Improve logging of active servers on start.

[Harkishen-Singh]

Signed-off-by: Harkishen-Singh harkishensingh@hotmail.com

Description

Fixes: #1182
This PR improves the logging of Promscale on the start.

The logs now will be

level=warn ts=2022-03-03T07:54:25.175Z caller=extension.go:59 msg="could not install promscale: the extension is not available. continuing without extension"
level=info ts=2022-03-03T07:54:25.186Z caller=client.go:146 msg="Prometheus HA is not enabled"
level=info ts=2022-03-03T07:54:25.192Z caller=client.go:116 msg="postgresql://postgres:%2A%2A%2A%2A@localhost:5432/timescale?application_name=promscale%400.11.0-dev.0&connect_timeout=60&sslmode=allow" numCopiers=8 pool_max_conns=80 pool_min_conns=8 statement_cache="512 statements"
level=warn ts=2022-03-03T07:54:25.215Z caller=telemetry.go:56 msg="Promscale collects anonymous usage telemetry data to help the Promscale team better understand and assist users. This can be disabled via the process described at https://docs.timescale.com/timescaledb/latest/how-to-guides/configuration/telemetry/#disabling-telemetry"
level=info ts=2022-03-03T08:03:54.300Z caller=runner.go:129 msg="Started Prometheus HTTP server" listening-port=:9201
level=info ts=2022-03-03T08:03:54.300Z caller=runner.go:207 msg="Started OpenTelemetry OTLP GRPC server" listening-port=:9202

Note: This PR adds run.Group (from github.com/oklog/run) that is used to manage the lifecycle of goroutines, which is a library used in Prometheus. This allows us to stop the execution even on partial error, which earlier was not being done.

TODO

• Update the changelog

Merge requirements

Please take into account the following non-code changes that you may need to make with your PR:

• CHANGELOG entry for user-facing changes
• Updated the relevant documentation