* 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.
* Add tools/merge_requirements.py
* Revert "Fix oldest tests by pinning Google DNS deps (#5000)"
This reverts commit f68fba2be2.
* Add tools/oldest_constraints.txt
* Remove oldest constraints from tox.ini
* Rename dev constraints file
* Update tools/pip_install.sh
* Update install_and_test.sh
* Fix pip_install.sh
* Don't cat when you can cp
* Add ng-httpsclient to dev constraints for oldest tests
* Bump tested setuptools version
* Update dev_constraints comment
* Better document oldest dependencies
* test against oldest versions we say we require
* Update dev constraints
* Properly handle empty lines
* Update constraints gen in pip_install
* Remove duplicated zope.component
* Reduce pyasn1-modules dependency
* Remove blank line
* pin back google-api-python-client
* pin back uritemplate
* pin josepy for oldest tests
* Undo changes to install_and_test.sh
* Update install_and_test.sh description
* use split instead of partition
* Remove extra le-auto tests from master
* Remove dockerfile-dev test from master
* Remove intermediate Python 3.x tests from master
* Reorder travis jobs for speed
The reason for this PR is many bug fixes in the nginx plugin for changes we haven't released yet are included in #5220 which may not make our next release. If it doesn't, we will (mostly) revert the nginx plugin back to its previous state to avoid releasing these bugs and will revert this PR after the release.
* Revert "Nginx IPv6 support (#5178)"
This reverts commit 68e37b03c8.
* Revert "Fix bug that stopped nginx from finding new server block for redirect (#5198)"
This reverts commit e2ab940ac0.
* Revert "Nginx creates a vhost block if no matching block is found (#5153)"
This reverts commit 95a7d45856.
* Remove assert_called_once from dns-route53
* Remove assert_called_once from main_test.py
* Remove assert_called() usage in dns-digitalocean
* Remove assert_called() usage in dns-route53
* Downgrade mock version in certbot-auto
* 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
* Add tools/pip_constraints.txt to pin all Python dependencies
* Use tools/pip_constraints.txt in tools/pip_install.sh
* Install dnsmadeeasy extras in dnsmadeeasy plugin
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.
* 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
Now we always check if we have root access if --cb-auto-has-root is not given
on the command line. This allows certbot-auto to properly acquire root when
upgrading from an older version. People upgrading from 0.18.0 to 0.18.1 may
check for root access twice, however, if root's user ID is 0, this check is
essentially a noop. If root's user ID is not 0, we'll request root access a 2nd
time during this upgrade.
* Add installer class
* Add wrapped reverter methods to common.Installer.
* Use Installer class in Apache plugin
* Use Installer class in Nginx plugin
* Don't create reverter in Apache and Nginx plugins
* Add version number to bootstrap scripts.
* Always determine Bootstrap function and version.
* Write bootstrap version into venv.
* Add PrevBootstrapVersion function.
* Add OS bootstrapping check to phase 2.
* Differentiate -n and renew when rebootstrapping.
* Quote all environment variables.
* Correct test condition
* Add loud warning about hardcoded version list.
* s/VENV_BOOTSTRAP_VERSION/BOOTSTRAP_VERSION_PATH
* Properly handle noop bootstrap functions.
* Update comment about root usage.
* run all of certbot-auto as root
* remove other $SUDO uses from template
* remove $SUDO usage from bootstrappers
* default venv path = /opt/eff.org/certbot/venv
* Create symlinks from old default venvs
* Delete old venv path when it exists.
Also, quote expansion of paths.
* fix typo
* Separate venv_dir and le_auto_path
* Deduplicate code with test_dirs()
* Ignore cleanup errors.
This is caused by subdirectories being owned by root.
* Split test into test_phase2_upgrade.
* Rename test_dirs to temp_paths for clarity.
* Check both venvs before bootstrapping again.
* Use OLD_VENV_PATH/bin
* Preserve environment with sudo.
* Remove "esp. under sudo" comment.
* Export *VENV_PATH.
* Change check for OLD_VENV installation.
This approach better handles manually set VENV_PATH values.
* Remove SUDO_ENV.
* Print message before requesting root privileges.
* Make a function for selecting root auth method.
* Address @erikrose's feedback.
We now use tools/pip_install_editable.sh which installs our packages using the
pinned versions from certbot-auto.
