自定义 Service Fabric 群集设置Customize Service Fabric cluster settings

本文介绍可以自定义的 Service Fabric 群集的各种结构设置。This article describes the various fabric settings for your Service Fabric cluster that you can customize. 对于 Azure 中托管的群集,可以通过 Azure 门户或使用 Azure 资源管理器模板自定义设置。For clusters hosted in Azure, you can customize settings through the Azure portal or by using an Azure Resource Manager template. 对于独立群集,可通过更新 ClusterConfig.json 文件并对群集执行配置升级来自定义设置。For standalone clusters, you customize settings by updating the ClusterConfig.json file and performing a configuration upgrade on your cluster. 有关详细信息,请参阅升级独立群集的配置For more information, see Upgrade the configuration of a standalone cluster.

有三种不同的升级策略:There are three different upgrade policies:

  • Dynamic - 对动态配置的更改不会导致 Service Fabric 进程或服务主机进程的任何进程重启。Dynamic - Changes to a dynamic configuration do not cause any process restarts of either Service Fabric processes or your service host processes.
  • Static - 对静态配置的更改会导致 Service Fabric 节点重启,以便使用该更改。Static - Changes to a static configuration will cause the Service Fabric node to restart in order to consume the change. 节点上的服务将重启。Services on the nodes will be restarted.
  • NotAllowed - 不能修改这些设置。NotAllowed - These settings cannot be modified. 若要更改这些设置,需要销毁该群集并创建一个新群集。Changing these settings requires that the cluster be destroyed and a new cluster created.

以下为可自定义的 Fabric 设置列表(按分区排序)。The following is a list of Fabric settings that you can customize, organized by section.

