1. Local image should be removed, otherwise docker 20 will not tigger
get manifest request to harbor;
2. E2E image Dockerfile update;
3. Fix nighlty test issue of tag retention, add execution refesh to get
result;
4. Fix nighlty test keyword 'Create An New Project And Go Into
Project' issue that waiting long enough time for list display;
5. Add nightly test case, in GUI, scan result will show if cve id exist in allow list configuration;
6. Move proxy cache test to schdule pipeline, it will save some time for
db pipeline.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Add operation id in ctx in baseapi before prepare
2. add operation id for registry proxy request
3. use url for other request
Signed-off-by: DQ <dengq@vmware.com>
1. Add retry for Get Statics Text keywords make this operation robust;
2. Replace image to one with slash in name for CNAB test, to cover more
condition;
3. Add replication rule deletion verfication in upgrade test;
4. Add non-admin user in pull operation for proxy cache test;
5. Add verification for quota display for upgrade test;
6. Add test for large size of image replication;
7. Add test checkpoint for system robot account py-test script;
Signed-off-by: Danfeng Liu (c) <danfengl@vmware.com>
1. Add GUI test for system robot account feature;
2. Fix issue of failing to catch docker pull image exception in API test.
Signed-off-by: Danfeng Liu (c) <danfengl@vmware.com>
Due to complicate logic of scan report in multi-level artifacts, should add scan tests
for into manifest list and CNAB bundle python test suit.
Signed-off-by: danfengliu <danfengl@vmware.com>
Change source of most of test image samples from docker-hub to local building ones, so it will cost less docker-hub pull requests.
And some of cases like push cnab, they have to use docker-hub, but image samples in cnab test will cost 17 quotas, in this PR, we
replace those samples, now cnab case will cost 6 quotas.
Signed-off-by: danfengliu <danfengl@vmware.com>
CI job timeout is 60 minutes, but proxy cache case timeout is 20 minutes, once
proxy cache case is timeout, it will reach CI job timeout, then cause CI job terminated
by timeout without running rest of test steps, expecially uploading useful harbor logs,
so move this test into a new clean job for saving time and debugging.
Signed-off-by: danfengliu <danfengl@vmware.com>
* updates on robot accounts
1, add patch method to refresh secret of a robot
2, fix robot account update issue
3, add editable attribute to handle the version 1 robot account
4, add duration for robot account
5, hide secret for get/list robot account
Signed-off-by: wang yan <wangyan@vmware.com>
* update code per review comments
1, change expirate creation func to AddDate().
2, remove the scanner duration specification, use the default value.
Signed-off-by: Wang Yan <wangyan@vmware.com>
1. In scan all py-test, missing a important test step for checkin scan all status,
2. which indicate the ending of scan all job.
In swagger, ongoig field is missing in stats struct.
Signed-off-by: danfengliu <danfengl@vmware.com>
author Wang Yan <wangyan@vmware.com> 1605849192 +0800
committer Wang Yan <wangyan@vmware.com> 1606361046 +0800
update code per review comments
Signed-off-by: wang yan <wangyan@vmware.com>
1. Use the task manager to manage the underlying execution/task
2. Use the pkg/scheduler to schedule the periodical job
3. Apply the new program model
4. Migration the old data into the new data model
Signed-off-by: Wenkai Yin <yinw@vmware.com>
Due to docker-hub pull request rate limitation, we will use registed account to pull image
from docker-hub, therefore add docker-hub login in API tests.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Since upgrade python to 3.7, some urllib3 warnings come out during
execution, because missing disconnection when using request by docker
library, it's not affect script execution, therefore ignore them first.
2. Replication tests failed in nightly due to new UI code, some element
locators are not valid anymore;
3. Remove switch scanner test steps, since Clair was removed.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Add 7 API python test scripts for tag immutability;
2. Swagger file has object struct defination issue on imuutability policy rule, because
this policy is in shared with retention rule, therefore, they should be unified to
the same type of rule;
3. Fixed a timeout issue of action keyword, waiting for a long period;
4. Add API test scripts for scan signed image.
Signed-off-by: danfengliu <danfengl@vmware.com>
Containerd in e2e image was using native one in ubuntu, it should be updated to the latest release.
And add proxy cache py-tests including pull image/manifest list by docker and ctr CLI.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Verify LDAP group admin in UI;
2. Update LDAP group admin need to verify the AdminRoleInAuth, get current user will return this field as true.
Signed-off-by: danfengliu <danfengl@vmware.com>
LDAP group has different role, user in group has the same role, as groups with different roles were added in project
member list, user should act like in different roles. for admin and dev role, there should be checkpoints to verify priviledges of each own.
