Moved boltdb to a driver implementation for such interface
Added CreateCacheDatabaseDriver in controller
Fixed default directory creation (boltDB will only create the file, not the dir
Added coverage tests
Added example config for boltdb
Re-added caching on subpaths, rewrote CreateCacheDatabaseDriver
Fix tests
Made cacheDriver argument mandatory for NewImageStore, added more validation, added defaults
Moved cache interface to own file, removed useRelPaths from config
Got rid of cache config, refactored
Moved cache to own package and folder
Renamed + removed cache factory to backend, replaced CloudCache to RemoteCache
Moved storage constants back to storage package
moved cache interface and factory to storage package, changed remoteCache defaulting
Signed-off-by: Catalin Hofnar <catalin.hofnar@gmail.com>
Issues fixed:
- the cli calls reaching out to the catalog endpoint used to request signature manifests
- resty was used instead of the cli http client to check if the discovery api was available
but it did not take into account TLS verification configuration
(testing locally withself-signed certificates did not work)
(cherry picked from commit ca42031ae9b1ceb459f5cd4f86cb82b3c9f78157)
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
- Digests were represented by different ways
- We needed a uniform way to represent the digests and enforce a format
- also replace usage of github.com/google/go-containerregistry/pkg/v1
with github.com/opencontainers/image-spec/specs-go/v1
Signed-off-by: Laurentiu Niculae <niculae.laurentiu1@gmail.com>
(cherry picked from commit 96b2f29d6d57070a913ce419149cd481c0723815)
(cherry picked from commit 3d41b583daea654c98378ce3dcb78937d71538e8)
Co-authored-by: Laurentiu Niculae <niculae.laurentiu1@gmail.com>
Should help with test flakiness
(the order in which the policies are read from the map impacted tested codepath)
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
* list all images that are base images for the given image + zli command
Signed-off-by: Lisca Ana-Roberta <ana.kagome@yahoo.com>
* Fix a failing test
The test expected the image size to be the size of the layer, not the manifest+config+layer
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
Signed-off-by: Lisca Ana-Roberta <ana.kagome@yahoo.com>
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
Co-authored-by: Lisca Ana-Roberta <ana.kagome@yahoo.com>
currently different subpaths can only point to same root directory only
when one or both of the storage config does not enable dedupe
different subpath should be able to point to same root directory and in
that case their storage config should be same i.e GC,Dedupe, GC delay
and GC interval
Signed-off-by: Shivam Mishra <shimish2@cisco.com>
- Decrease RLIMIT_NOFILE and the number of goroutines used to reach this limit (from 512 to 100)
- Reset RLIMIT_NOFILE to the initial value before the test finishes
- Remove panic
- Use temporary dir managed by test framework
- Swith to using test logging in pkg/cli/stress_test.go
- Execute commands without `bash -c` in pkg/cli/stress_test.go
First item is needed as the GH runner seems to stop the test if stressed too much.
The lower number is still good enough to reproduce the test conditions
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
We encountered some problems with using the existing folder structure,
but it looks like running the tooling with the latest versions works after
we regenerated the project using 'gql init' and refactoring to separate
the login previously in resolvers.go.
- the autogenerated code is now under the gql_generated folder
- the file resolvers.go now contains only the code which is not
rewritten by the gqlgen framework
- the file schema.resolvers.go is rewritten when gqlgen runs,
and we'll only keep there the actual resolvers matching query names
Changes we observed to schema.resolvers.go when gqlgen runs include
reordering methods, and renaming function parameters to match the
names used in schema.graphql
- we now have a gqlgen.yaml config file which governs the behavior of
gqlgen (can be tweaked to restructure the folder structure of the
generated code in the future)
Looks like the new graphql server has better validation
1 Returns 422 instead of 200 for missing query string - had to update tests
2 Correctly uncovered an error in a test for a bad `%` in query string.
As as result of 2, a `masked` bug was found in the way we check if images are
signed with Notary, the signatures were reasched for with the media type
of the image manifest itself instead of the media type for notation.
Fixed this bug, and improved error messages.
This bug would have also been reproducible with main branch if the bad `%`
in the test would have fixed.
Updated the linter to ignore some issues with the code which is
always rewritten when running:
`go run github.com/99designs/gqlgen@v0.17.13 generate`
Add a workflow to test gqlgen works and has no uncommitted changes
Signed-off-by: Andrei Aaron <andaaron@cisco.com>
Multiple go routines downloading trivy db
triggers data race on trivy internal db.Path().
In each go routine wait for db download to start.
closes#636
Signed-off-by: Petu Eusebiu <peusebiu@cisco.com>
config file may get removed before fsnotify starts watching it
make sure the config file gets removed when test ends, closes#608
Signed-off-by: Petu Eusebiu <peusebiu@cisco.com>
Files were added to be built whether an extension is on or off.
New build tags were added for each extension, while minimal and extended disappeared.
added custom binary naming depending on extensions used and changed references from binary to binary-extended
added automated blackbox tests for sync, search, scrub, metrics
added contributor guidelines
Signed-off-by: Alex Stan <alexandrustan96@yahoo.ro>
PR (linter: upgrade linter version #405) triggered lint job which failed
with many errors generated by various linters. Configurations were added to
golangcilint.yaml and several refactorings were made in order to improve the
results of the linter.
maintidx linter disabled
Signed-off-by: Alex Stan <alexandrustan96@yahoo.ro>
Signed-off-by: laurentiuNiculae <themelopeus@gmail.com>
Separated updateDistSpec functionality
Removed rewriting of config when distSpecVersion was wrong