ApplicationGateway/HttpApplicationGateway/Http

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ApplicationCertificateValidationPolicyApplicationCertificateValidationPolicy string,默认值为“None”string, default is "None" 静态Static 不会验证服务器证书;请求成功。This does not validate the server certificate; succeed the request. 请参阅配置 ServiceCertificateThumbprints,获取反向代理可以信任的远程证书的指纹列表(以逗号分隔)。Refer to config ServiceCertificateThumbprints for the comma-separated list of thumbprints of the remote certs that the reverse proxy can trust. 请参阅配置 ServiceCommonNameAndIssuer,获取反向代理可以信任的远程证书的使用者名称和证书颁发者指纹。Refer to config ServiceCommonNameAndIssuer for the subject name and issuer thumbprint of the remote certs that the reverse proxy can trust. 若要了解详细信息,请参阅反向代理安全连接To learn more, see Reverse proxy secure connection.
BodyChunkSizeBodyChunkSize Uint,默认值为 16384Uint, default is 16384 动态Dynamic 提供用于读取正文的区块大小(以字节为单位)。Gives the size of for the chunk in bytes used to read the body.
CrlCheckingFlagCrlCheckingFlag uint,默认值为 0x40000000uint, default is 0x40000000 动态Dynamic 应用程序/服务证书链验证的标记;例如 CRL 检查 0x10000000 CERT_CHAIN_REVOCATION_CHECK_END_CERT 0x20000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN 0x40000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN_EXCLUDE_ROOT 0x80000000 CERT_CHAIN_REVOCATION_CHECK_CACHE_ONLY 设置为 0 会禁用 CRL 检查,支持值的完整列表由 CertGetCertificateChain 的 dwFlags 记录: https://msdn.microsoft.com/library/windows/desktop/aa376078(v=vs.85).aspxFlags for application/service certificate chain validation; e.g. CRL checking 0x10000000 CERT_CHAIN_REVOCATION_CHECK_END_CERT 0x20000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN 0x40000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN_EXCLUDE_ROOT 0x80000000 CERT_CHAIN_REVOCATION_CHECK_CACHE_ONLY Setting to 0 disables CRL checking Full list of supported values is documented by dwFlags of CertGetCertificateChain: https://msdn.microsoft.com/library/windows/desktop/aa376078(v=vs.85).aspx
DefaultHttpRequestTimeoutDefaultHttpRequestTimeout 以秒为单位的时间。Time in seconds. 默认值为 120default is 120 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 提供用于 http 应用网关中正在处理的 http 请求的默认请求超时时间。Gives the default request timeout for the http requests being processed in the http app gateway.
ForwardClientCertificateForwardClientCertificate bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 如果设置为 false,反向代理不会请求客户端证书。如果设置为 true,反向代理将在 TLS 握手期间请求客户端证书,并将 base64 编码的 PEM 格式字符串转发到名为 X-Client-Certificate 的标头中的服务。检查证书数据后,服务可能无法处理请求,并返回相应的状态代码。When set to false, reverse proxy will not request for the client certificate.When set to true, reverse proxy will request for the client certificate during the TLS handshake and forward the base64 encoded PEM format string to the service in a header named X-Client-Certificate.The service can fail the request with appropriate status code after inspecting the certificate data. 如果此参数为 true 并且客户端不提供证书,反向代理将转发空标头,并让服务处理该情况。If this is true and client does not present a certificate, reverse proxy will forward an empty header and let the service handle the case. 反向代理将充当透明层。Reverse proxy will act as a transparent layer. 若要了解详细信息,请参阅设置客户端证书身份验证To learn more, see Set up client certificate authentication.
GatewayAuthCredentialTypeGatewayAuthCredentialType string,默认值为“None”string, default is "None" 静态Static 指示在 http 应用网关终结点处使用的安全凭据的类型,有效值为 None/X509。Indicates the type of security credentials to use at the http app gateway endpoint Valid values are None/X509.
GatewayX509CertificateFindTypeGatewayX509CertificateFindType string,默认值为“FindByThumbprint”string, default is "FindByThumbprint" 动态Dynamic 指示如何在由 GatewayX509CertificateStoreName(FindByThumbprint;FindBySubjectName)支持的值指定的存储中搜索证书。Indicates how to search for certificate in the store specified by GatewayX509CertificateStoreName Supported value: FindByThumbprint; FindBySubjectName.
