Forum Discussion

parscot's avatar
parscot
Frequent Visitor
3 years ago

Failed queries and capacity

Our capacity intermittantly spikes with query "failure" as seen in the Fabric Capacity app. How do we troubleshoot this?  We're using a P2 capacity and below the max. The issue doesnt seem to be related to the dataset as the same query could success without issue. 

 

 

 

4 Replies

  • Hey parscot ,

     

    This is a "check the logs" sort of situation.  You need to look at all of the capacity logs and see what is running simultaneously.  You need to look at the machine/service holding the dataset (example, if it is a SQL server, look at those logs, if it is a cloud tool like Dynamics, look at those logs) to see if anything is happening there.  You need to look at your network logs - is there an interesting spike at those moments where something totally not related (or maybe related) to reporting is taking all the bandwidth and causing timeouts?  I see on your screenshot all those background activites - are any of those conflicting in any way (using the same dataset, same server, same whatever)?  Then, you need to look at the Admin logs - is anything happening with O365 admin area that is affecting this?

    I know - there is no one magical bullet answer but this is all about finding the needle in the haystack when something "randomly" happens like this.

  • Hi parscot 

     

    We are experiencing the same issues. Did you manage to resolve this issue?

     

    Thanks