Generic metrics and context bases monitoring
Project description
jetblack-metrics
Generic metric classes and context based monitoring.
Installation
Install from the pie store.
pip install jetblack-metrics
Usage
First you need to implement a metric which interacts with an actual
instrumentation implementation. The following provides an HTTP request
metric using Prometheus to gather the metrics and the TimedMetric
to
provide a latency metric.
from jetblack_metrics import monitor, TimedMetric
from prometheus_client import Counter, Gauge, Histogram
class HttpRequestMetric(TimedMetric):
"""
A metric which holds HTTP information.
"""
def __init__(self, name: str, method: str, path: str) -> None:
super().__init__()
self.name = name
self.scope = method
self.info = path
self.status = 500
REQUEST_COUNT = Counter(
"http_request_count",
"Number of requests received",
["name", "method", "path", "status"]
)
REQUEST_LATENCY = Histogram(
"http_request_latency",
"Elapsed time per request",
["name", "method", "path"]
)
REQUEST_IN_PROGRESS = Gauge(
"http_requests_in_progress",
"Requests in progress",
["name", "method", "path"]
)
def on_enter(self):
super().on_enter()
self.REQUEST_IN_PROGRESS.labels(
self.name,
self.scope['method'],
self.scope['path']
).inc()
def on_exit(self) -> None:
super().on_exit()
self.REQUEST_COUNT.labels(
self.name,
self.scope['method'],
self.scope['path'],
self.status
).inc()
self.REQUEST_LATENCY.labels(
self.name,
self.scope['method'],
self.scope['path']
).observe(self.elapsed)
self.REQUEST_IN_PROGRESS.labels(
self.name,
self.scope['method'],
self.scope['path']
).dec()
Once we have the metric we can use the monitor
function to manage the process
of gathering the statistics.
def some_http_middleware(request, next_handler):
"""Some kind of HTTP middleware function"""
with monitor(HttpRequestMetric('MyApp', request.method, request.path)) as metric:
# Call the request handler
response = next_handler(request)
metric.status = response.status
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
jetblack-metrics-1.0.0.tar.gz
(6.9 kB
view hashes)
Built Distribution
Close
Hashes for jetblack_metrics-1.0.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 70a7b5e3bb57e1d7e5fb0a97d411cdfc641cc26541dfa4a44e893a227350b93f |
|
MD5 | cb24ee8c95dc7efe0c44d5b3ca3e94f2 |
|
BLAKE2b-256 | 38addb7dc5fdfc678c20dde4fe2cefb9dd0b4c881cc2a500b35b9989b1aece40 |