summaryrefslogtreecommitdiff
path: root/ci/verify-chef.sh
Commit message (Collapse)AuthorAgeFilesLines
* Replace Jenkins omnibus build/test pipeline with BuildkiteJeremiah Snapp2019-06-241-114/+0
| | | | Signed-off-by: Jeremiah Snapp <jeremiah@chef.io>
* update the ci verify tasks for chef-15lcg/update-ci-verifyLamont Granquist2019-05-011-0/+1
| | | | | | | | | | | | we need to bundle install now in both places (ChefDK style of appbundling means there is no Gemfile.lock here and we do the install against the somewhat novel-looking Gemfile that kinda looks like Gemfile.lock). and on windows side since there is no Gemfile.lock any more we need to avoid that test. Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
* Add some additonal RPM db cleanup steps missed in #7845Seth Chisamore2018-11-011-1/+3
| | | | Signed-off-by: Seth Chisamore <schisamo@chef.io>
* Automatically rebuild RPM database on EL6 testersschisamo/cleanup-rpmdbSeth Chisamore2018-11-011-0/+12
| | | | | | | | Our tests hammer YUM pretty hard and the EL6 testers get corrupted after some period of time. Rebuilding the RPM database clears up the underlying corruption. We'll do this each test run just to be safe. Signed-off-by: Seth Chisamore <schisamo@chef.io>
* Remove the existing acceptance testing frameworkkill-acceptanceTim Smith2018-06-261-25/+2
| | | | | | | | We haven't used this framework for 2 major releases of Chef and as time goes on it's more and more likely we'll build something very different for testing. We 100% do need acceptance testing, but between dokken testing in Travis and the potential for more complex integration testing with build-kite I think we should do the proper cleanup and remove this code now. Signed-off-by: Tim Smith <tsmith@chef.io>
* Revert "Run all tests in CI"Thom May2018-05-011-1/+1
|
* Run all tests in CItm/run_all_tests_in_ciThom May2018-04-251-1/+1
| | | | | | | | | | Previously, we ran only integration tests. But many of our unit tests are constrained by rootyness and platform, and this means the first time they get run is during the ChefDK verification stage, which uses the released Chef build. So we have no way of fixing test failures short of a new Chef release. Signed-off-by: Thom May <thom@chef.io>
* Kill any orphaned testsshain/test_killerScott Hain2018-02-201-0/+3
| | | | Signed-off-by: Scott Hain <shain@chef.io>
* Force deletion of tmpdir with sudoshain/force_delete_tmpdirScott Hain2018-01-161-1/+1
| | | | | | This makes sure we don't fail on the very slim chance that an OS generated file (lookin' at you, AIX) ends up in this directory and blocks deletion. Signed-off-by: Scott Hain <shain@chef.io>
* simplify omnibus config and greenify builds againLamont Granquist2017-05-091-16/+10
| | | | | | | | | | | | | | this is also necessary for bundler-1.14.x i'm still not entirely clear why we ever needed all the fussy software gem configs or what the build-chef / build-chef-gem infrastructure ever did for us. it seems to have been mostly micro-optimization around building the software gems before bundle installing the project in order to take advantage of git caching. i aggressively don't care about that, this is quite fast enough. we can install nokogiri and libgecode early and that should take care of 98% of the build optimization issue. Signed-off-by: Lamont Granquist <lamont@scriptkiddie.org>
* Reenable non-windows testsKartik Null Cating-Subramanian2016-07-211-2/+2
|
* Turn all those blasted acceptance tests offstupid1ksubrama/new_acceptanceKartik Null Cating-Subramanian2016-07-131-2/+2
|
* copy chef-acceptance suites into $WORKSPACESeth Chisamore2016-05-171-6/+8
| | | | | | | | | | | | | This will ensure the path to the suites can be fully resolved, it prevents errors like: ``` STDERR: /home/jenkins/workspace/chefdk-test/architecture/x86_64/platform/acceptance/project/chefdk/role/tester/vendor/bundle/ruby/2.1.0/gems/mixlib-shellout-2.2.6/lib/mixlib/shellout/unix.rb:183:in `chdir': No such file or directory @ dir_chdir - /home/jenkins/workspace/chefdk-test/architecture/x86_64/platform/acceptance/project/chefdk/role/tester/opscode-ci/.acceptance/acceptance-cookbook (Errno::ENOENT) ```
* Revert "exit 0 -- its how acceptance works"Seth Chisamore2016-05-171-20/+19
| | | | This reverts commit aac62e28ae3112c586f223c48ef7fe360f98e56a.