GatewayX509CertificateFindValueGatewayX509CertificateFindValue string,默认值为“”string, default is "" 动态Dynamic 用于查找 http 应用网关证书的搜索筛选器值。Search filter value used to locate the http app gateway certificate. 此证书在 https 终结点上配置,并且如果服务需要,还可用于验证应用的标识。This certificate is configured on the https endpoint and can also be used to verify the identity of the app if needed by the services. 首先查找 FindValue;如果其不存在,再查找 FindValueSecondary。FindValue is looked up first; and if that does not exist; FindValueSecondary is looked up.
GatewayX509CertificateFindValueSecondaryGatewayX509CertificateFindValueSecondary string,默认值为“”string, default is "" 动态Dynamic 用于查找 http 应用网关证书的搜索筛选器值。Search filter value used to locate the http app gateway certificate. 此证书在 https 终结点上配置,并且如果服务需要,还可用于验证应用的标识。This certificate is configured on the https endpoint and can also be used to verify the identity of the app if needed by the services. 首先查找 FindValue;如果其不存在,再查找 FindValueSecondary。FindValue is looked up first; and if that does not exist; FindValueSecondary is looked up.
GatewayX509CertificateStoreNameGatewayX509CertificateStoreName string,默认值为“My”string, default is "My" 动态Dynamic 包含 http 应用网关证书的 X.509 证书存储的名称。Name of X.509 certificate store that contains certificate for http app gateway.
HttpRequestConnectTimeoutHttpRequestConnectTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(5)TimeSpan, default is Common::TimeSpan::FromSeconds(5) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 提供用于正在从 http 应用网关发送的 http 请求的连接超时时间。Gives the connect timeout for the http requests being sent from the http app gateway.
IgnoreCrlOfflineErrorIgnoreCrlOfflineError bool,默认值为 TRUEbool, default is TRUE 动态Dynamic 是否要忽略应用程序/服务证书验证的 CRL 脱机错误。Whether to ignore CRL offline error for application/service certificate verification.
IsEnabledIsEnabled Bool,默认值为 falseBool, default is false 静态Static 启用/禁用 HttpApplicationGateway。Enables/Disables the HttpApplicationGateway. 默认情况下,禁用 HttpApplicationGateway,需要设置此配置以启用 HttpApplicationGateway。HttpApplicationGateway is disabled by default and this config needs to be set to enable it.
NumberOfParallelOperationsNumberOfParallelOperations Uint,默认值为 5000Uint, default is 5000 静态Static 要发布到 http 服务器队列的读取数。Number of reads to post to the http server queue. 此配置控制 HttpGateway 可以满足的并发请求数。This controls the number of concurrent requests that can be satisfied by the HttpGateway.
RemoveServiceResponseHeadersRemoveServiceResponseHeaders string,默认值为“Date; Server”string, default is "Date; Server" 静态Static 从服务响应中删除的响应标头列表(以分号/逗号分隔),这些标头将子啊转发到客户端之前删除。Semi colon/ comma-separated list of response headers that will be removed from the service response; before forwarding it to the client. 如果此值设置为空字符串,则会按原样传递服务返回的所有标头。If this is set to empty string; pass all the headers returned by the service as-is. 例如i.e 不会覆盖日期和服务器do not overwrite the Date and Server
ResolveServiceBackoffIntervalResolveServiceBackoffInterval 以秒为单位的时间,默认值为 5Time in seconds, default is 5 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 提供重试失败的解析服务操作之前的默认回退时间间隔。Gives the default back-off interval before retrying a failed resolve service operation.
SecureOnlyModeSecureOnlyMode bool,默认值为 FALSEbool, default is FALSE 动态Dynamic SecureOnlyMode:true:反向代理只会转发到发布安全终结点的服务。SecureOnlyMode: true: Reverse Proxy will only forward to services that publish secure endpoints. false:反向代理可以将请求转发到安全/不安全的终结点。false: Reverse Proxy can forward requests to secure/non-secure endpoints. 若要了解详细信息,请参阅反向代理终结点选择逻辑To learn more, see Reverse proxy endpoint selection logic.
ServiceCertificateThumbprintsServiceCertificateThumbprints string,默认值为“”string, default is "" 动态Dynamic 反向代理可以信任的远程证书的指纹的逗号分隔列表。The comma-separated list of thumbprints of the remote certs that the reverse proxy can trust. 若要了解详细信息,请参阅反向代理安全连接To learn more, see Reverse proxy secure connection.

