policy-reporter/CHANGELOG.md

Changelog

2.19.3

• Helm Chart
  • Fix ingress TLS rendering

2.19.3

• Helm Chart
  • Fix Ingress TLS block [#317 by rufusnufus]
  • Make inverval and scrapeTimeout configurable on ServiceMonitors [#318 by fhielpos]

2.19.2

• Policy Reporter
  • Fix mapping of mountedSecret for all missing targets
  • Implement retry for secretRef secret fetching

2.19.1

• Policy Reporter
  • AWS IRSA Authentication Support
  • Add source attribute to JSON ouput [#311 by nikolay-o]

2.19.0

• Policy Reporter
  • New AWS SecurityHub push target - See values.yaml for available configurations
  • External DB support (PostgreSQL, MySQL, MariaDB) - See values.yaml for available configurations
    • HA Mode support - only leader write into the DB
    • Versioned Schema, autoupdated when another version is detected
    • Configurable over values and secrets
  • Cache improvements to reduce duplicated pushes
  • Split Category API into namespaced scoped and cluster scoped API
  • Support search for contained words in the results API
• Policy Reporter UI
  • Update API requests

2.18.3

• Policy Reporter
  • new value to add priorityClassName to pods [#283 by boniek83]
  • fixed syntax error for policy reporter config.yaml [#295 by nikolay-o]
  • fixed customFields for kinesis targets [#295 by nikolay-o]
  • image signing and sbom generation for new Policy Reporter images

2.18.2

• Policy Reporter UI
  • Container signing and SBOM generation
  • New config api.overwriteHost to control the proxy host behavior

2.18.1

• Signed Helm Chart

• Policy Reporter

  • New channel property for Slack targets to define the Slack channel to send the results too
  • New mountedSecret property to read target configs from a mounted secret [#282 by rromic]
  • AWS KMS support for S3 target with new properties bucketKeyEnabled, kmsKeyId and serverSideEncryption [#281 by rromic]
    • Mountet secret needs to be in json format with keys defined in kubernetes/secrets Values struct.

• Monitoring

  • Add namespaceSelector to serviceMonitor values

2.18.0

• Policy Reporter
  • Improved logging configuration
    • Support JSON logging
    • Support log level
  • optional API access logging with api.logging set to true
  • New aggregation table for API performance improvements
  • Helm Ingress template
  • New Google Cloud Storage Target
    • Requires credentials as JSON String and the bucket name
    • Added in the helm valus under target.gcs
• Policy Reporter KyvernoPlugin
  • Helm Ingress template
  • Improved logging configuration
    • Support JSON logging
    • Support log level
• Policy Reporter UI
  • Improved logging configuration
    • Support JSON logging
    • Support log level
    • Proxy Logging

2.17.0

• Policy Reporter
  • Use metaclient to reduce informer memory usage
  • Use workerqueue to control concurrent processing of PolicyReports
  • Remove internal PolicyReport structures
  • Make sqlite volume configurable [#255 by monotek]
  • use defer to unlock when possible [#259 by eddycharly]
• Policy Reporter UI
  • New SSL configs for external clusters
    • skipTLS to disable SSL verification
    • certificate to configure a path to a custom CA for self signed URLs
  • New Helm values ui.volumes and ui.volumeMounts to add your custom CAs as mounts to the UI deployment.

2.16.0

• Add nameOverride to all charts [#254 by mjnagel]

2.15.0

• Add values to configure topologySpreadConstraints for all components [#241 by Kostavro]
• Fixing comment formats and deprecations [#250 by fengshunli]
• Add new APIs for PolicyReport and ClusterPolicyReport metadata (/v1/policy-reports, /v1/cluster-policy-reports) [#251
• search filter also checks the resource kind
• Use correct probes in core deployment [#236 by rgarcia89]
• Add source to PolicyReport Table and improve report-label API [#252