* exit 0 -- its how acceptance worksMatt Wrock2016-05-171-19/+20
|
* Execute chef-acceptance without elevated privilegesSeth Chisamore2016-05-131-5/+7
| | | | | | | | | | | | We output all data generated during a chef-acceptance run to the workspace of the executing Jenkins job using chef-acceptance's `--data-path` option. If the `chef-acceptance` commands are executed with elevated privileges (using `sudo`) the generated files are owned by `root`: https://gist.github.com/schisamo/7ce7262813f2bc81b7314d9eab53afa0 This is an issue as the generated data cannot be properly archived OR cleaned up when the next job runs: https://gist.github.com/schisamo/b7246987d49534b27b8a4ad72f9ad965
* Add `KITCHEN_CHEF_CHANNEL` value now `ARTIFACTORY_USERNAME` that was removedSeth Chisamore2016-05-101-1/+1
| | | See https://github.com/chef/chef/blob/master/acceptance/.shared/kitchen_acceptance/libraries/kitchen.rb#L19
* Write chef-acceptance data to the Jenkins `$WORKSPACE`Seth Chisamore2016-05-101-1/+1
| | | | | | | | Previously data was being written to the Chef package install path of `/opt/chef/*/acceptance/*` which is bad for numerous reasons. Now that chef/chef-acceptance#39 has been merged we can take advantage of Chef Acceptance's new `--data-path` option which allows us to relocate this directory into the Jenkins job's workspace for easy archiving.
* Remove legacy `ARTIFACTORY_*` env variablesSeth Chisamore2016-05-101-4/+4
| | | | These have not been needed since: chef/mixlib-install#90
* Run the intergration tests in FIPS modeKartik Null Cating-Subramanian2016-04-251-7/+1
|
* Turn off GIT_SSL_NO_VERIFYjk/use-ssl-againJohn Keiser2016-04-191-3/+1
|
* Just use the actual installed bundle instead of reusing the Gemfile to run ↵John Keiser2016-04-181-1/+1
| | | | Chef tess
* Don't update bundler when running acceptance testJohn Keiser2016-04-181-1/+3
|
* Install only core chef plus test when testingJohn Keiser2016-04-181-3/+1
|
* Don't bundle install on testsJohn Keiser2016-04-181-1/+1
|
* Run acceptance from the acceptance dir in the Chef gemJohn Keiser2016-04-181-1/+1
|
* Remove Gemfile.windows and put both platforms in Gemfile.lockJohn Keiser2016-04-181-6/+20
|
* Add --force-destroy to chef-acceptancedanielsdeleo2016-03-221-1/+1
|
* remove final cleanup from ci verify scripts to return proper exit codepw/fix_verify_shPatrick Wright2016-02-161-3/+0
|
* Disable unit tests in later stages of CD pipelineschisamo/disable-unitSeth Chisamore2016-02-111-1/+1
| | | | | | | Travis/Appveyor already run the unit tests at the appropriate gate earlier in the build process. Running the tests on each platform we build a package for should give the same result and really just serves to slow the overall delivery process down.
* Create special tempdir and destroy it on each ci runJohn Keiser2016-02-101-0/+9
|
* Expand the list of acceptance tests (run all suites)salam/acceptance-updateSalim Alam2016-02-081-1/+1
|
* Make sure acceptance tests for chef can run in Jenkins.sersut/ec2-acceptanceSerdar Sutay2016-01-271-17/+31
|
* Fix sh spacing issuesJay Mundrawala2016-01-271-1/+1
| | | | The if statement was always being evaluated to true.
* Fix bug in sh script that runs test in jenkinsjdm/fix-pipelineJay Mundrawala2016-01-261-3/+4
| | | | | | | | | | | Some testers fail with ``` ci/verify-chef.sh: test: ] missing` ``` This regression was introduced by https://github.com/chef/chef/commit/2bcb7d06dcb49c953b9e666b24c8bd3cb56da2c5
* export CHEF_FIPS=1 in the chef-fips jenkins pipelineJay Mundrawala2016-01-251-1/+6
|
* Move Jenkins verification scripts to top-levelSeth Chisamore2016-01-111-0/+89
View Lorry Controller Status