Skip to content

bcdock v0.5.0

Choose a tag to compare

@github-actions github-actions released this 30 Sep 11:55
· 1 commit to main since this release

bcdock v0.5.0

Beta CLI. Surface and JSON shapes may change between releases - see CHANGELOG.md for what moved.

Install:

go install github.com/bcdock/cli/cmd/bcdock@v0.5.0

Or download an archive below and put the bcdock binary on PATH.

Changes in this release

Fixed

  • A failed environment stops a --wait at once, with the reason, and exits 1. The
    API reports a failure as status error (or failed-debug); the CLI waited for
    failed, which the API never sends. So env create --wait and env resume --wait on a
    failed environment ran to the full timeout (30 minutes) and printed
    timed out (current status: error) without saying why. They now stop at the first poll
    that sees the failure, print the environment's error message, and exit 1.
    env hibernate --wait and env delete --wait do the same; a delete that ends in
    error is no longer reported as a success.
  • A refused create says why and what to do. env create with a short version
    (--version 27) printed only error: invalid_input; the API's sentence ("No BC artifact
    for 27 ... Run 'bcdock artifacts list ...'") was dropped. Any error in that shape now
    prints its message.
  • A version with no ready image says so. The API refuses it with
    request_version_required and no message, and the CLI printed that code and nothing
    else. It now names the version and the region and says to pick from
    bcdock artifacts list --region <region> --fast-only, or to run bcdock env create
    with no version for the picker. With -o json the error code is unchanged.
  • Confirming a platform upgrade on resume gives the CLI command. The API's message
    said "Resubmit with targetVersion"; the CLI now ends it with
    bcdock env resume <env> --version <version> --wait.

Changed

  • env wait exits 1, not 124, when the environment fails. Old: an environment
    that reached a failure status you did not ask for was documented as exit 124, and in
    practice the wait ran to --timeout and then exited 124. New: it exits 1 at once,
    with the environment's error message. 124 still means the wait gave up: --timeout
    elapsed, or the environment settled in another status you did not ask for (such as
    hibernated). A wait that asks for --status failed now exits 1 with the reason too,
    since the API never reports failed. To treat a failure as a success, ask for
    --status error.
  • Help text and docs name the real statuses: error and failed-debug, not failed.
  • Help examples and the --version flag text use full versions (or <version> from
    bcdock artifacts list --region <r> --fast-only) instead of 25.5, which the API
    refuses. The help topics list the error status and give resume as ~7-15 minutes.

Full changelog: CHANGELOG.md
Report issues: support@bcdock.io · Security: see SECURITY.md