2.14.1

• Policy Reporter
  • Fix generate multiple custom metrics

2.14.0

• Policy Reporter
  • Persist also PolicyReport labels
  • API
    • New API to get available labels for PolicyReports: /v1/namespaced-resources/report-labels
    • New API to get available labels for ClusterPolicyReports: /v1/cluster-resources/report-labels
  • Metrics
    • special syntax to add report labels to metric labels: label:report-label-name, special characters like -, /, ., : will be transformed to _ in metrics
  • New Target Filter reportLabel to, filter results based on labels of the related (Cluster)PolicyReport
• Monitoring
  • New values to disable dedicated Grafana Dashboards:
    • grafana.dashboards.enable.overview, default true
    • grafana.dashboards.enable.policyReportDetails, default true
    • grafana.dashboards.enable.clusterPolicyReportDetails, default true
  • New values to configure the Grafana Dashboard datasource label, pluginName, pluginId
    • grafana.datasource.label, default Prometheus
    • grafana.datasource.pluginName, default Prometheus
    • grafana.datasource.pluginId, default prometheus
  • New value grafana.dashboards.labelFilter to add custom report labels as dashboard filter, default []. Label has to be a valid prometheus label, e.g. created-by => created_by.
  • New values grafana.dashboards.multicluster.enabled and grafana.dashboards.multicluster.label to add an optional cluster label.
• Kyverno Plugin
  • New HTML Compliance Reports
    • Grouped by Policy with Details per Namespace and Resource
    • Grouped by Namespace with Details per Policy and Resource
  • Go update to 1.19
• UI
  • Integrate new Compliance Reports
  • New PolicyReport label based filter, use ui.labelFilter to define a list of labels to add
  • Go update to 1.19

2.13.5

• Add configuration target.s3.pathStyle for the S3 output

2.13.4

• Fix customFields mapping in TargetFactory

2.13.3

• Fix customFields property in values.yaml
• Fix PolicyReporter image.tag version

2.13.2

• Policy Reporter
  • Add customFields property to missing targets: Elasticsearch, S3, Webhook, Kinesis
• Policy Reporter UI
  • Create Links out of URL property values
• Monitoring
  • New monitoring.serviceMonitor.honorLabels and monitoring.kyverno.serviceMonitor.honorLabels value: chooses the metrics labels on collisions with target labels [#216 by monotek]

2.13.1

• Policy Reporter
  • Fix persist error for duplicated IDs
  • Disable UI SA automount

2.13.0

• Policy Reporter
  • New certificate config for loki, elasticsearch, teams, webhook and ui, to set the path to your custom certificate for the related client.
  • New skipTLS config for loki, elasticsearch, teams, webhook and ui, to skip tls if needed for the given target.
  • New secretRef for targets to reference a secret with the related username, password, webhook, host, accessKeyID, secretAccessKey information of the given target, instead of configure your credentials directly.
• Policy Reporter UI
  • New value refreshInterval to configure the default refresh interval for API polling. Set 0 to disable polling.
• Policy Reporter Kyverno Plugin
  • Fix the creation of duplicated results for PolicyReportResults.

2.12.0

• Policy Reporter
  • New Helm Chart value to add extra volumes to PolicyReporter deployment [#186 by preved911]
  • HTTP Basic authentication for Elasticsearch targets with username and password configuration fields
  • target.slack.customFields map property for Slack pushes to add additional metadata to notifications like clustername
  • Add timestamp to Result REST APIs
  • Overwrite the installation target namespace via the new global.namespace value.

2.11.3

• Policy Reporter
  • New emailReports.smtp.secret configuration to use an existing external secret to configure your SMTP connection
    • You can set all or a subset of the available keys in your secret: host, port, username, password, from, encryption
    • Keys available in your secret have a higher priority as your Helm release values.

2.11.2

• Policy Reporter
  • Add new Severity values info and critical
  • Update PolicyReport ID generierung
• Policy Reporter UI
  • Fix Grouping by Policy and Categories
  • Fix ReverseProxy RequestHost
  • New configuration ui.clusterName which is used in the ClusterSelect, if you configure additional Clusters
• Policy Reporter Kyverno Plugin
  • Add time property to PolicyReportResults

2.11.1

• Policy Reporter
  • Fix CronJob Resources by [#157 by MaxRink]
• Policy Reporter UI
  • Fix API Proxy for APIs behind ReverseProxy (like NGINX Ingress)

2.11.0

• Policy Reporter

  • High Availability support with leaderelection for necessary features like target pushes, to avoid duplicated pushes by multiple instances
    • Add new role and rolebinding to manage lease objects if leaderelection is enabled
  • Add redis configuration to the Helm Chart for external cache storage
  • Add PodDisruptionBudget for HA Deployments (replicaCount > 1)
  • Add skipTLS configuration for MS Teams Webhook