ApplicationGateway/Http/ServiceCommonNameAndIssuerApplicationGateway/Http/ServiceCommonNameAndIssuer

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup X509NameMap,默认值为 NoneX509NameMap, default is None 动态Dynamic 反向代理可以信任的远程证书的使用者名称和颁发者指纹。Subject name and issuer thumbprint of the remote certs that the reverse proxy can trust. 若要了解详细信息,请参阅反向代理安全连接To learn more, see Reverse proxy secure connection.

BackupRestoreServiceBackupRestoreService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
MinReplicaSetSizeMinReplicaSetSize int,默认值为 0int, default is 0 静态Static BackupRestoreService 的 MinReplicaSetSizeThe MinReplicaSetSize for BackupRestoreService
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 静态Static BackupRestore 服务的 PlacementConstraintsThe PlacementConstraints for BackupRestore service
SecretEncryptionCertThumbprintSecretEncryptionCertThumbprint string,默认值为“”string, default is "" 动态Dynamic 机密加密 X509 证书指纹Thumbprint of the Secret encryption X509 certificate
SecretEncryptionCertX509StoreNameSecretEncryptionCertX509StoreName string,推荐的值为“My”(无默认值)string, recommended value is "My" (no default) 动态Dynamic 这表示用来加密和解密 X.509 证书存储的凭据名称的证书,备份还原服务用此加密解密存储凭据This indicates the certificate to use for encryption and decryption of creds Name of X.509 certificate store that is used for encrypting decrypting store credentials used by Backup Restore service
TargetReplicaSetSizeTargetReplicaSetSize int,默认值为 0int, default is 0 静态Static The TargetReplicaSetSize for BackupRestoreServiceThe TargetReplicaSetSize for BackupRestoreService

