Repository navigation
Fix more tests and tag ones requiring a server - #792
Merged
Merged
Conversation
initialize_temp_dirs failed when there was no server for every test that uses the temp file helpers, including tests that only touch agents. parser_validate's default manifest step uses mk_tmp_environment_with_teardown, which reads the server's environmentpath, so it now runs only where there is a server. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
The Windows and MacOS cells run the openvox suite without a server and skip the tests tagged server. These 35 untagged tests use the server host or run puppet agent against a server, so they need the tag as well. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
Disabling and enabling the agent need no server, since a disabled agent stops before it contacts one. Without a server the test now runs those steps alone and skips the final agent run. With a server nothing changes. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
Without a server, each agent builds the code tree and applies it with puppet apply, then the same checks run on the results. With a server nothing changes. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
Member
Author
Their file and report checks were blocks passed to Result#stdout, which ignores blocks, so they never ran. The file checks now allow the newline the heredoc writes, and the expected content moves out of the step block that two of the tests set it in, where the checks could not see it. supports_utf8's asset_equal typo is fixed too. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
The clientcert data file was named after $fqdn, a Ruby global that is never set, so the first hierarchy level was always empty and the assertions expected that. Each agent now gets a data file named after its certname, the assertions expect that level first, and two failure messages no longer print nil names. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
v3_config_and_data opened the server's whole confdir, SSL private keys included, with chmod -R 775 and never put it back. It now opens only the hiera.yaml it writes, as v4_hieradata_with_v5_configs does. v4_hieradata_with_v5_configs wrote global data next to its global hiera.yaml, where hiera never reads it and where the teardown left it behind, so those writes are gone. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
catalog_uuid passed :acceptance_exit_codes, which Beaker ignores. puppet_lookup_cmd's trusted facts check passed even when lookup fell back without the CA, since the output is the same, so it now also checks that the fallback warning is absent. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
nmburgan
force-pushed
the
openvox_suite_windows_macos
branch
2 times, most recently
from
November 13, 2026 22:48
3d4f917 to
3784b20
Compare
GitHub's Windows runners name the built-in administrator (RID 500) runneradmin, and Beaker connects as it. Use the account Beaker connects as where the tests act as that user, which stays Administrator under Beaker's Windows defaults, and find the built-in account by its RID where should_purge checks that the purge kept it. QA-506's winexitcode module no longer sets an owner, whose failure skipped every exec the test runs. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
OpenVox printed the runner's domain in lower case, while the test upper cased the hostname. Windows account names are case-insensitive. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
Beaker's local connection, which the macOS cells use, never fills a result's output. The errors these tests expect go to stderr, and calling_all_functions logs at several levels, so it matches on both streams. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
QA-506 applied its manifests without --detailed-exitcodes, so an exec that returned an unexpected exit code failed its resource while puppet apply still exited 0, and the test passed. Now each apply has to exit 2, which means its exec ran and returned the expected code. Signed-off-by: nmburgan <13688219+nmburgan@users.noreply.github.com> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
nmburgan
force-pushed
the
openvox_suite_windows_macos
branch
2 times, most recently
from
November 14, 2026 00:39
0367ad0 to
50b22b8
Compare
bastelfreak
approved these changes
Nov 14, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This includes more fixes for silently broken tests, marks several tests with the
servertag that need one, and fixes up some tests to run serverless where it makes sense to do so so that we can run as many tests under Windows and MacOS as possible.