• Policy Reporter KyvernoPlugin

  • High Availability support with leaderelection for necessary features like PolicyReport management for blocked resources
    • Add new role and rolebinding to manage lease objects if leaderelection is enabled
  • Add PodDisruptionBudget for HA Deployments (replicaCount > 1)
  • Internal refactoring for better CRD management

• Policy Reporter UI

  • Add redis as possible log storage to support high availability deployments
  • Add PodDisruptionBudget for HA Deployments (replicaCount > 1)

2.10.3

• Policy Reporter
  • Add new config target.loki.path to overwrite the deprected prom push API

2.10.2

• Policy Reporter UI
  • New option ui.clusters makes it possible to configure additional external Policy Reporter APIs (details)
  • General UI improvements for loading state and error handling

2.10.1

• Monitoring
  • Fix Datasource for Metrics and Filters in the preconfigured Dashboards
  • Add Datasource as additional Select to the preconfigured Dashboards

2.10.0

• Policy Reporter
  • Email Reports
    • Send Summary Reports over SMTP to different E-Mails
    • Supports channels and filters to send different subsets of Namespaces or Sources to dedicated E-Mails
    • Reports are generated and send over dedicated CronJobs, this makes it easy to send the reports as often as needed
    • Currently a basic summary and a more detailed violation report is available and can be separatly enabled and configured
  • Metrics
    • Add metrics.mode for less or custom metric values, to reduce cardinality
  • Monitoring
    • Fix Source Column for result tables
    • Fix Warn counter for ClusterPolicyReport Details

2.9.5

• Fix Policy Reporter Version in the Helm Chart values.yaml

2.9.4

• Policy Reporter
  • Add AWS Kinesis compatible target
  • Add new Helm value profiling.enabled to enable pprof profiling, disabled by default
  • Improved Informer handling

2.9.3

• Policy Reporter
  • Fix grafana.dashboards.value type conversion [fix #158]

2.9.2

• Policy Reporter
  • Add grafana.dashboards.value value to configure the ConfigMap label value for the Prometheus Operator by [#157 by stone-z]

2.9.1

• Policy Reporter
  • Name Configuration for Target (Channels) to customize UI Labels
• Policy Reporter UI
  • Fix table on chip selection
  • Order labels
  • Return 404 Status Code for non existing URL paths

2.9.0

• Policy Reporter
  • New configuration to use Redis as external result caching store
  • SQLite Improvement: Use batch insertion for PolicyReportResults
  • PolicyReport Informer Update: Use typed informer to improve performance and memory usage
  • Drop support for v1alpha1 of the PolicyReport CRD
  • Serverside Pagination for better Dashboard performance
  • Concurrent PolicyReport processing
• Policy Reporter UI
  • Serverside Pagination support
  • Dynamic Chart sizes
• Policy Reporter Kyverno Plugin
  • Generate Policy Reports for enforcement violations

2.8.0

• Policy Reporter
  • New target filter and channels to define multiple configurations of the same target
    • Filter target results by exclude and include rules for namesapces, priorities and policies
    • Support wildcards for policies and namespaces
  • New webhook target
    • this target is a simple way to send notifications to custom tools and APIs
    • results are send as POST requests with a JSON representation of the result
    • the headers properties allows you to send custom header with the request to allow for example authentication

2.7.1

• Policy Reporter
  • Add Resource APIVersion to the Results REST APIs

2.7.0

• Policy Reporter
  • PolicyReport Filter:
    • PolicyReporter CRD Filter by Namespaces
    • Disable ClusterPolicyReport CRD processing

2.6.3

• Policy Reporter
  • Fix Debouncer has wrong reference to OldPolicyReport when a result was cached.

2.6.2

• Policy Reporter

  • Update Go to 1.17.8
  • Add serviceMonitor.relabelings and serviceMonitor.metricRelabelings for ServiceMonitor configuration in the monitoring Subchart.
  • Add kyverno.serviceMonitor.relabelings and kyverno.serviceMonitor.metricRelabelings for the KyvernoPlugin ServiceMonitor configuration in the monitoring Subchart.

