Hi. I’m first using tempo and testing java springboot application with mongodb(sharded).
I found something below
if batch size is not set, first ‘find’ bring 101 documents.
then, ‘getMore’ bring 16MB of documents in each times.
In my environment, there is huge size of document in mongodb collection.
So, ‘find’ query one time ‘getMore’ query two times occur. Screenshot is below.
For each span an begin and end is recorded, that is done by the software.
Each span is represented with a name at the left side.
The length of the blue bar is the duration of the span, so how long it took to execute that action.
The place of the blue bar shows where it happened in time.
Some actions do have sub actions, these are shown as child items at the left.
You can have child items which are executed in parallel or in sequence.
In your example:
The first ‘batch-no’ is a parent, it has 1 child ‘batch-no’ DemoApplication_home.
That has 3 child items, which were executed in sequence.
The first is ‘find’ and after that ‘getMore’.
Now to your question, what does that blank section mean:
That is time being spent in ‘batch-no’ DemoApplication_home, the ‘find’ action is done, the ‘getMore’ action is not started yet.
This can have several reasons, but it might be calculations being done (CPU / IO usage for example).
Metrics can help you to find out what is happening with your JVM heap memory for example.
And profiling might help to see what happens in that time.
