Skip to content

build(deps): Bump github.com/prometheus/common from 0.70.1 to 0.71.0 - #1

Open
dependabot[bot] wants to merge 32 commits into
mainfrom
dependabot/go_modules/github.com/prometheus/common-0.71.0
Open

dependabot[bot] wants to merge 32 commits into
mainfrom
dependabot/go_modules/github.com/prometheus/common-0.71.0

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Sep 13, 2026

Copy link
Copy Markdown

Bumps github.com/prometheus/common from 0.70.1 to 0.71.0.

Release notes

Sourced from github.com/prometheus/common's releases.

v0.71.0

What's Changed

New Contributors

Full Changelog: prometheus/common@v0.70.1...v0.71.0

Commits
  • 9a4aff0 build(deps): bump github.com/stretchr/testify in /assets (#979)
  • 483bb89 Update dependabot config (#978)
  • d29e1ea build(deps): bump golang.org/x/net (#975)
  • e531bd2 build(deps): bump google.golang.org/protobuf from 1.36.11 to 1.36.12 (#976)
  • bc3fc3a build(deps): bump github.com/stretchr/testify from 1.11.1 to 1.12.1 (#977)
  • 15e9f45 expfmt: fix OpenMetrics 2.0 decoder error, format docs, and encoder version d...
  • 0acfdb3 expfmt: prevent st@ leaking to Gauge and Untyped samples in OpenMetrics 2.0 (...
  • eb72e27 model: add Duration unit constants and conversion methods (#952)
  • 67b7d90 fix: drop invalid OpenMetrics 2.0 exemplars instead of failing exposition (#970)
  • 715ac36 expfmt: format OpenMetrics 2.0 float values and validate units (#969)
  • Additional commits viewable in compare view

./main.go breaks as soon as package main gains a second file, and
CGO_ENABLED=1 makes build-arm64 fail on an amd64 host for a binary
that needs no cgo.
http.Client treats a zero timeout as no timeout, so --timeout 0 would
let every scrape hang forever against an unresponsive device.
--version exited 1 with a usage dump because --target was required, so
target is now validated in run instead. A --web.metrics-path (renamed
from --web.telemetry-path) without a leading slash panicked inside
ServeMux.Handle, and is now rejected with an error.

The device collector is registered per scrape, so that a device query
stops as soon as Prometheus gives up on the scrape rather than running
to the device timeout.
signal.NotifyContext keeps its handler installed until stop() runs, so
a SIGINT or SIGTERM sent while a scrape was draining was swallowed and
the process could not be interrupted for up to one device timeout.

Signal handling moves into run so that the shutdown goroutine can call
stop() as soon as the context is cancelled, which restores the default
handling for the next signal.
CI already runs GoReleaser, but the repository had no configuration,
so it ran on defaults and built no image. The Dockerfile takes the
binaries GoReleaser compiled from the build context, as in
kimsufi-notifier-bot.
The binary now lands in build/<os>/<arch>/bin/<name>, the layout
GoReleaser passes as Docker build context, so docker/Dockerfile builds
from the build directory with plain docker as well.

run-docker ran a bot command from another project.
Logging output changes from key=value text to JSON, and the module
gains zerolog plus go-colorable and go-isatty, which zerolog pulls in
for ConsoleWriter.

The collector takes a zerolog.Logger by value rather than a pointer,
following zerolog's own convention for a struct that small.

promhttp.HandlerOpts.ErrorLog needs a Println, and zerolog.Logger
implements io.Writer only at no level, so bridging through the standard
library would drop the level field from every promhttp error. A small
promErrorLog adapter logs them at error level instead.
Logging was hardcoded to zerolog's global logger, so the level could
not be raised to drop the per-scrape errors an unreachable device
produces, and the JSON output stayed unreadable when running the
exporter by hand.

The names follow the existing web.listen-address style rather than
--log-level. Levels go through zerolog.ParseLevel, so trace, fatal and
panic work too, though the usage string lists only the useful four.

main logs the error run returns through the zerolog global, which the
flags would otherwise miss, leaving that one line as JSON under
--log.format console. run now points the global at the configured
logger.
List every metric with its description, and point at the collector and the
NerdQAxe firmware v1.0.37.2-LTS API it is built from. Move the make targets,
linting and release steps out to CONTRIBUTING.md.
Query the device once before listening, so that a wrong target or an
unreachable device is reported right away instead of only through
nerdqaxe_up on every scrape. Include the queried URL in client errors.
Match the Makefile TARGET default, which is the hostname the device
announces over mDNS, instead of a documentation IP.
The logo combines the NerdQaxe wordmark with the Prometheus mark.
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file go Pull requests that update go code labels Sep 13, 2026
@dependabot
dependabot Bot requested a review from TheoBrigitte as a code owner September 13, 2026 12:56
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file go Pull requests that update go code labels Sep 13, 2026
The device names itself on each scrape, so hostname and mac become
variable labels on every metric instead of labels of nerdqaxe_info
alone. Several devices can then be told apart without joining on
nerdqaxe_info.

Descriptors move from package-level vars to a metrics struct built per
collector, and New takes the SystemInfo read at startup, which names the
device of a scrape that fails before any has succeeded.
--target is now repeatable, and NERDQAXE_TARGET reads a comma-separated
list. Every device is queried in parallel on each scrape, and the
hostname and mac labels every metric already carries tell them apart.

Devices share the same descriptors, so they cannot be registered as
separate collectors; collector.Group registers them as one.

A target that cannot be queried at startup still stops the exporter, and
the same device listed twice is rejected rather than emitting duplicate
metrics.
@TheoBrigitte

Copy link
Copy Markdown
Owner

@dependabot rebase

Bumps [github.com/prometheus/common](https://github.com/prometheus/common) from 0.70.1 to 0.71.0.
- [Release notes](https://github.com/prometheus/common/releases)
- [Changelog](https://github.com/prometheus/common/blob/main/CHANGELOG.md)
- [Commits](prometheus/common@v0.70.1...v0.71.0)

---
updated-dependencies:
- dependency-name: github.com/prometheus/common
  dependency-version: 0.71.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot
dependabot Bot force-pushed the dependabot/go_modules/github.com/prometheus/common-0.71.0 branch from 1d43173 to 1ced07e Compare September 13, 2026 21:46
@TheoBrigitte
TheoBrigitte force-pushed the main branch 3 times, most recently from edefd7c to 127b0e8 Compare September 13, 2026 23:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file go Pull requests that update go code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant