Forum Discussion
API - ActivityEvents returns duplicate/excess logs
Hi,
We've started using the Activity Events API to build up a better understanding of usage across our organisation; currently the focus is on report views so specifically filtering down to that.
I've been monitoring the daily load and a couple of outliers have popped up whereby the data tells me a user has accessed a report ~400 times per day. Not only that, the views are all within seconds of each other, and on speaking with the user, they're not aware of any network issues other than the Power BI service feeling 'slow'... I can't see anything on the capacity metrics to suggest they were particularly active during this period, or whether there were any issues.
Has anyone else experienced this before? I've not found anything on the forum or elsewhere online.
Below is an example of the data when I bring it into SQL from the following call: https://api.powerbi.com/v1.0/myorg/admin/activityevents?startDateTime='2022-10-20T08:00:00.000Z'&endDateTime='2022-10-20T08:59:59.999Z'&$filter=Activity eq 'ViewReport'
Have raised with Microsoft and turns out it is a known issue which is being investigated by the dev team; it's in their hands now.
8 Replies
- lbendlinSuper User
Is the report dataset's data source in Direct Query mode? Does the report have an excessive number of visuals on each page?
- BM4291Resolver I
Thanks lbendlin,
I've had a look at the three separate reports that have flagged this issue and each of them use DirectQuery, as for the number of visuals, I'm not sure what would be classed as excessive? One of the reports has what I'd say is quite a lot (20+) whereas another has 5.
The majority of our reports are linked via DirectQuery but have only seen this flag up on two days (since 1st October) and for two users, is there a known issue, or something I should be proactively filtering out for these views?- lbendlinSuper User
In my opinion anything more than 10 visuals on a page is excessive. See if you can combine some of them via small multiples, and drop others that do not provide value.
Keep doing what you are doing - good monitoring is a very important part of keeping your Power BI environment healthy.