Fixes#1473.
writes privkey.pem to 0600 by default for new lineages
on renewals where a new privkey is generated, preserves group mode and gid
Things this PR does not do:
we talked about forcing 0600 on privkeys when a Certbot upgrade is detected. Instead, this PR only creates new lineages with the more restrictive permission to prevent renewal breakages.
this doesn't solve many of the problems mentioned in #1473 that are not directly related to the title issue!
* safe_open on archive keyfiles
* keep group from current lineage
* clean up integration test
* safe_open can follow symlinks
* fix tests on windows, maybe
* Address Brad's comments
* Revert changes to safe_open
* Test chown is called when saving new key
* Reorder chown operation
* Changelog and documentation
* Fix documentation style
After #6485 and #6435, it appears that there is no good reason to not fail fast when test, cover or linting scripts are executed.
This PR ensures to fail fast by invoking commands throught subprocess.check_call instead of subprocess.call, and by removing the handling of non-zero exit code at the end of theses scripts.
As now coverage on Windows is executed with thresholds, I added specific thresholds for this platform. Because some portions of code that are done for Unix platform will not be executed on Windows.
Note that coverage reports from Travis and AppVeyor are accumulated on Codecov. So if a file is covered up to 50 % on Linux, and all other parts are covered on Windows, then coverage is 100 % for Codecov.
Note: that PR also fixes the ability of coverage tests to fail if thresholds are exceeded.
* Use check_call to fail fast in all scripts related to tests/lint/coverage/deploy
* Make specific coverage threshold for windows
Certbot relies heavily on bash scripts to deploy a development environment and to execute tests. This is fine for Linux systems, including Travis, but problematic for Windows machines.
This PR converts all theses scripts into Python, to make them platform independant.
As a consequence, tox-win.ini is not needed anymore, and tox can be run indifferently on Windows or on Linux using a common tox.ini. AppVeyor is updated accordingly to execute tests for acme, certbot and all dns plugins. Other tests are not executed as they are for Docker, unsupported Apache/Nginx/Postfix plugins (for now) or not relevant for Windows (explicit Linux distribution tests or pylint).
Another PR will be done on certbot website to update how a dev environment can be set up.
* Replace several shell scripts by python equivalent.
* Correction on tox coverage
* Extend usage of new python scripts
* Various corrections
* Replace venv construction bash scripts by python equivalents
* Update tox.ini
* Unicode lines to compare files
* Put modifications on letsencrypt-auto-source instead of generated scripts
* Add executable permissions for Linux.
* Merge tox win tests into main tox
* Skip lock_test on Windows
* Correct appveyor config
* Update appveyor.yml
* Explicit coverage py27 or py37
* Avoid to cover non supported certbot plugins on Windows
* Update tox.ini
* Remove specific warnings during CI
* No cover on a debug code for tests only.
* Update documentation and help script on venv/venv3.py
* Customize help message for Windows
* Quote correctly executable path with potential spaces in it.
* Copy pipstrap from upstream
* feat(nginx): add and test new parsing abstractions
* chore(nginx parser): fix mypy and address small comments
* chore(nginx parser): clean up by removing context object
* fix integration test and lint
Main piece of #5810.
* Rename Certbot integration tests
* Remove nginx from certbot tests
* allow for running individual integration tests
* fail under 65
* Add set -e
* Track Nginx coverage and omit it from report later.
* Use INTEGRATION_TEST in script
* add INTEGRATION_TEST=all
* update min certbot percentage
Debian Wheezy is no longer supported (see https://wiki.debian.org/LTS) and
Amazon shut down their Debian 7 mirrors so let's stop trying to use Debian 7
during testing.
* Initial work on new version of --reuse-key
* Test for reuse_key
* Make lint happier
* Also test a non-dry-run reuse_key renewal
* Test --reuse-key in boulder integration test
* Better reuse-key integration testing
* Log fact that key was reused
* Test that the certificates themselves are different
* Change "oldkeypath" to "old_keypath"
* Simply appearance of new-key generation logic
* Reorganize new-key logic
* Move awk logic into TotalAndDistinctLines function
* After refactor, there's now explicit None rather than missing param
* Indicate for MyPy that key can be None
* Actually import the Optional type
* magic_typing is too magical for pylint
* Remove --no-reuse-key option
* Correct pylint test disable
The value for FAKE_DNS is now always the same because Boulder's
docker-compose hardcodes it, so skip some sed.
Set a time limit on how long we'll wait for boulder to come up.
This change will allow registering/updating account with multi emails.
Detail is enclosed in #4242
* support multi emails register
* add more test cases
* update test to unregister before register
* update create path to support multi emaill
* refactor payload updating
* fix typo
* move command line doc to another place
* revert the change for updating account registration info, added unit test
* rearrange text for consistency
See https://wiki.ubuntu.com/Releases.
Ubuntu 15.* repositories have been shut down for months now causing our tests
to always fail on these systems. While the tests on Ubuntu 12.04 still work, it
has been unsupported by Canonical for almost a year and I don't think we should
hamstring ourselves trying to continue to support it ourselves.
* Remove aauthzr instance variable
* If domain begins with fail, fail the challenge.
