brianr

We've been using this internally for the last couple days. It's strangely awesome to see real stack traces for Javascript errors. Makes it feel just a little more like a "real" programming language. If anyone has any questions, we're here.

This looks really nice, will be spending some time with it. Seems like it could benefit a lot from a rich component library -- is there anything like that on the web somewhere or in the works?

Live-tailing of images looks really, really cool. Do the updates stream in automagically?

Try Rollbar (https://rollbar.com) - we have an official PHP library.

There's also a Symfony2 bundle contributed by the community: https://github.com/staffim/StaffimRollbarBundle

Rollbar (https://rollbar.com) - San Francisco (SOMA)

We're hiring a back-end engineer to be employee #2. You'll own the reliability, performance, and scalability of our systems, spending about half your time on infrastructure automation/maintenance and the other half building and improving back-end systems.

Our tech stack currently includes Python, Node.js, MySQL, Redis, Memcache, and Beanstalkd; plus Nagios for monitoring and Fabric for automation.

One key challenge we're having is ensuring uptime in the face of datacenter-level issues (we've had some) -- our customers rely on us to tell them when they're having problems, so we really need to be rock solid, at least on our API and data processing tiers. So far we're in 6 datacenters with DNS failover and a few other techniques, but there's much much more to do.

If this sounds interesting, email brian@rollbar.com and let's find a time to talk or meet up.

LivingSocial has thousands of employees, many in sales and other non-technical roles. That first sentence reminds them "this affects you, it's not just for the engineers." That seems prudent to me.

It's correct. Business days = weekdays minus holidays, so about 20 per month. So 30 business days ~= 1.5 months.

Rollbar (https://rollbar.com): San Francisco.

Looking for a senior backend engineer. You'll join our three-person team (all of us code) and help shape the future of the product and company.

We're building tools to help developers solve problems faster, and our customers love us. As back-end engineer, you'll own the reliability, scalability, and performance of our systems. Rock-solid reliability is required. Your hats will include devops, DBA, performance optimization, and building new tools and systems.

If this sounds interesting, let's grab coffee or a beer. Email brian@rollbar.com

One trick I didn't mention in the article: partially nullable unique indexes. The idea is you want multiple rows to be able to half-claim a value, but only one to really claim it. The value goes in the non-nullable column, and the second column is a nullable "claimed" column that is set to 1 to claim it. I haven't been able to think of any real use-cases for this though.

Congrats to the team at Exceptional!

If any Airbrake/Exceptional users are looking for an alternative, check out Rollbar (https://rollbar.com). Here's a write-up we did on how we're different from Airbrake: https://rollbar.com/vs/airbrake/

The same issue will likely apply for Facebook apps that run inside of iframes... and I think these days that's all of them.

This looks really nice, great work!

Setup looks a little more daunting than I was hoping though. Could it be possible to make this work without me having to install your gem and run this through Travis?

First one: dropped out of undergrad, if things had totally failed going back (with financial aid) would've been an option.

Current one: ~1 year personal runway saved.

Another category: SDKs, or any code you give your customers to embed within their own. Not only will this make them more comfortable (since they can inspect the code), you're likely to get free, quality work in the form of contributions.

I've been pleasantly surprised at how many people have contributed to the Rollbar (http://rollbar.com) libraries... we've had 10 different contributors to our Ruby gem and a handful to our other libraries. Would highly recommend this strategy to others.

Yes, via email. At least one customer I know of is doing this. A proper integration is in the works.

Co-founder here. We know this is getting to be a crowded space, but we hope you like our take on the problem. Feedback appreciated!

Operational monitoring: Nagios

Error monitoring: Rollbar (http://rollbar.com) - yes, we use it on itself

Performance monitoring: Rollbar, Pingdom, and some custom instrumentation

No firsthand experience, but I've been hearing good things about Pagoda Box (https://pagodabox.com/).

Have you see Koding? http://koding.com . Web-based IDE and terminal access to cloud servers they host for you. I've played around with it a bit and it is really, really cool.

For Rollbar (http://rollbar.com) it's been a fairly random mix: friends, twitter, reddit ads, sponsoring the Node Knockout, the occasional HN comment, and a surprising amount of word of mouth. More recently we've done a bit of PR but it hasn't been targeted at customers.

They look similar design-wise, but both also look similar to Asana's API docs: http://developer.asana.com/documentation/ . It's just a nice way of laying out documentation.

Looking through the PayPal docs quickly, they don't show any particular similarity to Stripe, besides both being REST APIs for a payments platform.

Indeed, maybe it would be better the "smart" or "deliberate" way.

Agree with most of what you're saying. What I was trying to communicate is that trying to win by doing things that seem more difficult than everybody else isn't a strategy, it's a cop-out. It can feel unmatchable, but often it's actually a waste of effort.

There absolutely are times when the work required really is hard. Your internationalization example is a great one -- you had to do 3 months of hard work, but the payoff was huge. And you didn't do this until the lower hanging fruit was already done.

That's a great approach for the cases where the migration takes only a few seconds. When you start running into situations where the migration takes hours or days, it's back to the "normal" way.

Add one for error/exception tracking:

Ratchet.io (http://ratchet.io)

Private beta but public launch is very soon; mention HN for immediate invite.

Good defaults also let you give power users complex features while keeping the common case manageable.

For http://ratchet.io , we're currently working on overhauling our notification system. The new system is rule-based, extremely flexible, and can be as complex to configure as you want it to be. But we're providing sane defaults -- e.g. "email everyone on the project when a new error occurs in production", "if Pivotal Tracker is set up, create a story for each new error" -- so it will (hopefully!) work just fine for most people out of the box.

(self plug) Check out http://ratchet.io - we have libraries for PHP, Django/python, and Node.js, as well as Rails, client-side JS, and Flash.

Nice work triv.io!

Since you asked... I'm working on a product in this space. Check out http://ratchet.io . Collects and aggregates errors from rails, javascript, php, python, etc.

UberX cars still have towncar licenses though, judging by the TCP stickers on their bumpers.