HN Explorer

Ask HN: Review my winter break project (statusdot - hosted website status pages)

38 points by samdk ↗ HN
Hi HN!

Eli Fox-Epstein (elitheeli here) and I decided that we should build something this week, so we built Statusdot (http://statusdot.com). It's a service to provide status pages for websites (similar to status.github.com or status.37signals.com).

https://statusdot.com

It's pretty basic right now: You set up a CNAME pointing at statusdot.us for (for your 'status' subdomain), so you can access the site by going to status.yourdomain.com. You give us a URL, we do a GET request every 5, 10, 15, 30, or 90 minutes. If that request doesn't come back with a 200 response or doesn't contain a string you specify, we consider it down. We display that information on a status page, along with your twitter account and status messages you can post. (We also send you an SMS message when your site goes down.)

We've got a few more features and a ton more polish left to do, but we wanted to soft-launch to get some feedback, so right now it's free for a couple of weeks. (After that we're planning to start charging monthly.)

We'd love any feedback you can give. Would you use it? If not, why not? Is there anything we could add that would make it more useful for you?

Thanks!

21 comments

[ 4.1 ms ] story [ 52.0 ms ] thread
pvsnp
I just put my site here, it looks like a great start. Getting some more data in future would be interesting, like how much time it's taking to load the pages and show that on a graph.
jackowayed
Looks good. A few places you could go with it:

* It looks like you currently allow custom messages and overriding of the up/down-ness, but only through SMS. A web interface for that would be nice.

* An API for that would be nice too. Eg. GitHub's monitoring of the uptime of the git interface could make an API call to tell you when it goes down and comes back up.

Why do you have us CNAME to statusdot.us instead of to statusdot.com?

I'm also wondering what the best way to handle plans is. I'm using it to monitor my personal site and am highly unlikely to ever pay you for that. But letting me use it for free would make you likely to get some money out of me when I launch a real site.

Some ideas:

* To use SMS, it costs money

* To have multiple sites, it costs money.

* To use the API, it costs money (?)

* To CNAME, it costs money (?)

But it seems like you should offer some basic version for free.

samdk
Yep, the custom messages/status overriding will be accessible through the web interface soon, and an API is definitely in the pipeline. We wanted to get something out the door quickly though.
aik
The landing page seems overly dark and techno. A professional business looking for a client-facing status page you would expect to be lighter and friendlier looking, and the front-page doesn't agree with this. I was a little turned off by it, but the example was great.
samdk
Yeah, the homepage definitely needs to be redone. I did that, did the rest of the site (in a completely different style), and haven't gotten a chance to take a look back at the homepage yet. Thanks for the feedback.

(I'd also appreciate others' opinions on this, although I do agree.)

llgrrl
I love your service. Some thing that I'm thinking about now:

- Ability to customize the status page (theming) - Embeddable image (just in case someone wants to shove a lot of status for different pages into one page) - API? This, you can charge money for.

samdk
Custom CSS/branding and an API are the next two big things we're planning on getting working, so that's good to hear. :) thanks!
jessor
Great idea!

In the admin, when one adds his site and there's an error, all fields clear out. Also i tried putting in a german mobile number in various formats and it always said "Sms number must be at most 10 characters long".

Best of luck, looking forward to where you'll go with this.

elitheeli
Thanks, we'll try to get these fixed ASAP.
fookyong
If you're just doing a GET request then this doesn't solve the problem.

Correct me if I'm wrong but you're trying to solve the problem of reliably displaying uptime status to answer the "huh it's down for me, is it down for you?" question that users sometimes have.

Just sending a GET doesn't really solve this, as it will only give you the status of the server from the point of view of that node.

Anyway, it depends what your intended use-case is.

If it's for publicly displaying uptime and therefore taking the reputation of the client company into your own hands, I would make sure that the method of reporting is as accurate as possible. I don't even think multiple nodes is the answer. You basically have to have a daemon installed on the client server.

samdk
Yep, you're right that a single GET request has some problems. We don't right now (because we wanted to get feedback ASAP), but we will definitely be checking from multiple nodes (running on multiple different providers) by the time we're charging people money so that we're as sure as we can be that the problem isn't us.

Getting an API together (to enable a server daemon) is also very high on our priority list. (Although the disadvantage with that is that then you're not actually checking to see if the site's up: you're checking whether the server's alive, and that doesn't necessarily mean the site is up.) We do recognize we're dealing with something tricky and we want to get it right. Thanks for the feedback.

fookyong
no probs.

I guess to make my point more relevant in a business sense, I recognise the problem being solved here. In fact it's not just users who have the "is it down?" question, it's non-technical coworkers in a large organisation too. I'm the web director at a medium-sized company and I've had my fair share of "is it down?" phone calls at inconvenient times in the night :)

However, for me to put my faith in a 3rd party service where I could refer coworkers or users to confirm the uptime instead of contacting me directly, and - more importantly - for my to get my wallet out and pay for it, I would need to know first that the data is as accurate as it can be.

Perhaps that's your freemium model right there:

Free - a simple uptime page

Paid - a more detailed uptime page with multiple nodes, OS service status via API, latency from various locations, response times over the last hour, etc.

jeffmould
Looks nice, but what is the selling point here? How do you differentiate yourself from a service like Pingdom? I would put some selling points on the homepage.
ernestipark
Looks pretty good so far. I like that I got up and running real quick and it's simple. Two quick suggestions (although some of these might be things you already thought of but just haven't implemented yet: (1) error messages for form mis-entry, and (2) in the change password form, I'd move the old password field to be first. That's how I've seen it everywhere and it feels unintuitive to type the new password first. I messed up the first time I tried to change my password because of it. Best of luck!
obilgic
Why do not you just integrate micro-blogging system to this ?
obilgic
Also customizable themes/colors. Maybe add some space for website logo?
dholowiski
It looks great. The only thing I would be concerned with is that it would be pretty easy for a developer to replicate on their own.

Being totally honest, the price would have to be pretty low, or offer X sites (like 10) for the price for me not to just do it myself.

That or there would have to be some kick-ass value-add. The simplicity is a nice value-add, but doesn't quite cut it for me.

Mithrandir
Will there be a free version at the end of the beta?
kevinburke
Pretty awesome - puts my winter project(s) to shame :-)