Why are there no new metrics in Prometheus?

Question

Monitoring

Hello.
developed an application in python - a web server that, when accessed by port and URL, returns metric values ​​​​in this form

Some_Metrics{host="server01", key_="system.cpu.util.usage"} 3
Some_Metrics{host="server01", key_="vfs.fs.size[/data,free]"} 7016763392
Some_Metrics{host="server01", key_="system.uname"} Linux server01 4.1.12-94.3.9.el7uek.x86_64 #2 SMP Fri Jul 24 03:09:40 PDT 2017 x86_64
Some_Metrics{host="server02", key_="system.cpu.util.usage"} 2
Some_Metrics{host="server02", key_="vfs.fs.size[/data,free]"} 6016763392
Some_Metrics{host="server02", key_="system.uname"} Linux server02 4.1.12-94.3.9.el7uek.x86_64 #2 SMP Fri Jul 24 01:09:40 PDT 2017 x86_64

registered in the Prometheus config

job_name: some_metrics 
metrics_path: /metrics 
static_configs:
targets: ['server01:8001'] 
params: 
collect[]:
textfile

restarted Prometheus, judging by the application logs, Prommetheus connected and the process started
Tried to write several queries to try to get the values ​​of new metrics,

http://server01:9090/api/v1/query_range?query={host=%22server01%22}&start=1513629658.0&end=1513633258.0&step=60


http://server01:9090/api/v1/query_range?query=some_metrics&start=1513629658.0&end=1513633258.0&step=60

Nothing in return

{"status":"success","data":{"resultType":"matrix","result":[]}}

What am I doing wrong and how to solve this problem?

Answer 1

In my opinion, the format of the metrics is not respected. Values ​​can only be of real type and certainly not of string type.