This required some other changes WRT verification:
- Change __cmp_like to allow for not accepting <pending> for an IP address.
- Change __verify_with_retry to use a timeout rathan than number of retries. This is a more intuitive interface and aligns with the way we do retries in istio/istio. I also got rid of exponential backoff and allow both the timeout and delay between retries to be configured.
* Try longer timeout for tasks/security/authorization/authz-ingress/test.sh
* Go back to old 5m timeout and add failure on timeout back in
* Test with individual wait_for_istio call updates
* Revert to simply changing timeout (but keep duration in output)
* Fix call
* Comment out istioctl wait call
* Add TODO remark
* Fix lint error
* Add temporary sleep until _wait_for_istio is re-enabled
* Add wait for sleep deployment to complete
* Update test reference
* Test framework changes
* Another required change
* Update Tag to 1.8
* Pick istio/istio commit that actually exists
* Disable ISTIO_META_DNS_CAPTURE
* Add --skip-confirmation to istioctl installl commands
* Increase test timeout. First pass at fixes.
* Update to later istio/istio that fixes DNS and minor fixes
* test fixes
* Pick up go.mod `replace` changes from #8118
* Fix istioctl-analayze and mirror
* Fix mtls-migration test
* Update istio to include commit to fix egress
* Re-enable verify with fix
* Update istio/istio ref for egress fix
* Fix tasks/security/authorization/authz-td-migration - remove ns
* Shorten wait timeout so tests complete in under an hr
* Let tests continue after wait timeout
* Fix --skip-confirmation to -y and use yes | in tests
* revert yes | to echo y |
* Additional echo y fix
* Code review comments
* Change verify from same to contains as k8s 1.19 has extra warning lines.
* automated test - insecure remote access to telemetry addons
* Change profile to demo and add snippets to clean up DestinationRule (besides Gateway and VirtualService)
* automated doc test = secure remote access to telemetry addons through gateway
* Automated test for secure access of common telemetry addons and factor out common snippets to test_utils.sh
* moved test_utils to tests/util/addons.sh so that it can be reused across tests
* refactor addons with a more flexible deploy/undeploy function and fix naming of functions to access addons
* fix cleanup after tests
* fix sourcing
* remove verification for istio config deployment
* don't assume the existence of load-balancer in CI environment, but use _set_ingress_environment_variables instead of snippet in the documentation page
* add verification steps back
* fix output snippets and add verifications
* automated test for viewing traces from zipkin dashboard (expectation is that traces are seen correctly)
* fix shellcheck linting errors
* remove verification of traces and just check if zipkin dashboard is accessible through port-forward
* Modify snippet generation logic to take {{< boilerplate >}} into account
* snippets for boilerplate
* fix shellcheck (linting) issues
* snippets generated with new logic
* automated test - check if zipkin dashboard is accessible by port-forwarding (as we cannot verify screenshot and the trace is already verified in istio/istio integration test
* clean up ZIPKIN_URL comment
* Address code review comments
1. Simplify boilerplate snippet filenames (remove .md_snips)
2. Sourced filenames should be in double quote(") instead of single quote(')
* fix gencheck_istio flake
* do not generate or include boilerplate snippets if there is not {{<text>}} in them
* fix linting errors in snip.py
* fix bug - filter out boilerplates without snippets very early
* add an example task to test
* main test function: save progress
* a working example: routing request
* improve log info and error handling
* introduce makefile
* run each test as a subtest; remove common setup from test.sh
* add another test.sh: fault-injection
* improve error handling
* check test environment
* add two more test.sh files
* fix make command for istio setup
* update two test.sh files from upstream
* add comments and update README.md
* update test.sh files from upstream
* support multiple test names
* update README
* update README.md for new framework
* remove documentation of migration steps
* undo format changes
* change separation line to '# @cleanup'
* move go code and makefile from content/ to tests/
* change package name
* make for loop more readable
* change the set of auto-sourced scripts
* add docs for all functions
* approach to deal with folders with the same name
* minor fixes to ensure everything still runs
* fix make gen error
* add a TIMEOUT argument
* make sure util/debug.sh works with new framework
* make lint-go happy
* [BIG CHANGE] allow different istio setup configs
* make linters happy
* make linters happier
* changed wording and function orders
* make error return as the 2nd argument
* add TODOs
* Update content/en/docs/tasks/traffic-management/traffic-shifting/test.sh
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* only test english docs
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* allow test.sh as suffix
* move adding setup configs to tests/setup
* recommend full paths
* Update tests/README.md
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
* require full test paths
Co-authored-by: Frank Budinsky <frankb@ca.ibm.com>
When tests are run, shell script debugging is enabled and written to a
file called "<script_name>_debug.txt" in the same directory as the
"<script_name>_output.txt" file.
* verify.sh: Don't exit on error in _run_and_verify_*() funcs.
Most tests include "set -e", which causes the script to exit if a
statement returns a non-true return value. In some cases, the command
may exit with a non-true return value, but we want to retry the command
later. We want to temporarily disable that "errexit" behavior.
* verify.sh: Clarify __verify_with_retry() comment and variable name.
* snip.py: Replace github file token with release-specific URL.
* verify.sh: Show the expected output as well as the actual output.
* snip.py: Update the githubfile regex to not include email addresses.
When generating snip scripts, pairs of "@" signs indicate a link to
GitHub repo content. However, JWT attribute values contained pairs of
email addresses such as:
`testing@secure.istio.io/testing@secure.istio.io`
which would be treated as an email address and mangled. This commit
rewrites the regex to not match on email addresses.
* Add authz-jwt user guide test.
Nathan Mittler
Eric Van Norman
Frank Budinsky
Suchith J N
Mitch Connors
Hongyi Zhang
Justin Pettit