CHANGELOG.md

Changes by Version

1.14.0 (2019-09-04)

• Add commonSpec to other jobs (es-index-cleaner, es-rollover, cassandr… (#640)
• Add common spec to dependencies (#637)
• Add resource limits for spark dependencies cronjob (#620)
• Add Jaeger version to Elasticsearch job images (#628)
• Add badger to supported list of storage types (#616)
• Get rid of finalizer, clean sidecars when no jaeger instance found (#575)
• Deploy production ready self provisioned ES by default (#585)
• Always deploy client,data nodes with master node (#586)
• Configure index cleaner properly when rollover is enabled (#587)
• Agent service ports with correct protocol (#579)
• Renamed the ManagedBy label to OperatedBy (#576)
• Added htpasswd option to the OpenShift OAuth type (#573)
• Changed Operator to set ownership of the instances it manages (#571)
• Added upgrade mechanism for managed Jaeger instances (#476)
• Check and update finalizers before setting APIVersion and Kind (#558)
• Remove sidecar when instance is deleted (#453)
• Allow setting es-operator-image (#549)
• Use zero redundancy when number of ES nodes is 1 (#539)
• Use es-operator from 4.1 branch (#537)
• Reinstated the service metrics (#530)
• Use ES single redundancy by default (#531)
• Change replace method, to remain compatible with golang 1.11 (#529)
• Avoid touching the original structure of the options. (#523)
• Prevented the Operator from overriding Secrets/ImagePullSecrets on ServiceAccounts (#526)
• Added support for OpenShift-specific OAuth Proxy options (#508)
• Allowed usage of custom SA for OAuth Proxy (#520)
• Make sure the ES operator's UUID is a valid DNS name (#515)
• Set the ES node GenUUID to explicit value based on jaeger instance namespace and name (#495)
• Add linkerd.io/inject=disabled annotation (#507)

1.13.1 (2019-07-05)

• Bump Jaeger to 1.13 (#504)
• Disable the property ttlSecondsAfterFinished (#503)
• Set default redundancy policy to zero (#501)

1.13.0 (2019-07-02)

• Changed to always use namespace when a name is involved (#485)
• Sanitize names that must follow DNS naming rules (#483)
• Added instructions for daemonsets on OpenShift (#346)
• Enable completion time-to-live to be set on all jobs (#407)

1.12.1 (2019-06-06)

• Removed 'expose metrics port' to prevent 'failed to create or get service' error (#462)
• Add support for securityContext and serviceAccount (#456)
• Add install SDK goal to make (#458)
• Upgraded the operator-sdk version to 0.8.1 (#449)
• Switch to go modules from dep (#449)
• Do not set a default Elasticsearch image (#450)
• Log the operator image name when created (#452)
• Add label to the common spec (#445)
• Fix injecting volumes into rollover jobs (#446)
• Remove race condition by disabling esIndexCleaner till after SmokeTes… (#437)
• Fix runtime panic when trying to update operator controlled resources that don't have annotation or labels (#433)

1.12.0 (2019-05-22)

• Update to 1.12 and use new admin ports (#425)
• Use ephemeral storage for Kafka tests (#419)
• Fix csv example and add spec.maturity (#416)
• Add resources requests/limits to oauth_proxy (#410)
• Check that context is not nil before calling cleanup (#413)
• Improve error message when queries fail (#402)
• Add resource requirements to sidecar agent (#401)
• Add streaming e2e tests (#400)
• Make sure to call ctx.cleanup if perpare()) fails (#389)
• Change how Kafka is configured for collector and ingester (#390)
• Use storage namespace in index cleaner test (#382)
• Fix rbac policy issue with blockOwnerDeletion (#384)
• Reinstate gosec with fix for OOM error (#381)
• Enhance ES index cleaner e2e test to verify indices have been removed (#378)
• Add owner ref on operator's service to ensure it gets deleted when op… (#377)
• Update CSV description to comply with guidelines (#374)
• Include elasticsearch statefulset nodes in availability check (#371)
• Fail lint goal if not empty (#372)

1.11.1 (2019-04-09)

• Include docs for common config (#367)
• Reinstated the registration of ES types (#366)
• Add support for affinity and tolerations (#361)
• Support injection of JAEGER_SERVICE_NAME based on app or k8s recommended labels (#362)
• Change ES operator apiversion (#360)
• Update test to run on OpenShift (#350)
• Add prometheus scrape 'false' annotation to headless collector service (#348)
• Derive agent container/host ports from options if specified (#353)

1.11.0 (2019-03-22)

Breaking changes

• Moved from v1alpha1 to v1 (#265)
• Use storage flags instead of CR properties for spark job (#295)
• Changed from 'size' to 'replicas' (#271). "Size" will still work for the next couple of releases.