• Policy Reporter UI

  • Update Go to 1.17.8

• Policy Reporter KyvernoPlugin

  • Update Go to 1.17.8

2.6.1

• Update Policy Reporter UI to v1.3.2
  • Support access over Subpaths, e.g. Rancher Reverse Proxy
• Update Policy Reporter Monitoring to v2.1.0
  • Fix Failing ClusterPolicyRules Columns of the PolicyReports Dashboard
  • Add Filter to the PolicyReports Dashboard

2.6.0

• Add seccomp profile support [#120 by eddycharly]

2.5.0

• New Policy Reporter API to get a list of available resources
• New Filter for Policies, Kinds, Categories and Results APIs

2.4.0

• Policy Reporter
  • Add Support for custom Loki labels

2.3.0

• Policy Reporter

  • Add Support for linux/s390x [#115 by skuethe]

• Policy Reporter UI

  • Add Support for linux/s390x [#98 by skuethe]

• Policy Reporter KyvernoPlugin

  • Add Support for linux/s390x [#13 by skuethe]

2.2.6

• Use upper case on drop capabilities [#113 by skuethe]

2.2.5

• Policy Reporter

  • Update Go to 1.17.6 [#110 by realshuting]
  • Update Helm Chart with new component versions
  • Update dependencies

• Policy Reporter UI

  • Update Go to 1.17.6 [#93 by realshuting]
  • Update dependencies

• Policy Reporter KyvernoPlugin

  • Update Go to 1.17.6 [#12 by realshuting]

2.2.4

• Fix PolicyReport Napper - string casting

2.2.3

• Fix Helm Chart uihost template function.

2.2.2

• Fix Helm Chart values.yaml. Cleanup unused default configurations. [#103 by AndersBennedsgaard]

2.2.1

• Fix Typo in values.yaml [#102 by christophefromparis]

2.2.0

• Policy Reporter UI v1.2.0
  • New configurations to customize the dashboard by disable PolicyReport- or ClusterPolicyReport information

2.1.1

• Fix KyvernoPlugin Metrics ServiceMonitor Port [#96 by z0rc]
• Remove unused Port from KyvernoPlugin Deployment and Service

2.1.0

• KyvernoPlugin v1.1.0
  • New KyvernoPlugin API - VerifyImages Rules (details)
• Policy Reporter UI v1.1.0
  • New Kyverno VerifyImages view in Policy Reporter UI
  • New configurations to disable views (details)

2.0.1

• Remove NetworkPolicy ingress rule for UI if not enabled
• Update Policy Reporter UI
  • Fix: Show PolicyReportResult Properties in Tables

2.0.0

Chart

• Removed deprecated values crdVersion, cleanupDebounceTime
• Simplify policyPriorities, policyPriorities.enabled was removed along with the watch feature
  • Priority determined mainly over severity
• Add sources filter to target configurations
• Improved NetworkPolicy configuration for all components
• Metrics now an optional feature
• Each component expose a single Port 8080

See Migration Docs for details

Policy Reporter

• modular functions for separate activation/deactivation
  • REST API
  • Metrics API
  • Target pushes
• PolicyReports are now stored in an internal SQLite
• extended REST API based on the new SQLite DB for filters and grouping of data
• metrics API is now optional
• metrics and REST API using the same HTTP Server (were separated before)
• improved CRD watch logic with Kubernetes client informer
• Yandex changed to a general S3 target.

Policy Reporter UI

• Rewrite with NuxtJS
• Simplified Proxy
• Improved SPA file handling

Policy Reporter Kyverno Plugin

• modular functions for separate activation/deactivation
  • REST API
  • Metrics API
• metrics and REST API using the same HTTP Server (were separated before)
• improved CRD watch logic with Kubernetes client informer

1.12.6

• Update Go Base Image for all Components
  • Policy Reporter [#90 by fjogeleit]
  • Policy Reporter UI [#11 by realshuting]
  • Policy Reporter Kyverno Plugin [#9 by realshuting]

1.12.5

• Dependency Update

1.12.4

• Fix policy-reporter-ui ServiceName function [#87 by m-yosefpor]

1.12.3

• Fix policy-reporter-ui backend name [#85 by m-yosefpor]

1.12.2

• Fix CRD registration for PolicyReport and ClusterPolicyReport

1.12.0

• Add Yandex as new Target for Policy Reporter

1.11.0

• Add Yandex as new Target for Policy Reporter

1.10.0

• Update Policy Reporter UI to v0.15.0
  • Add Filters as Query Parameters, make them shareable over links
• Hosting all new Images on the GitHub Container Registry instead of DockerHub
• Go Version updates to Go 1.17 of all components

