Running `make gobenchmark` locally creates individual benchmarks files.
This PR adds them to git-ignore list. The combined benchmarks.txt was
already ignored.
<!--Ex. Fixing a bug - Describe the bug and how this fixes the issue.
Ex. Adding a feature - Explain what this achieves.-->
#### Description
Contrib has this folder, and I find it helpful for stashing working
files (e.g. config files) for testing running the Collector from a
binary.
Happy to take another approach if someone is doing something else to
store files used to test the Collector locally.
The repo uses markdown-link-check and not having a package.json file to
pin the version causes dependabot security checks sadness. Following the
same pattern as the specification repo for storing package.json in the
root of the repo.
---------
Signed-off-by: Alex Boten <223565+codeboten@users.noreply.github.com>
This is generated from the opentelemetry-configuration repository using
gojsonschema. Updated makefile to support this w/ `make genjsonschema`
Follow up to
https://github.com/open-telemetry/opentelemetry-collector/pull/7678
---------
Signed-off-by: Alex Boten <aboten@lightstep.com>
This is replicating the setup in the contrib repo that doesn't use
go-acc & gcovmerge for coverage. It uses tools built into go 1.20
instead
---------
Signed-off-by: Alex Boten <aboten@lightstep.com>
We were not able to release yesterday because the release CI job failed.
It failed because two sibling jobs persisted the same file(s) (bin/*) to the workspace.
- This commit fixes the issue by storing the packaged collector in `dist/`
instead of `bin/`.
- In addition to that, the commit also adds job to catch any issues that
might make the publish job fail two weeks down the line. It doesn't add
any significant overhead. It just makes sure the job can run (no
conflicting persisted files in workspace) and that the files expected by
the real publish job are generated.
* initial switch to circleci
* update docs
* test packages in batches of 10.
* update golangci to have go1.14 vet support
* roll back go.mod, move to separate pull.
* allow running longer to avoid circleci timeout.
* further increase timeout.
* further increase timeout.
* fix data race, adjust timeout.
* 10 min timeout (will decrease with parallelism later)
* batch/parallelize all test executions.
* modularize circleci
* add loadtest tools
* optimize running time
* address review feedback
Cijo Thomas
Evan Bradley
Alex Boten
Alex Boten
Dmitrii Anoshin
Sean Marciniak
Bogdan Drutu
Paulo Janotti
Owais Lone
James Bebbington
Liz Fong-Jones
Paulo Janotti
rghetia
Tigran Najaryan
Steve Flanders
Bogdan Drutu
Yang Song
Emmanuel T Odeke