Other changes

• Initialise menu to include Log Out option when using OAuth Proxy (#344)
• Change Operator provider to CNCF (#263)
• Added note about the apiVersion used up to 1.10.0 (#283)
• Implemented a second service for the collector (#339)
• Enabled DNS as the service discovery mechanism for agent => collector communication (#333)
• Sorted the container arguments inside deployments (#337)
• Use client certs for elasticsearch (#325)
• Load back Elasticsearch certs from secrets (#324)
• Disable spark dependencies for self provisioned es (#319)
• Remove index cleaner from prod-es-deploy example (#314)
• Set default query timeout for provisioned ES (#313)
• Automatically Enable/disable depenencies tab (#311)
• Unmarshall numbers in options to number not float64 (#308)
• Inject archive index configuration for provisioned ES (#309)
• update #305, add grps and health port to jaeger collector service (#306)
• Enable archive button if archive storage is enabled (#303)
• Fix reverting ingress security to oauth-proxy on openshift if set to none (#301)
• Change agent reporter to GRPC (#299)
• Bump jaeger version to 1.11 (#300)
• Enable agent readiness probe (#297)
• Use storage flags instead of CR properties for spark job (#295)
• Change operator.yaml to use master, to keep the readme uptodate with latest version (#296)
• Add Elasticsearch image to CR and flag (#289)
• Updated to Operator SDK 0.5.0 (#273)
• Block until objects have been created and are ready (#279)
• Add rollover support (#267)
• Added publishing of major.minor image for the operator (#274)
• Use only ES data nodes to calculate shards (#257)
• Reinstated sidecar for query, plus small refactoring of sidecar (#246)
• Remove ES master certs (#256)
• Store back the CR only if it has changed (#249)
• Fixed role rule for Elasticsearch (#251)
• Wait for elasticsearch cluster to be up (#242)

1.10.0 (2019-02-28)

• Automatically detect when the ES operator is available (#239)
• Adjusted logs to be consistent across the code base (#237)
• Fixed deployment of Elasticsearch via its operator (#234)
• Set ES shards and replicas based on redundancy policy (#229)
• Update Jaeger CR (#193)
• Add storage secrets to es-index-cleaner cronjob (#197)
• Removed constraint on namespace when obtaining available Jaeger instances (#213)
• Added workaround for kubectl logs and get pods commands (#225)
• Add -n observability so kubectl get deployment command works correctly (#223)
• Added capability of detecting the platform (#217)
• Deploy one ES node (#221)
• Use centos image (#220)
• Add support for deploying elasticsearch (#191)
• Replaced use of strings.ToLower comparison with EqualFold (#214)
• Bump Jaeger to 1.10 (#212)
• Ignore golang coverage html (#208)

1.9.2 (2019-02-11)

• Enable single operator to monitor all namespaces (#188)
• Added flag to control the logging level (#202)
• Updated operator-sdk to v0.4.1 (#200)
• Added newline to the end of the role YAML file (#199)
• Added mention to WATCH_NAMESPACE when running for OpenShift (#195)
• Added openshift route to role (#198)
• Added Route to SDK Scheme (#194)
• Add Jaeger CSV and Package for OLM integration and deployment of the … (#173)

1.9.1 (2019-01-30)

• Remove debug logging from simple-streaming example (#185)
• Add ingester (and kafka) support (#168)
• When filtering storage options, also include '-archive' related options (#182)

1.9.0 (2019-01-23)

• Changed to use recommended labels (#172)
• Enable dependencies and index cleaner by default (#162)
• Fix log when spak depenencies are used with unsupported storage (#161)
• Fix serviceaccount could not be created by the operator on openshift. (#165)
• Add Elasticsearch index cleaner as cron job (#155)
• Fix import order for collector-test (#158)
• Smoke test (#145)
• Add deploy clean target and rename es/cass to deploy- (#149)
• Add spark job (#140)
• Automatically format imports (#151)
• Silence 'mkdir' from e2e-tests (#153)
• Move pkg/configmap to pkg/config/ui (#152)
• Fix secrets readme (#150)

1.8.2 (2018-12-03)