1.9.4

• Make the Image Registry configurable with image.registry [#74 by stone-z]

1.9.3

• Fix loki target messages for labels with dots

1.9.2

• Add additional egress rules to kyvernoPlugin and UI subchart with networkPolicy.egress

1.9.1

• Configure the Kubernetes API Port for NetworkPolicy with networkPolicy.kubernetesApiPort

1.9.0

• Implement NetworkPolicy for Policy Reporter and related Components [#68 by windowsrefund]
• Customize liveness- and readinessProbe for Policy Reporter [#67 by windowsrefund]

1.8.10

• Fix ServiceMonitor Namespace overwrite with monitoring.serviceMonitor.namespace instead of monitoring.namespace

1.8.9

• Ensure Backward Compatibility for monitoring.namespace configuration

1.8.8

• Optional Namespace Configuration for Monitoring ServiceMonitor
• Separat Namespace Configuration for Monitoring ConfigMaps with monitoring.grafana.namespace

1.8.7

• Update Policy Reporter UI to 0.14.0
  • Colored Diagrams
  • Suppport SubPath Configuration
• Restart CRD Watches when no CRDs are found
• Fix Ingress Resource in the UI Subchart
• Allow to override namespace for serviceMonitor [#57 by Issif]

1.8.6

• Update Policy Reporter UI to 0.13.1
  • Hide Rule Chips if rule name is empty
• Update Policy Reporter Kyvern Plugin to 0.3.2
  • Improved LivenessProbe, checks now if Kyverno CRDs are available
• Update Policy Reporter to 1.8.4
  • Improved LivenessProbe, checks now if any PolicyReport CRD is available

1.8.5

• Added trusted root CA's [#52 by frezbo]

1.8.4

• Changed Organization

1.8.3

• Update Policy Reporter UI to 0.13.0
  • Change Result Grouping between by Status and by Category
  • Add source filter to ClusterPolicyReports

1.8.2

• Fix scored mapping for v1alpha2/policyreports
• Disable KyvernPlugin as default as expected
• Support source and properties for policyreports/v1alpha2 in Policy Reporter UI
  • Update Policy Reporter UI to 0.12.0

1.8.1

• Customize label and annotation for Grafana dashboards [#43 by nlamirault]
• ARM64 Support for all Components

1.7.3

• Update Policy Reporter - Kyverno Plugin to 0.2.0
  • New APIs for Liveness and Readiness Probes

1.7.2

• Update Policy Reporter - Kyverno Plugin to 0.1.2
  • Fix Handling of Validations with empty messages

1.7.1

• Fix HelmChart - Deployment Probes for Policy Reporter

1.7.0

• Enable REST API by default
  • Add /healthz and /ready APIs as new endpoints for readinessProbe and livenessProbe
• Helm Chart Updates
  • Add global.labels to add labels on every resource created
  • Add default labels on every resource

1.6.2

• Increase Result Caching Time to handle Kyverno issues with Policy reconcilation Issue
• Fix golint errors

1.6.1

• Add .global.fullnameOverride as new configuration for Policy Reporter Helm Chart
• Add static manifests to install Policy Reporter without Helm or Kustomize

1.6.0

• Internal refactoring
  • Unification of PolicyReports and ClusterPolicyReports processing, APIs still stable
  • DEPRECETED crdVersion, Policy Reporter handels now both versions by default
  • DEPRECETED cleanupDebounceTime, new internal caching replaced the debounce mechanism, debounce still exist with a fixed period to improve stable metric values.

