Many users have asked for the ability to get weekly and/or monthly uptime ratios to integrate more values into their websites/apps.

Today, we have added a new variable to the getMonitors method which can return the uptime ratio of any given period.

It can provide values back to 2 months (this is how long the logs are kept for).

The new feature is smart enough to return the ratios of multiple intervals in a single API call.

In order to use it, just add: customUptimeRatio=7 to the getMonitors request for getting the 7 days uptime. Or you can go with customUptimeRatio=7-30-60 to get uptime ratios of all 3 days.

To learn more about the usage, just check the getMonitors method in the API docs.

Updates

• alertContactIDs should be sent when creating or editing monitors to define which alert contacts to be notified of up/down events
• getMonitors method can now return the alert contacts of a monitor with adding showMonitorAlertContacts=1 to the querystring

New Methods

There are 3 new methods and the names define their functionality well:

• getAlertContacts

• newAlertContact

• deleteAlertContact


The API documentation is updated with all the new information

We have rolled an exciting update today where you can assign different alert contacts for each monitor.

Let's say:

• you have 10 monitors, 3 of them are your own websites and others belong to customers. And, you want Uptime Robot to notify your customers for "only" the up/down events of their websites,
• you want to notify the web developer when your company website goes down and notify the IT admin when e-mails server don't work.

These are all possible now.

How to use it?

Alert contacts are still defined to your account from the "My Settings" page. After that, they can be selected per monitor from the "add and edit monitor" dialogs.

What about my current alert contact settings?

We have assigned your alert contacts to your monitors accordingly. So, everything will keep working how they were.

If you prefer to change the alert contacts of a monitor, simply click the "edit" icon besides it and you'll see how easy it is.

Did these updates change effect the API?

Yes. There are now new methods and features reflecting the changes. Another blog post is coming about it within a few hours.

Also, several wrappers and code samples are shared which eases consuming the "easy-to-use" API more.

As you know, the standard Uptime Robot apiKey can pull and push data for every monitor in an account. You can create create monitors, edit or delete them. But, what if you need to use the API that pulls data from your account but displays it inside a client's website? Revealing the apiKey would be a security issue.

Monitor-specific apiKeys

A new type of apiKey comes to the rescue: monitor-specific apiKeys are now added to the system which enables pulling data of "only" a given monitor.

Each monitor can have their own apiKey and they are locked to use the getMonitors API method only for that monitor.

That is specially functional for web agencies and freelancers controlling multiple clients websites from their own Uptime Robot accounts.

Improvements

The getMonitors method now returns:

• monitorURL (can't find a good reason why this wasn't in the initial release : ))
• httpusername and httppassword (as monitoring password-protected websites is now possible)

Tighter integration for web apps

If you have a web application/service and looking for a tighter integration with Uptime Robot (like creating users via API), please feel free to contact us at support@uptimerobot.com for more details.

That's all for now : ).

Uptime Robot now has support for monitoring password-protected websites by optionally mentioning the username-password in the monitor's settings.

The feature exists for both HTTP and Keyword monitors and can be found in the "optional settings" link of "Add New and Edit Monitor dialogs".

P.S. The API requests, responses and documentation are updated accordingly.

Uptime Robot already has a wide range of notification methods: e-mail, text message, Twitter DMs, RSS and API.

Today, we are introducing a new notification method: push notifications for İOS -iPhone, iPad And iPod- through the popular and free iOS notification application Boxcar.

Usage is pretty simple:

• Sign-up for a free Boxcar account
  
• Install the free Boxcar app to your iOS device
• Add your Boxcar e-mail as an alert contact in Uptime Robot's "My Settings>Add Alert Contacts>iOS Push" screen
• And that's it.

Also, Boxcar has a Mac Desktop application which can also display the same notifications in your computer.

Does this mean Uptime Robot won't have an iOS app?

No, it doesn't. An iOS app for Uptime Robot is planned however we are currently focused on several other "most requested" features like reporting, website widgets and response time analysis.

Once, such core features are ready, we would love to launch a full-featured iOS app. But thought getting push notifications right now would be awesome.

This is the v1 of the API which covers most of the important stuff including:

• getting the list of monitors, their details (uptime ratio, URL, friendly name, type, etc.)
• getting the notification logs for each monitor
• getting the alert contacts notified after each notification
• adding new monitors
• editing monitors
• deleting monitors.

It is a REST API that can return:

• XML
• JSON
• JSON-p

responses.

In order to use it, simply get your apiKey from the My Settings page and head to the API documentation.

What's next with the API?

In the next version, we'll be adding:

• "monitor-specific apiKeys" so you can securely use the apiKey in a client's website without revealing your main apiKey
• methods for managing alert contacts (adding/editing/deleting them)

Please share your thoughts about the API and/or any other methods that you think should exist in the next version.

The good news : it is fixed now and working as expected.

Sorry for that guys -and, we have experienced that a serious number of things have changed on PHP 5.3 : )-.

Although things work close to perfect, there are 2 downsides:

• all the monitoring nodes are located in US datacenters which makes it possible to get false/positives if the "website being monitored" is not hosted in US and there is a communication problem between country/continent-wide networks (it sometimes happens).
• Uptime Robot's remote monitors run on shared accounts which can not check port monitors that require full system ownership.

In order to perfect the system, we started running 2 new remote monitoring servers: in Europe (Ireland) and Asia (Singapore) that are placed in Amazon's Cloud.

So, with the new setup, every monitor will be controlled from 3 continents (North America, Europe and Asia) before it will be marked as down (including port monitors).

These new remote monitoring locations will be activated by 1 August 2011 to make sure that you whitelist the new IPs below if needed / a firewall is used (if you didn't whitelist any IPs until now and things work ok, there is probably no need to take any actions):

• 46.137.190.132
• 122.248.234.23

P.S. And, we'll be using these new nodes to check and report multi-continent response-times soon .

Happy monitoring everyone!

When a monitor goes "down" first and "up" afterwards, Uptime Robot used to send the info: "The monitor XY (http://www.xy.com/) is back up".

Right now, it includes the amount of downtime so you won't have to compare the date-times of the notifications for finding the value. It is now like: "The monitor XY (http://www.xy.com/) is back up (it was down since 1 days, 2 hours, 07 minutes and 57 seconds)".

Hope you like it : ).