We have measures tracking the response times and counts for several of the key programs run by one of our applications. The issue we are seeing is that the measure data itself does not remain current.
At any given point in time we can look in application logs to see how many times a certain action is taken. However, when we look at Dynatrace, there is often no data. If we wait 20-30 minutes, we see that the data finally populates and the charts match what we see in the logs.
It is nice that the data eventually shows up, but we need it to be available in a timely manner in order to be able to establish Incidents off of it or otherwise perform real-time monitoring effectively.
Here is an example. We took a screenshot of a sample chart, but no data shows from the 10:00-10:20 timeframe. Looking at the application logs at this time, however, we know that data is coming through.
If we go back and screenshotting the same chart 20 minutes later, we see that the data we were missing before finally populates. However, there is a good chance that much of the more current data is missing.
Has anybody else experienced anything like this? We are not seeing it for all different measures from what I can tell.
Any help you could give regarding this issue would be greatly appreciated!
- Kasey C.
I appreciate your suggestion! I had taken a look at the Measure Health and Performance Warehouse Overview dashboards in deployment health, but nothing stood out. From what I can tell, we are still well below the base thresholds that Dynatrace recommends:
I have also not seen anything that would lead me to believe that the server itself is backed up. We are still receiving PurePaths and stitching them together at the expected rate. 😃
- Kasey C.
Yes, we have seen similar behavior. There have been numerous problems we have tried to work through with support. Check your Dynatrace Server Health dashboard (from Start Center, Deployment Health) to see if your PurePath Buffer size is increasing, or your RTA Queues are building, Or from the Dynatrace Collector Sizing dashboard, check for Buffer Saturation from any of your collectors. These metrics are also available in the dynatrace Self-Monitoring profile.
In any case you will need to get advice from support as to what needs to be adjusted. We made sure Compression was on for each remote collector. There are a few DEBUG options that have provided some relief. And we are exploring a server upgrade.
Thank you for the response.
Would you happen to have the numbers for the support cases handy so that I can reference them if I go to open a ticket? We are still on 6.2, so the ticket level would automagically go to priority 4.
Having something to reference with potential solutions to the problems may expedite research into our issue.
- Kasey C.