[Celerdata Integration]Some metrics were not reported to Datadog

[yandongxiao]

Background

On Datadog's dashboard, the celerdata.fe.query.timeout.count metric cannot be found. Additionally, the celerdata.fe.table_num metric only has the metric for the information_schema table, i.e., starrocks_fe_table_num{db_name="information_schema"}.

First, from the code perspective, these two metrics exist.
celerdata.fe.query.timeout.count is here

celerdata.fe.query.timeout.count,count,,,,Total number of timeout queries,0,celerdata,,

celerdata.fe.table_num is here

celerdata.fe.table_num,gauge,,,,Counter of tables,0,celerdata,,

In the metrics.py, the related content is:

...
"starrocks_fe_table_num": "fe.table_num",
"starrocks_fe_query_timeout": "fe.query.timeout",
...

Secondly, from the perspective of FE Node, these metrics exist.

# HELP starrocks_fe_query_timeout total timeout query
# TYPE starrocks_fe_query_timeout counter
starrocks_fe_query_timeout 0


# HELP starrocks_fe_table_num count of table
# TYPE starrocks_fe_table_num gauge
starrocks_fe_table_num{db_name="information_schema"} 54
starrocks_fe_table_num{db_name="_statistics_"} 10
starrocks_fe_table_num{db_name="custom_reporting"} 19
starrocks_fe_table_num{db_name="sys"} 8

Additionally, other metrics can be queried, e.g. celerdata.fe.thread_pool and celerdata.fe.query.latency_ms.count

celerdata.fe.thread_pool,gauge,,,,Statistics of the thread pool,0,celerdata,,
celerdata.fe.query.latency_ms.count,count,,,,,0,celerdata,,

In the metrics.py, the related content is:

"starrocks_fe_thread_pool": {
    "name": "fe.thread_pool",
    "type": "gauge",
},

"starrocks_fe_query_latency_ms": "fe.query.latency_ms",
"starrocks_fe_query_latency_ms_count": "fe.query.latency_ms.count",
"starrocks_fe_query_latency_ms_sum": "fe.query.latency_ms.sum",

Are there some misconfigurations in celerdata integration

Expected behavior

All metrics should be reported to Datadog, and can be searched.

Version

Datadog agent 7.71.2
CelerData Integration 1.2.1