Signed-off-by: danfengliu <danfengl@vmware.com>
Robot tests have been always executed in docker container, but API pyt-test is not,
so rebuild Dockerfile for it. During that, some of tools have been upgraded, so API py-test
scripts need some modification.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Fix ldap script issue, test case name should be in lowercase;
2. Add manifest list checkout point for proxy cache issue;
3. Add P2P Preheat policy CRUD Test.
Signed-off-by: danfengliu <danfengl@vmware.com>
1.Separate Dockerfile to drone and nightly e2e, one Dockerfile should cover
both scenarios, it will be resolved later;
2.Fix docker api err type issue, API error has no message attibute;
3.Upgrade docker in nightly e2e images.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Upgrade robot-framework to 3.1 and Selenium library to 4.4.0.
2. Fix a registry issue for clear filter text input.
Signed-off-by: danfengliu <danfengl@vmware.com>
* add debugging env for GC time window
For debugging, the tester/users wants to run GC to delete the removed artifact immediately instead of waitting for two hours, add the env(GC_BLOB_TIME_WINDOW) to meet this.
Signed-off-by: wang yan <wangyan@vmware.com>
two phases:
1, mark, select the gc candidates bases on the DB and mark them as status delete.
2, sweep, select the candidate and mark it as status deleting and remove it from backend and database.
Signed-off-by: wang yan <wangyan@vmware.com>
1. Fix issue of keyword Go Into Repo, the verification logic could be more strict;
2. Add API E2E pytest of GC with untag flag enabled;
3. Add sleep in test_user_view_logs.py for delete log ocurred;
4. Test Case - Tag CRUD is not stable. Although add button was clicked, but the tag was'nt added successfully.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Add oras cli py-test;
2. Add env for notary url, allow to input different notary port instead of solid 4443;
3. Add retry for keyword Cannot Pull Image and make it longer during retry.
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Manifest list can be pulled by ctr;
2. ui-test missing key checkpoint, fix it by add checking scan detail result;
3. add tag retension untag image test;
Signed-off-by: danfengliu <danfengl@vmware.com>
1. Enable `security` in the swagger.yaml.
2. Include `basic` auth in `security` to make the generated python
client by `swagger-codegen-cli` work with basic authorization.
3. Include `anonymous` auth in `security` to make APIs of v2.0 generated
by `goswagger` work with `security` middleware.
Closes#11771
Signed-off-by: He Weiwei <hweiwei@vmware.com>
1. Add docker prune cmd to release some space;
2. Add tag for pytest in case of debugging requirement;
3. Replace image to smaller size in robot account pytest;
Signed-off-by: danfengliu <danfengl@vmware.com>
Fixes#11241
1, remove count quota from quota manager
2, remove count in DB scheme
3, remove UI relates on quota
4, update UT, API test and UI UT.
Signed-off-by: wang yan <wangyan@vmware.com>
Docker CLI fails if it's not logged in upon seeing "basic" realm challenging while pinging the "/v2" endpoint. (#11266)
Some CLI will send HEAD to artifact endpoint before pushing (#11188)(#11271)
To fix such problems, this commit re-introduce the token auth flow to the CLIs.
For a HEAD request to "/v2/xxx" with no "Authoirzation" header, the v2_auth middleware populates the
"Www-Authenticate" header to redirect it to token endpoint with proper
requested scope.
It also adds security context to based on the content of the JWT which has the claims of the registry.
So a request from CLI carrying a token signed by the "/service/token" will have proper permissions.
Signed-off-by: Daniel Jiang <jiangd@vmware.com>
Fixes#11225
As registry changes to basic auth, the push action lost the pull permission.
Add it in the robot security context.
Signed-off-by: wang yan <wangyan@vmware.com>
1, enable user view log api test case
2, update project logs api permission check
3, use project ctl instead in permission check base method
Signed-off-by: wang yan <wangyan@vmware.com>
use the tag controller to handle CRUD of tags, especially the delete scenario, it could validate
the immutable and signature. And move the code of tag handling from artifact controller to tag controller
Signed-off-by: wang yan <wangyan@vmware.com>
1. Add middleware to record the accepted blob size for stream blob
upload.
2. Add middleware to create blob and associate it with project after blob upload
complete.
3. Add middleware to sync blobs, create blob for manifest and associate blobs
with the manifest after put manifest.
4. Add middleware to associate blob with project after mount blob.
5. Cleanup associations for the project when artifact deleted.
Signed-off-by: He Weiwei <hweiwei@vmware.com>