1.5.0

• Support multiple Resources for a single Result
  • Mapping Result with multiple Resources in multiple Results with a single Resource
  • Upate UI handling with Results without Resources

1.4.1

• Update Kyverno Plugin
  • Fix Rule Type mapping
• Update Policy Reporter UI
  • Fix Chart rerender when values are the same

1.4.0

• Add Kyverno Plugins to the Helm Chart

1.3.4

• Configure Debounce Time in seconds for Cleanup Events over Helm Chart
  • Helm Value cleanupDebounceTime - default: 20
• Improved securityContext defaults

1.3.3

• Update Policy Reporter UI to v0.9.0
  • expand Tables with Validation Message
• Reduce log messages

1.3.2

• Compress REST API with GZIP
• Update Policy Reporter UI to 0.8.0
  • Support for GZIP Responses

1.3.1

• Debounce reconcile modification events for 10s to prevent resending violations

1.3.0

• New Helm Configuration
  • crdVersion changes the version of the PolicyReporter CRD - v1alpha1 is the current default

1.2.3

• Fix resend violations after KubeAPI reconnect

1.2.2

• Fix PolicyReportResult.timestamp parsing

1.2.1

• Support PolicyReportResult.status as well as PolicyReportResult.result for newer CRD versions

1.2.0

• Support for (Cluster)PolicyReport CRD Properties in Target Output
• Support for (Cluster)PolicyReport CRD Timestamp in Target Output
• Fix resend violations after Kyverno Cleanup with ResultHashes

1.1.0

• Added PolicyReport Category to Metrics
• New (Cluster)PolicyReport filter for Grafana Dashboards
  • Add All Selection for Policy Filter
  • Category Filter
  • Severity Filter
  • Kind Filter
  • Namespacefilter (PolicyReports only)
• New (Cluster)PolicyReport filter for Policy Reporter UI
  • Category Filter
  • Severity Filter
  • Kind Filter

1.0.0

• Support Priority by Severity
  • high -> critical
  • medium -> warning
  • low -> information
• Severity is added as label to result metrics
• Severity is added in Policy Reporter UI tables
• Add "Critical" as new Priority to differ between Errored Policies and Failed priorities with High Severity
• Use "Warning" as new default Priority instead of Error which should now used for Policies in Error Status

0.22.0

• New Target Policy Reporter UI
  • New Log View in the Policy Reporter UI to see the latest log entries
  • Default: latest 200 logs with priority >= warning

0.21.0

• New Target MS Teams

0.20.2

• Policy Reporter UI update
  • Select All option for Policy Filter
  • New Namespace Filter for PolicyReport View

0.20.0

• [Breaking Change] rename policy-reporter-ui Subchart to ui
  • Simplify the customization by configure all PolicyReporter UI values under ui

0.19.0

• PolicyResult Priority mapping is now configurable over the Helm Chart

0.18.0

• Helm Chart updates #16 fixes #14
  • Target Configuration are now configured under target in the HelmChart values.yaml
  • config.yaml are now deployed as Secret with encoded data body (plain stringData before)

0.17.0

• New Helm Linting Workflow by kolikons #15
• Improved Helm Chart by kolikons #13
  • More configuration possibilities like UI Ingress, ReplicaCount
  • Role and RoleBindings for ConfigMaps are now optional (required for Priority configuration)

0.16.0

• New Optional REST API
• New Optional Policy Reporter UI Helm SubChart

0.15.1

• Add a checksum for the target configuration secret to the deployment. This enforces a pod recreation when the configuration changed by a Helm upgrade.

0.15.0

• Customizable Dashboards via new Helm values for the Monitoring Subchart.

0.14.0

• Internal refactoring
  • Improved test coverage
  • Removed duplicated caching
• Updated Dashboard
  • Filter zero values from Policy Report Detail after Policies / Resources are deleted

0.13.0

• Split the Monitoring out in a Sub Helm chart
  • Changed naming from metrics to monitoring
• Make Annotations for the Deployment configurable
• Add two new Grafana Dashboard (PolicyReport Details, ClusterPolicyReport Details)

0.12.0

• Add support for a special default key in the Policy Priority. The default key can be used to configure a global default priority instead of error

0.11.1

• Use a Secret instead of ConfigMap to persist target configurations

0.11.0

• Helm Chart Value metrics.serviceMonitor changed to metrics.serviceMonitor.enabled
• New Helm Chart Value metrics.serviceMonitor.labels can be used to add additional labels to the SeriveMonitor. This helps to fullfil the serviceMonitorSelector of the Prometheus Resource in the MonitoringStack.

0.10.0

• Implement Discord as Target for PolicyReportResults

0.9.0

• Implement Slack as Target for PolicyReportResults

0.8.0

• Implement Elasticsearch as Target for PolicyReportResults
• Replace CLI flags with a single config.yaml to manage target-configurations as separate ConfigMap
• Set loki.skipExistingOnStartup default value to true