omnibus build fixes #1821
Merged
omnibus build fixes #1821
Conversation
This file contains 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
Berkshelf 7 has gotten stable and includes a fix for handling tar archives that have extended headers (that are appearing in cookbooks these days). This should fix our omnibus build erroring out while attempting to vendor the cookbooks required by the internal cookbook. Signed-off-by: Robb Kidd <[email protected]>
In an effort to reduce the number of dependencies in the omnibus-internal cookbook, this removes the unicorn cookbook as a dependency. The unicorn cookbook was being used for its configuration template only. Opted to copy the template directly into the omnibus-internal cookbook and manage it ourselves. Signed-off-by: Robb Kidd <[email protected]>
The inspec software definition in omnibus-software does not (yet?) also install the inspec-bin gem to get binstub executables created. Because the supermarket-ctl command is the thing that needs inspec as a command to shell out to _and_ it is managed as a software definition _and_ the software definition is already bundle-installing, I've added inspec and inspec-bin to the Gemfile used by the software definition. This results in the latest _release_ inspec, as opposed to whatever is current on the master branch, getting installed within Supermarket as an executable. Licensing is handled at the call site in the -ctl command that uses inspec. Since we're bring in the latest, we need to accept the license for the duration of the -ctl test. Signed-off-by: Robb Kidd <[email protected]>
robbkidd
added
Aspect: Packaging
By adding a `reconfigure` command to the `general` category, we can override the one provided by omnibus-ctl that performs license checking for a project. It was decided that the use of chef-client within Supermarket's -ctl commands was a "library use," so a non-persistent acceptance at the call sites is acceptable. This reconfigure command looks much like omnibus-ctl's, but with the license review and acceptance removed. This command still uses the `run_chef` method and passes in the non-persistent acceptance as an argument. This change also updates omnibus-ctl because it was woefully behind. Signed-off-by: Robb Kidd <[email protected]>
Updating these -ctl commands so that they appear under appropriate categories instead of floating up at the top of the help output without any context. Signed-off-by: Robb Kidd <[email protected]>
"Properly format header comments." OK. Signed-off-by: Robb Kidd <[email protected]>
robbkidd
force-pushed
the
robb/update-berkshelf
branch
from
19b19a8 to
680d115
Compare
|
Here's how it is. 🎉 This branch successfully builds an omnibus package again. 😭 Travis testing fails because the Travis builder fails to start PostgreSQL 9.3. PG 9.3 went end-of-life in roughly November 2018, so it is reasonable that Travis would stop making that available in their builders. I think these changes to fix the omnibus build are required for whatever else we do to address PostgreSQL. |
Signed-off-by: Christopher A. Snapp <[email protected]>
Signed-off-by: Christopher A. Snapp <[email protected]>
Add EL8 tester to build pipeline
Signed-off-by: Robb Kidd <[email protected]>
Because 9.3 isn't supported any more. Supermarket really needs to upgrade the version of PostgreSQL embedded in the install. In the meantime, we'll test one version up. Signed-off-by: Robb Kidd <[email protected]>
Nokogiri and yard needed updating. Easy. omniauth has a vulnerability that isn't resolved with a newer version of that gem. While we're sorting out that resolution, disable the check for only that CVE. Signed-off-by: Robb Kidd <[email protected]>
Bring in the latest defaults, particularly new git, libarchive, and openssl. Signed-off-by: Robb Kidd <[email protected]>
|
I bumped the version of PostgreSQL the CI tests use from v9.3 to v9.4, but we still ship 9.3 in the omnibus package. I don't think an upgrade of PostgreSQL is in scope for this PR—which is only about getting the build working again—but I am wary of how far to kick that can down the road. Anyone have ideas about what sort of timer we should put on making tests fail again if we're still embedding PG 9.3? |
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.
Description
What started as an update of the berkshelf used for the omnibus-internal cookbook to solve #1817 turned into a bit of an ordeal. There are a few distinct changes that all related to getting the omnibus package to build and behave like it always has. It may make more sense to walk the commits with the explanations in the commit messages than to review the PR's collective set of changes.
Related Issue
Closes #1817
Types of changes
New feature (non-breaking change which adds functionality)Breaking change (fix or feature that would cause existing functionality to change)Chore (non-breaking change that does not add functionality or fix an issue)Checklist:
I have updated the documentation accordingly.I have added tests to cover my changes.