if Sentry is more popular atm or if the UI looks better; I want to know that I can have absolute confidence in my error tracking platform and sleep better at night. If error tracking services were facial hair, Rollbar would be a full, lush beard, and Sentry would be the teenage kid with sporadic prickly hairs here and there and some peach fuzz on the side. Rollbar doesn't rate-limit by default, which means you get all of your exception occurrences notified and tracked. It is 4K Ultra HD, if you will. (Though, to help manage costs and temper noise, you can set custom rate limits per API key--this is so powerful!) Sentry rate-limits by default, resulting in "sampling" error tracking which isn't full coverage and leaves you erroneously thinking that your app is in better health than it actually is.
Rollbar is "multi-tenant" (similar to GitHub) in the sense that you can have one user account affiliated with multiple organizations and projects. This is a nice added convenience.
Rollbar is enterprise-ready and has on-premise deployments.
As of this review, it's 2016--why aren't you using Rollbar yet? If you're still trying to hand-roll your own error logging system, I would seriously question you or your company's technical competence.
Our team has been testing Rollbar as a possible replacement for Honeybadger and has liked it enough to decide to make the switch! Some of the features that stood out to us:
Better UI/UX in general that makes the app feel more comfortable to use
Pricing is more affordable (assuming a reasonable number of events per month)
Powerful notification settings, letting us set up different issue severity and only receive notifications on what matters most (more signal, less noise)
Mute option (another way to cut down on the noise)
More powerful options for auto-resolving and cleaning up old errors
Nice "person tracking" functionality to see who is affected by various errors
Good asynchronous error reporting options in the Ruby adapter
RQL query language for powerful searching (although it's still a little rough around the edges)
We ran into a few minor annoyances:
The gem providing Resque support is developed by a third-party and doesn't seem to get much use (1 watcher and 4 stars as of the time of this review). Honeybadger provides native Resque support in the official Ruby gem.
There doesn't seem to be any search functionality on the documentation page.
No documentation and a slightly unintuitive workflow for one of the integrations we use.
Overall, Rollbar looks like a solid service that was easy to set up, easy to use, and has some powerful features for searching through past error data. We're looking forward to making the switch.
Rollbar became a must for me in post-deployment exception handling and production-related problems debugging.
Here's a list of features that made Rollbar a lifesaver for me:
You can manually report anything you want, this feature helps a lot in debugging production-related bugs. You can also attach a custom payload to be inspected in the dashboard later.
You can use Rollbar-agent internally in your server to prevent Rollbar from blocking your business logic and routines.You can also use the common background job processors like sidekiq, resque ... etc
You have a very rich documentation to read, but actually you'll just need the first 2 paragraphs to go ahead with it!
Wide set of helpers
For me as a rails developer i found a lot of helpers related to rails like: Capistrano deployment hooks, ActiveJob integration, Async error reporting and Rails booting error handling.
You don't need to refresh the page each time to see the new exceptions, everything is real-time !
You can use Rollbar for free to try it without getting engaged in any financial operation, and you can also keep using it if your rate is under 5000 report per month.
I had a very good experience with their support, they respond in a short time and give flexible solutions.
The service has been easy to use and reliable. It's very well documented, and was easy to integrate into our stack. Obviously, you only get out of the error messages what you put in. They will log the person's IP, user agent, etc, but you will need to use descriptive log messages to know what's going on and where.
We are interested in converting some of our APIs over to rollbar as well to try to get a more wholistic view of errors as they happen. This is a very good service and I recommend them.
The guys from Rollbar made very useful error monitoring service.
We have implemented it in our startup weblium.com. This allowed us to see all errors centrally in one place and now our site has become significantly more stable than it was previously.
The main advantages of Rollbar:
1. Reporting all exceptions and errors We just add simple Rollbar code and it catches all unhandled exceptions. Very useful and simple feature, that we use.
2. Custom error reporting. We add some asserts on the most complex part of the code, to catch potential logic errors at the very beginning.
3. Real-time log. When you have to test the production. Instant error logging helps save time.
4. Assign Owner Each error can be assigned to the developer and can be marked as resolved when done. Very useful.
5. RQL Console The new tool, it allows us to create detail data queries. Very powerful tool for the analysis of problem areas in our code.
We have a very good experience with Rollbar. I highly recommend this service.
Lots of companies I respect were using it + open-source + great features and UI. Went for the hosted version, since it plays nice with Heroku. I like how they group together similar errors, give you the ability to mute events or mark them as solved. P.S. check out the Founder Stories feature we did on Sentry if you want to know how they started, its an awesome story.
We've only just added Rollbar but it's shaping up well. Being able to collate all of our errors with stack traces along with the current user and time is going to save more time you could ever imagine as we currently have to troll through docker logs one after each other just to find a single error where as with Rollbar we can easily search and even be alerted when something goes wrong.
The error and event tracking in Sentry is superb. Being able to assign the raw error to people along with all information at the time the event occurred means that we're tracking and fixing problems before they become apparent to customers. We use the self-hosted instance of Sentry.
Sentry is a very powerful error reporting tool. We use it both on front-end and back-end of Ataccama One. It proved to be invaluable in providing insights on our errors - what caused it, what user did before the error occured, stack trace, release tracking and more.
Rollbar handles any unhandled exception, express route error, and any log entry of level warning or error. It notifies our slack ops channel and is integrated with github to allow us to create issues directly from reported errors.
We use Sentry to gather our thrown non-checked exceptions in one place, so we don't have to crawl through all our logs manually. All standalone-applications, our website aswell as our game-servers are linked into sentry.
We use Rollbar for exception tracking. It’s fantastic. I've used other things, but Rollbar is just really, really fast. Their speed at development is amazing. The features, you can tell it’s developers building it.
Sentry is amazing, most of our systems send their exceptions to sentry. We couldn't live without it and get a much better understanding of how our code is behaving in the wild. Plus it integrates with Github.
We are in love with Rollbar. Its deep integration into our Slack Channel keeps us updated at all times. We are able to push bug fixes in less than an hour. The pricing is very suitable for us as a Startup