I’ve started Grafana as a docker image from grafana/grafana (the latest version I assume) and am accessing it using Chrome and Firefox in Ubuntu 20.
I have a postgres (TimescaleDb) datasource.
I have a minor UI issue with the query builder when creating a dashboard. As I select the table a query request fails with the message “column “time” does not exist”. This makes sense, since my time column is named “timestamp” not “time”. Once I try to select column the first element in the list is called time, followed by the actual columns of my table. If I select “timestamp” and then try to select column again the first element is now “timestamp”, the second element is also “timestamp”, followed by the rest of the actual columns of the table.
This is the response to the query made when I open the column select:
{
“results”: {
“tempvar”: {
“frames”: [{
“schema”: {
“refId”: “tempvar”,
“meta”: {
“executedQueryString”: “SELECT quote_ident(column_name) FROM information_schema.columns WHERE \ntable_schema IN (\n SELECT\n CASE WHEN trim(s[i]) = '”$user"’ THEN user ELSE trim(s[i]) END\n FROM\n generate_series(\n array_lower(string_to_array(current_setting(‘search_path’),’,’),1),\n array_upper(string_to_array(current_setting(‘search_path’),’,’),1)\n ) as i,\n string_to_array(current_setting(‘search_path’),’,’) s\n) AND table_name = ‘c’ AND data_type IN (‘timestamp without time zone’,‘timestamp with time zone’,‘bigint’,‘integer’,‘double precision’,‘real’,‘smallint’,‘numeric’) ORDER BY column_name"
},
“fields”: [{
“name”: “quote_ident”,
“type”: “string”,
“typeInfo”: {
“frame”: “string”,
“nullable”: true
}
}]
},
“data”: {
“values”: [
["“timestamp”", “var_double”, “var_float”, “var_int32”]
]
}
}]
}
}
}
I don’t see any issue here. Maybe a bit strange that the first column is quoted. In the database it appears normal.
Otherwise there are no functional issues.