Uptimerobot + Opsgenie + StatusPage service and component configuration

[Leigh Elliott]

Hi Everyone,

Does anyone know if there is a how-to guide that explains the end-to-end process to set up new components and services in both Opsgenie and Statuspage that are fed from a Uptimerobot monitoring feed?

I've successfully created the API integrations between Opsgenie and StatusPage with Uptimerobot. However, I'm struggling when configuring individual monitors within Uptimerobot with individual services and components within Opsgenie and StatusPage.

I was expecting there to be a way to do the following:

1. Monitor X in Uptimerobot monitors website www.xyz.com
2. When an alert fires in Uptimerobot for monitor X, an alert is created in Opsmanager automatically linked to service "X."
3. Additionally, an incident is created in StatusPage, and this is also linked automatically to component "X", and its status is automatically updated to "Major Outage" (and its downtime metrics are also automatically updated)
   
   

Currently, I am getting an alert in Opsmanager and an incident in Statuspage, but they are not linked to a service or component (respectively).

How do I configure the incoming alert from Uptimerobot so that service and component assignment is automatic?

Thanks,
Leigh.

[Darryl Lee · Accepted solution]

Hi @Leigh Elliott,

This is Darryl. I am here to help. 😀

Understand that you would like to accomplish the integration flow from UptimeRobot -> Opsgenie -> Statuspage, and the end goal is to automatically update the status of the Statuspage Component.

First of all, there are a few items I would like to clarify:

• Opsmanager --> I assume that you mean Opsgenie.
• API integrations between Opsgenie and StatusPage with Uptimerobot. --> I assume this means you've configured the UptimeRobot integration on Opsgenie & Statuspage integration on Opsgenie.

The way to update the status of Statuspage Components is by the "tags" adding onto the associated Opsgenie Alert, as mentioned in this section.

With that, the key criteria is "Is there any particular information in the data from UptimeRobot to indicate which component is going to be changed to what status?".

Once this is cleared, you can leverage the Dynamic Fields at the Advanced Integration Settings on the UptimeRobot integration to accomplish this.

The concept is:

When an alert is created from UptimeRobot to Opsgenie, auto-add a tag (following the accepted format as instructed here) to this alert, through the well-configured Statuspage integration, there will be an Incident created on Statuspage and auto-update the affected component's status.

---

For further assistance, we recommend reaching out to Opsgenie Support via this link for us to help you better.

Thank you.

Best regards,

Best regards,
Darryl Lee
Support Engineer, Atlassian

[Leigh Elliott]

Hi Darryl,

Thanks for your reply.

Point 1 - Yes, you are correct. I meant Opsgenie, not Opsmanager (whoops).
Point 2 - Also correct. In Opsgenie, I have two integrations setup, one with UptimeRobot and the other with StatusPage

Yesterday I was able to progress this a bit further. Your article references are helpful and confirm that I am on the right track. I've managed to get UptimeRobot monitor alerts captured by Opsgenie and then have the same alert create an incident in StatusPage.

I've also got tags working and can now see the StatusPage incident correctly referencing the relevant component.

One thing I am still trying to work out now is why when I configure an Opsgenie Tag against the Opsgenie Alert of:

cmp_Tutor Warehouse (www.tutorwarehouse.co.nz):major_outage

(Where the component is named: "Tutor Warehouse (www.tutorwarehouse.co.nz)")

The StatusPage incident still shows the component as "Operational" and isn't logging the downtime. Is there anything else I need to add to the Integration rules to record the downtime?

Thanks,
Leigh.
 

[Darryl Lee]

Hi @Leigh Elliott,

Glad to know my suggestion works.

The reason why the tag for this particular component is not working properly is that each Opsgenie Alert Tag can only contain at most 50 characters (reference), and the "cmp_Tutor Warehouse (www.tutorwarehouse.co.nz):major_outage" is actually 59, which eventually gets truncated to be "cmp_Tutor Warehouse (www.tutorwarehouse.co.nz):maj" in my testing.

As a suggested resolution, please consider updating the component name accordingly, thank you.

Best regards,
Darryl Lee
Support Engineer, Atlassian

[Leigh Elliott]

Thanks, Darryl.  That is interesting and good to know.  I have updated the service and component name with fewer characters so that the Opsgenie tag is less than 50 characters, and the auto-updating of the StatusPage component status is now working.

The documentation overall is good, but a real-world example showing a typical setup end to end would be helpful too. 

[Darryl Lee]

Well done, @Leigh Elliott, it's awesome to know my suggestion can help!

We appreciate your feedback on the documentation content.

While I am delivering this feedback to the associated team, may I confirm with you which document you mean that could be more helpful with real-word examples?

On this section, there are actually some examples listing on the page about the tags format.

Thank you.

[Leigh Elliott]

I guess I'm thinking more of a tutorial providing an example of how you might consider implementing monitoring and alerting integration with Opsgenie & StatusPage.  I realise there are many different tools out there, so perhaps use one of the more common or popular tools as an example.  The tutorial could go through the setup process end to end.

Now that I have the integration setup, my next challenge is working out how best to scale.  I may need to set up webhook accounts in UptimeRobot for each monitor I have set up.  Again, I am unsure how best to approach this, as I would prefer not to set up webhook accounts for each monitor in UptimeRobot and Opsgenie.  Any thoughts on this?

[Darryl Lee]

@Leigh Elliott, I will deliver this to our product team for consideration. Perhaps publishing this kind of tutorial in a fresh article would make it more useful.

For the extended inquiry regarding the scaling between UptimeRobot and Opsgenie, we suggest reaching out to Opsgenie Support Team to share with us for the details regarding:

• Your current configuration.
• Your expectations for the workflow going forward.

It would be more effective for us to assist you in this manner.

Thank you.

Best regards,
Darryl Lee
Support Engineer, Atlassian

[Leigh Elliott]

Sorry about all the replies, managed to work it out today.  Didn’t realise I could add multiple alert rules on the one integration, once I did that all good.

Do you know if Monitor URLs that contain “#” are supported?

[Darryl Lee]

@Leigh Elliott, no worries at all.

Theoretically, the URL parsing should be working even with a hashtag. If it's not working, please try to encode the hashtag to be %23 following the ASCII formmating.

Please feel free to reach out to Opsgenie Support Team if none of these works for further assistance. The reason why I keep suggesting this is due to the data privacy and it forms a more effective communication in terms of attachments and tracking and so on.

Hope you can understand this.

Best regards,
Darryl Lee
Support Engineer, Atlassian