Hi,
I am looking for a Kibana visualization that displays the count of errors that took place in a time range, grouped by an exception and a stack trace.
the relevant mappings:
{
"index_patterns": [ ... ],
"mappings": {
"_source": {
"enabled": true
},
"count": {
"type": "double",
"index": false
},
"stacktrace": {
"type": "keyword",
"index": true
},
"exception": {
"type": "keyword",
"index": true
}
}
}
}
I've decided to use a Data Table visualization and configure it as follows:
- Metrics: sum of count
- Buckets:
split rows: terms aggregation on fieldexception
split rows: terms sub aggregation on fieldstacktrace
This basically accomplished what i wanted, with one exception: when the time range does not
include any errors of particular type (particular exception + stack trace), it is also shown in a table with a sum of count
aggregation equal to 0. Those irrelevant rows make it impractical to find the errors of interest.
Am I approaching this problem in a right way? Is there a better way to do this?
If not, how could I show only the errors that actually occurred during the selected time range?