* test --allow-subset-of-names
* Fix renewal and add extra check
* test after hook checks
* Use newer boulder config
* Use ACMEv2 endpoint if requested
* Add v2 integration tests
* Work with unset variables
* Add wildcard issuance test
* quote domains
* Drop support for EOL Python 2.6
* Use more helpful assertIn/NotIn instead of assertTrue/False
* Drop support for EOL Python 3.3
* Remove redundant Python 3.3 code
* Restore code for RHEL 6 and virtualenv for Py2.7
* Revert pipstrap.py to upstream
* Merge py26_packages and non_py26_packages into all_packages
* Revert changes to *-auto in root
* Update by calling letsencrypt-auto-source/build.py
* Revert permissions for pipstrap.py
* Fix install verb
* Fix error message, tests and remove global pylint change
* Fix boulder integration test keypath
* Also use chain_path from lineage if not defined on CLI
* Fix rebootstrapping before venv move
* add regression test
* dedupe test
* Cleanup case when two venvs exist.
* Add clarifying comment
* Add double venv test to leauto_upgrades
* Fix logic with the help of coffee
* redirect stderr
* pass VENV_PATH through sudo
* redirect stderr
* Make leauto_upgrades do a real upgrade
* Cleanup vars and output
* Sleep until the server is ready
* add simple_http_server.py
* Use a randomly assigned port
* s/realpath/readlink
* wait for server before getting port
* s/localhost/all interfaces
* Add --delete-after-revoke flags
* Use delete_after_revoke value
* Add delete_after_revoke unit tests
* Add integration tests for delete-after-revoke.
* Use pipstrap to install a good version of pip
* Use pytest in cb-auto tests
* Remove nose usage in auto_test.py
* remove nose dev dep
* use pytest in test_tests
* Use pytest in tox
* Update dev dependency pinnings
* remove nose multiprocess lines
* Use pytest for coverage
* Use older py and pytest for old python versions
* Add test for Error.__str__
* pin pytest in oldest test
* Fix tests for DNS-DO plugin on py26
* Work around bug for Python 3.3
* Clarify dockerfile comments
Occasionally a network error prevents Docker from starting boulder causing
Travis tests to fail like it did at
https://travis-ci.org/certbot/certbot/jobs/282923098. This works around the
problem by using travis_retry to try to start boulder again if it fails.
This also moves the logic of waiting for boulder to start into
tests/boulder-fetch.sh so people running integration tests locally can benefit.
* Switching from old branch (issue-4109) and addressing changes requested
in last iteration of review:
80aa857fd2
Requested changes that were addressed:
- fixed outdated docstring for `cert_path_to_lineage`
- removed `full_archive_dir_from_renewal_conf` amd replaced with `full_archive_path` (and `_full_archive_path` -> `full_archive_path`)
- matching on `cert` instead of `chain` in `cert_manager.cert_path_to_lineage`
- fixed the two coding wrongs make a right issue
Requested changes which were not addressed:
- moving `cert_path_to_lineage` from `cert_manager` to `storage`,
as it would introduce a hard to resolve circular dependency.
* Update integration tests to handle default deletion after revoke.
* Swapping test domains.
* Addressing PR feedback:
- calling storage.full_archive_path with a ConfigObj instead of None
- Removing lambda x: x.chain_path as an option to match against
* Addressing PR feedback: it's expected that len(pattern) is 0, so handle that case properly.
* Testing of conflicting values of --cert-name and --cert-path non-interactive mode.
* Silly test for when neither certname nor cert-path were specified.
* Changing archive_files to a private function, because mocking nested functions seems impossible.
* Tests for storage.cert_path_for_cert_name
* Splitting out _acceptable_matches
* Some tests for cert_manager.cert_path_to_lineage
* Offerings to the Lint God
* Cleaner way of dealing with files in archive dirs
* Handling the two different use cases of match_and_check_overlaps a bit better
* late night syntax errors
* Test for when multiple lineages share an archive dir
* Tests for certbot.cert_manager.match_and_check_overlaps
* Removing unneeded nesting
* Lint errors that Travis caught that didn't show up locally
* Adding two integration tests (matching & mismatched --cert-path, --cert-name) based on feedback.
* Asking the user if they want to delete in interactive mode.
* Add hook dir constants
* Add hook dir properties to configuration
* test hook dir properties
* reuse certbot.util.is_exe
* Add certbot.hooks.list_hooks
* test list_hooks
* Run pre-hooks in directory
* Run deploy-hooks in directory
* Run post-hooks in directory
* Refactor and update certbot/tests/hook_test.py
* Add integration tests for hook directories
* Have Certbot create hook directories.
* document renewal hook directories
* Add --no-directory-hooks
* Make minor note about locale independent sorting
Up until now, this test was written incorrectly. In addition, when it has
failed, it simply prints error messages rather than reporting that the test
failed. This fixes both of these problems.
* prevent regressions of #5082
* Fix test_leauto_upgrades.sh
test_leauto_upgrades.sh has been incorrectly been succeeding because while peep
doesn't work with newer versions of pip and letsencrypt-auto would crash,
the output included the version number so we reported the test as passing.
This updates letsencrypt-auto to the oldest version that still works for the
purpose of the test and sets pipefail so errors are properly reported.
* Test symlink creation in test_leauto_upgrades.sh
* Pin dependencies in test_sdists.sh.
* Fix permissions errors in test_tests.sh