• Configure sampling strategies (#139)
• Add support for secrets (#114)
• Fix crd links (#132)
• Create e2e testdir, fix contributing readme (#131)
• Enable JAEGER_SERVICE_NAME and JAEGER_PROPAGATION env vars to be set … (#128)
• Add CRD to install steps, and update cleanup instructions (#129)
• Rename controller to strategy (#125)
• Add tests for new operator-sdk related code (#122)
• Update README.adoc to match yaml files in deploy (#124)

1.8.1 (2018-11-21)

• Add support for UI configuration (#115)
• Use proper jaeger-operator version for e2e tests and remove readiness check from DaemonSet (#120)
• Migrate to Operator SDK 0.1.0 (#116)
• Fix changelog 'new features' header for 1.8 (#113)

1.8.0 (2018-11-13)

Notable new Features

• Query base path should be used to configure correct path in ingress (#108)
• Enable resources to be defined at top level and overridden at compone… (#110)
• Add OAuth Proxy to UI when on OpenShift (#100)
• Enable top level annotations to be defined (#97)
• Support volumes and volumeMounts (#82)
• Add support for OpenShift routes (#93)
• Enable annotations to be specified with the deployable components (#86)
• Add support for Cassandra create-schema job (#71)
• Inject sidecar in properly annotated pods (#58)
• Support deployment of agent as a DaemonSet (#52)

Breaking changes

• Change CRD to use lower camel case (#87)
• Factor out ingress from all-in-one and query, as common to both but i… (#91)
• Remove zipkin service (#75)

Full list of commits:

• Query base path should be used to configure correct path in ingress (#108)
• Enable resources to be defined at top level and overridden at compone… (#110)
• Fix disable-oauth-proxy example (#107)
• Add OAuth Proxy to UI when on OpenShift (#100)
• Refactor common spec elements into a single struct with common proces… (#105)
• Ensure 'make generate' has been executed when model changes are made (#101)
• Enable top level annotations to be defined (#97)
• Update generated code and reverted change to 'all-in-one' in CRD (#98)
• Support volumes and volumeMounts (#82)
• Update readme to include info about storage options being located in … (#96)
• Enable storage options to be filtered out based on specified storage … (#94)
• Add support for OpenShift routes (#93)
• Change CRD to use lower camel case (#87)
• Factor out ingress from all-in-one and query, as common to both but i… (#91)
• Fix operator SDK version as master is too unpredicatable at the moment (#92)
• Update generated file after new annotations field (#90)
• Enable annotations to be specified with the deployable components (#86)
• Remove zipkin service (#75)
• Add support for Cassandra create-schema job (#71)
• Fix table of contents on readme (#73)
• Update the Operator SDK version (#69)
• Add sidecar.istio.io/inject=false annotation to all-in-one, agent (da… (#67)
• Fix zipkin port issue (#65)
• Go 1.11.1 (#61)
• Inject sidecar in properly annotated pods (#58)
• Support deployment of agent as a DaemonSet (#52)
• Normalize options on the stub and update the normalized CR (#54)
• Document the disable ingress feature (#55)
• dep ensure (#51)
• Add support for JaegerIngressSpec to all-in-one

1.7.0 (2018-09-25)

This release brings Jaeger v1.7 to the Operator.

Full list of commits:

• Release v1.7.0
• Bump Jaeger to 1.7 (#41)

1.6.5 (2018-09-21)

This is our initial release based on Jaeger 1.6.

Full list of commits:

• Release v1.6.5
• Push the tag with the new commit to master, not the release tag
• Fix git push syntax
• Push tag to master
• Merge release commit into master (#39)
• Add query ingress enable switch (#36)
• Fix the run goal (#35)
• Release v1.6.1
• Add 'build' step when publishing image
• Fix docker push command and update release instructions
• Add release scripts (#32)
• Fix command to deploy the simplest operator (#34)
• Add IntelliJ specific files to gitignore (#33)
• Add prometheus scrape annotations to Jaeger collector, query and all-in-one (#27)
• Remove work in progress notice
• Add instructions on how to run the operator on OpenShift
• Support Jaeger version and image override
• Fix publishing of release
• Release Docker image upon merge to master
• Reuse the same ES for all tests
• Improved how to execute the e2e tests
• Correct uninstall doc to reference delete not create (#16)
• Set ENTRYPOINT for Dockerfile
• Run 'docker' target only before e2e-tests
• 'dep ensure' after adding Cobra/Viper
• Update the Jaeger Operator version at build time
• Add ingress permission to the jaeger-operator
• Install golint/gosec
• Disabled e2e tests on Travis
• Initial working version
• INITIAL COMMIT