26 comments

[ 3.1 ms ] story [ 58.5 ms ] thread
Wow, this looks great!

What kind of event will trigger something showing up in the "Recent History" section?

Thanks! The recent history is populated by automated status changes (i.e. outages and degradations) and manually created issues. Status changes are automated via watchdogs, which support a variety of integrations, including our own built-in monitoring solution that provides both the statuses and metrics you see here. For an example, take a look at https://status.sherlock.stanford.edu.
I guess I'm used to nearly a megabyte of javascript coming over the network, but...

Four embedded fonts? Just to get the character ⓘ and a checkmark, both of which are just regular unicode characters that these font files mis-encode?

Does anyone even look at anything any more?

ahh... so thats what all those those empty square boxes are

they don't fail very gracefully

Most of the fonts loaded are variations of Open Sans, which are used throughout the pages. We'll look into improving the degradation when the icon font loads; it's just the Octicon font set. Thanks for pointing this out.
Instead of using Octicon, you could use the Unicode characters for circle-i U+24D8, checkmark U+2713, and calendar U+1F4C5.
In FireFox you can set gfx.downloadable_fonts.enabled to false in about:config and have a much faster web experience.

It's especially super useful on mobile.

That renders this site more broken though, since there's just brown squares all over the page instead of characters from whatever font. I suppose due to no fallback?
doesn't load, not a really great status page then
It appears that a lot of the links aren't real links. I tried to open them in a new tab and it also navigated the current tab.
(comment deleted)
Hi Corben - I like the marketing angle! Just some feedback from a potential customer. I wouldn't use hund.io today because of this: "Our self-healing infrastructure is distributed across multiple AWS regions and availability zones. We run critical services in in."

I want my monitoring service and status page to have 0% overlap with my infrastructure because we all know that cloud providers sometimes have global downtime. It would be awesome if you could let users decide which cloud platform their service was hosted on: Azure, GCP, AWS, DO, etc.

Besides that the product looks great! And there is definitely a need for a combined monitoring / status page service!

On a side note, can you talk a bit about how you've approached the design theming/customization feature?

Thanks! We understand that it's important for monitoring and status page's to be distributed, and separate from your infrastructure. That said, our application is designed to be highly-available at all levels. With our enterprise plan, we are capable of hosting on different infrastructure.

The design customization allows uploading brand images, and gives full CSS customization as well as custom HTML. More about the customization can be found in our knowledge base here: https://hund.io/help/documentation/customization. We also plan on providing support for "page fixtures" to allow rearranging and changing how the information on the page is presented. You can see a little more on them in our roadmap here: https://trello.com/c/HeHVXFkQ/84-page-fixtures

What are you using to generate those svg graphs ?
A very modified dimple.js: http://dimplejs.org/ We plan on making our own library with improved rendering speeds and extended functionality.
How about a Status/Issues page in the style of the Hacker News or Reddit homepage?

Each link is an issue you should fix with your site, or a TODO item you could finish, or a bug report, or a chart of some site data. Automatically generated and refreshed to keep the addictive fresh content factor high. Comments from your coworkers. Trick you into doing work rather than surfing HN.

Pretty nice but the scaling for the "Website TCP Connection Time" graph seems off: average time seem to be in the mSec range but scale max is 100 so you just see a nearly flat line at 0.