These are 13 type of panels.
2020 and your suggestion fixed my problem. Spent a whole day trying to figure out what i did wrong. Thanks. You’re the man.
Hi,
Where did all these statistics come from? JMETER? if so, how can I use it to test Influx?
Hi all,
i even face similar issue loading data in dashboard panel, i have almose 42 panel, each fetching by datasource as pg_foriegn data wrapper.
Its highly critical that our business look at grafana dashboard with different set of data. Takes much time than the expected. initially i upgraded the EC2 instance from t3.small to m5.large to improve performance. i still notice the issue persist.
pls help me with your suggestion pls.
I think we have identified a defect in JMeter Load Test in GrafanaLabs (JMeter Load Test | Grafana Labs) - $timeFilter was missed from Test Start/Test End Annotation Queries:
“annotations”: {
“list”: [
{
“datasource”: “${DS_JMETER}”,
“enable”: true,
“iconColor”: “rgb(23, 255, 0)”,
“name”: “Test Start”,
“query”: “SELECT * FROM testStartEnd WHERE type=‘started’”,
“tagsColumn”: “type”,
“titleColumn”: “testName”
},
{
“datasource”: “${DS_JMETER}”,
“enable”: true,
“iconColor”: “rgba(255, 96, 96, 1)”,
“name”: “Test End”,
“query”: “SELECT * FROM testStartEnd WHERE type=‘finished’”,
“tagsColumn”: “type”,
“titleColumn”: “testName”
}
]
},
Which caused 978 data points returning instead of 11 in our case:
SELECT count(*) FROM testStartEnd WHERE type=‘started’
name: testStartEnd
time count_testName
0 978
SELECT count(*) FROM testStartEnd WHERE type=‘started’ AND time >= ‘2020-12-04 22:25:00’ and time < ‘2020-12-04 23:15:00’
name: testStartEnd
time count_testName
1607120700000000000 11
Our Grafana Dashboard got slower and slower due to this defect when more and more test results were saved into influx database.
After adding $timeFilter to Test Start/Test End Annotation Queries:
SELECT * FROM “$database”.“autogen”.testStartEnd WHERE type=‘started’ AND $timeFilter
SELECT * FROM “$database”.“autogen”.testStartEnd WHERE type=‘finished’ AND $timeFilter
Our Grafana Dashboard loading time reduced from 1.5m to 5s.