We also use letsencrypt-auto-source/letsencrypt-auto instead of certbot-auto in
the root to:
1. Make sure OS bootstrappers are up to date with master.
2. Copy letsencrypt-auto-source into our tree so it can be used by
tools/pip_install_editable.sh later.
* Be careful with new interaction from enabling nginx
* Fix py3 compataibility & better docs
* Make minor changes to @pde's PR to unhide nginx
* unhide nginx plugin
* Only protect against nginx interaction in cb-auto
* Hide exceptions that occur during session.close()
This fixes#4840. Exceptions that are raised out of __del__ methods are caught
and printed to stderr. By catching any exceptions that occur, we now prevent
this from happening.
Alternative solutions to this would have been either not calling
session.close() at all or adding a close() method to acme.client.ClientNetwork,
acme.client.Client, and certbot.client.Client and using certbot.client.Client
in a context manager to ensure close() is called. The former means that users
of the ACME library never properly close their connections until their program
exits and the latter adds a lot of complexity and nesting of client code for
little benefit.
* Only catch Exceptions
* Revert "Don't save keys/csr on dry run (#4380)"
This reverts commit e034b50363.
* Don't save CSRs and keys during dry run
* Factor out _test_obtain_certificate_common
* Add test_obtain_certificate_dry_run
* Wrap key from make_key in util.Key
* Wrap result from make_csr in util.CSR
* Pin oldest version of packaged python deps
* Install security extras in oldest tests
* Revert "bump requests requirement to >=2.10 (#4248)"
This reverts commit 402ad8b353.
* Use create=True when patching open on module
* Remove py26 oldest tests.
The only systems where we support Python 2.6 use certbot-auto so the oldest
supported versions of our dependencies are never used when using supported
installation methods. Let's remove this unnecessary and slow test.
* Make tox.ini happy
* Remove py26-oldest from Travis
* Remove dependency on git from pip_install.sh.
Using git allowed this file to continue to work even if it was moved to another
directory. This slight increase in robustness wasn't worth it though as it
broke our development Dockerfile (see #4703), the certbot website's Dockerfile
(see certbot/website#226), and our test farm tests (see
certbot/tests/letstest/scripts/test_apache2.sh for an example that calls
tools/venv.sh without installing git). Rather than continuing to find and patch
these things, let's just allow this script to fail if it's moved rather than
propagating the git dependency all over the place.
* Add readlink.py.
This is the equivalent of `readlink -f` on many Linux systems. This is useful
as there are often differences in readlink on different platforms.
* Use readlink.py in pip_install.sh.
This allows us to work around differences in readlink on macOS.
The program readlink used to be used in integration tests so an environment
variable was used to handle differences in the executable on different systems.
This command is no longer used though so the variable can be removed.
Using ArgumentTypeError causes Certbot to report an unexpected error occurred
while using ArgumentError causes argparse to print more usage information and
call sys.exit().
* Automatically delete temp log file when not used.
This allows close() calls in logging.shutdown() to cause the file to be
deleted when no logging output has been written to the file.
* Make certbot.log.MemoryHandler.flush() a noop.
This causes MemoryHandler.flush() calls in logging.shutdown to be a noop,
allowing us to control when the handler is actually flushed. This prevents log
records from being sent to a temporary file handler for things like
`certbot --version`.
* Keep reference to certbot.log.MemoryHandler.target
In Python 2.7+, the logging module only keeps weak references to created
logging handlers. Because of this, the MemoryHandler's target will not be
properly flushed and closed when logging.shutdown() is called on program exit
unless we keep a reference to it in the MemoryHandler.
* Fixes#4719.
This completes the changes necessary to fix#4719. Now temporary log files are
not created if sys.exit() is called before logging is fully set up. These files
are still created if Certbot crashes for any other reason.
* Document pre_arg_parse_except_hook args.
* Creates SupportedChallengesAction
This fixes#3987 as the call to set_by_default can be removed entirely.
Additionally, logger.warning can be used rather than writing to stderr directly
because #3184 has been resolved and we're guaranteed to having logging setup.
* Move validator to SupportedChallengesAction
supported_challenges_validator was moved to SupportedChallengesAction so
argparse.ArgumentError can be easily used to provide nice error output. Tests
in standalone_test.py were also updated so the module still has 100% test
coverage.
* Better document ArgumentError usage
