Best way to monitor endpoints added under custom extension

[AK]

Hi Team,

We have several custom extension in our environment. For which, there are huge number of endpoints added.

Sometimes, few endpoints goes into error state with any reason like authentication error or some other error.

One possible way I think of is to extract a list of endpoints with their state using API, This will give me API URL which I can put into monitoring under HTTP monitor and ingest event (2nd request) to same HTTP monitor to post list of endpoints that are in error. However, I can see, limitations with HTTP monitor like, response body is limited to grab 50kb string. While developing script, code debugging is also difficult because api.info also has limitation of 200 character. Let me know if anyone has achieved this successfully using any synthetic monitor.

Or can you guys think any better way to monitor endpoints? Need your expert advise.

Regards,

AK

[AntonioSousa]

@AK, in our extensions we are "baking" several ways to get around the issues you describe. Some of them include:

• Sending logs to Dynatrace as processing errors are detetected. You can then use Log analysis to see what might be happening.
• Sending events to Dynatrace when something important happens. You can even raise problems in Dynatrace through this method.
• Collecting metrics that show the amount of work being done. I typically setup SLOs for these metrics, so that I can monitor and eventually alert how things are evolving along time.

There are other approaches, and one that I have not tried yet is an extension that seems to upload logs from other extensions directly to Dynatrace. I tried to search it now, but couldn't find it. I'll try to find it and post it back here.

[AK]

Hi @AntonioSousa,

Thank you for the response. It seems, we have several other options to get endpoints monitored.

However, I would like to describe my idea in depth i.e. to use HTTP monitor,

So this HTTP monitor will have two requests,

First will fetch all endpoint details with their individual status and store it.

Second will have ingest API call, which can use list of endpoints in error (captured from previous request) and ultimately trigger an availability event. This will generate problem notification and have those errored endpoint list.

I will definitely explore on the suggestion you gave but just shared my thoughts.

Regards,

AK

[AntonioSousa]

@AK,

Looks good to me. Even if you can put some failure logic in the extensions, and that's of course good, it really doesn't help when it fails by itself. Having another technology (synthetics) doing that monitoring does help in those and other circumstances...

[AK · Accepted solution]

Hi,

I'm able to achieve the desired result via HTTP monitor and it is working awesome.

@AntonioSousa, Thank you for your suggestion and help in this case.

Regards,

AK