ClusterManagerClusterManager

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AllowCustomUpgradeSortPoliciesAllowCustomUpgradeSortPolicies Bool,默认值为 falseBool, default is false 动态Dynamic 是否允许自定义升级排序策略。Whether or not custom upgrade sort policies are allowed. 启用此功能后,这用于分两个阶段进行升级。This is used to perform 2-phase upgrade enabling this feature. Service Fabric 6.5 现支持在群集或应用程序升级期间为升级域指定排序策略。Service Fabric 6.5 adds support for specifying sort policy for upgrade domains during cluster- or application upgrades. 支持的策略为 Numeric、Lexicographical、ReverseNumeric 和 ReverseLexicographical。Supported policies are Numeric, Lexicographical, ReverseNumeric and ReverseLexicographical. 默认设置为 Numeric。The default is Numeric. 为了能够使用此功能,在 SF 6.5 代码完成升级之后,必须将群集清单设置 ClusterManager/ AllowCustomUpgradeSortPolicies 设置为 True,使其成为第二个配置升级步骤。To be able to use this feature, the cluster manifest setting ClusterManager/ AllowCustomUpgradeSortPolicies must be set to True as a second config upgrade step after the SF 6.5 code has completed upgrade. 有必要分两个阶段完成此操作,否则代码升级可能会在第一个升级期间混淆升级顺序。It is important that this is done on two phases, otherwise the code upgrade may get confused about the upgrade order during the first upgrade.
EnableDefaultServicesUpgradeEnableDefaultServicesUpgrade Bool,默认值为 falseBool, default is false 动态Dynamic 在应用程序升级期间启用升级默认服务。Enable upgrading default services during application upgrade. 升级后,会覆盖默认服务说明。Default service descriptions would be overwritten after upgrade.
FabricUpgradeHealthCheckIntervalFabricUpgradeHealthCheckInterval 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 受监视结构升级期间的运行状况检查频率The frequency of health status check during a monitored Fabric upgrade
FabricUpgradeStatusPollIntervalFabricUpgradeStatusPollInterval 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 轮询结构升级状态的频率。The frequency of polling for Fabric upgrade status. 此值确定任何 GetFabricUpgradeProgress 调用的更新速率This value determines the rate of update for any GetFabricUpgradeProgress call
ImageBuilderTimeoutBufferImageBuilderTimeoutBuffer 以秒为单位的时间,默认值为 3Time in seconds, default is 3 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 所允许的映像生成器特定超时错误用于返回到客户端的时间量。The amount of time to allow for Image Builder specific timeout errors to return to the client. 如果此缓冲区太小,客户端会先于服务器超时,并收到一个泛型超时错误。If this buffer is too small; then the client times out before the server and gets a generic timeout error.
InfrastructureTaskHealthCheckRetryTimeoutInfrastructureTaskHealthCheckRetryTimeout 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在对基础结构任务进行后处理时,重试失败的运行状况检查所花费的时间。The amount of time to spend retrying failed health checks while post-processing an infrastructure task. 观察通过的运行状况检查将重置此计时器。Observing a passed health check will reset this timer.
InfrastructureTaskHealthCheckStableDurationInfrastructureTaskHealthCheckStableDuration 以秒为单位的时间,默认值为 0Time in seconds, default is 0 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在成功完成基础结构任务的后处理之前,用于观察连续通过的运行状况检查的时间量。The amount of time to observe consecutive passed health checks before post-processing of an infrastructure task finishes successfully. 观察失败的运行状况检查将重置此计时器。Observing a failed health check will reset this timer.
InfrastructureTaskHealthCheckWaitDurationInfrastructureTaskHealthCheckWaitDuration 以秒为单位的时间,默认值为 0Time in seconds, default is 0 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 对基础结构任务的后处理完成后,开始运行状况检查之前的等待时间量。The amount of time to wait before starting health checks after post-processing an infrastructure task.
InfrastructureTaskProcessingIntervalInfrastructureTaskProcessingInterval 以秒为单位的时间,默认值为 10Time in seconds, default is 10 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 基础结构任务处理状态机使用的处理时间间隔。The processing interval used by the infrastructure task processing state machine.
MaxCommunicationTimeoutMaxCommunicationTimeout 以秒为单位的时间,默认值为 600Time in seconds, default is 600 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. ClusterManager 与其他系统服务(即命名服务、故障转移管理器等)之间的内部通信的最大超时时间。The maximum timeout for internal communications between ClusterManager and other system services (i.e.; Naming Service; Failover Manager and etc.). 此超时时间应小于全局 MaxOperationTimeout(因为对于每个客户端操作,系统组件之间可能有多个通信)。This timeout should be smaller than global MaxOperationTimeout (as there might be multiple communications between system components for each client operation).
MaxDataMigrationTimeoutMaxDataMigrationTimeout 以秒为单位的时间,默认值为 600Time in seconds, default is 600 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 发生 Fabric 升级后,数据迁移恢复操作的最大超时时间。The maximum timeout for data migration recovery operations after a Fabric upgrade has taken place.
MaxOperationRetryDelayMaxOperationRetryDelay 以秒为单位的时间,默认值为 5Time in seconds, default is 5 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 遇到故障时,内部重试的最大延迟时间。The maximum delay for internal retries when failures are encountered.
MaxOperationTimeoutMaxOperationTimeout 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 用于内部处理 ClusterManager 上的操作的最大全局超时时间。The maximum global timeout for internally processing operations on ClusterManager.
MaxTimeoutRetryBufferMaxTimeoutRetryBuffer 以秒为单位的时间,默认值为 600Time in seconds, default is 600 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 因超时导致内部重试时,最大操作超时时间是 <Original Time out> + <MaxTimeoutRetryBuffer>The maximum operation timeout when internally retrying due to timeouts is <Original Time out> + <MaxTimeoutRetryBuffer>. 以 MinOperationTimeout 为增量添加额外超时时间。Additional timeout is added in increments of MinOperationTimeout.
MinOperationTimeoutMinOperationTimeout 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 用于内部处理 ClusterManager 上的操作的最小全局超时时间。The minimum global timeout for internally processing operations on ClusterManager.
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 3Int, default is 3 不允许Not Allowed ClusterManager 的 MinReplicaSetSize。The MinReplicaSetSize for ClusterManager.
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 不允许Not Allowed ClusterManager 的 PlacementConstraints。The PlacementConstraints for ClusterManager.
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 不允许Not Allowed 指定以秒为单位的时间范围。Specify timespan in seconds. ClusterManager 的 QuorumLossWaitDuration。The QuorumLossWaitDuration for ClusterManager.
ReplicaRestartWaitDurationReplicaRestartWaitDuration 以秒为单位的时间,默认值为 (60.0 * 30)Time in seconds, default is (60.0 * 30) 不允许Not Allowed 指定以秒为单位的时间范围。Specify timespan in seconds. ClusterManager 的 ReplicaRestartWaitDuration。The ReplicaRestartWaitDuration for ClusterManager.
ReplicaSetCheckTimeoutRollbackOverrideReplicaSetCheckTimeoutRollbackOverride 以秒为单位的时间,默认值为 1200Time in seconds, default is 1200 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 如果 ReplicaSetCheckTimeout 设置为 DWORD 的最大值,则出于回滚目的,将用此配置的值对其进行重写。If ReplicaSetCheckTimeout is set to the maximum value of DWORD; then it's overridden with the value of this config for the purposes of rollback. 永远不会重写用于前滚的值。The value used for roll-forward is never overridden.
SkipRollbackUpdateDefaultServiceSkipRollbackUpdateDefaultService Bool,默认值为 falseBool, default is false 动态Dynamic CM 会在应用程序升级回滚过程中跳过恢复更新的默认服务。The CM will skip reverting updated default services during application upgrade rollback.
StandByReplicaKeepDurationStandByReplicaKeepDuration 以秒为单位的时间,默认值为 (3600.0 * 2)Time in seconds, default is (3600.0 * 2) 不允许Not Allowed 指定以秒为单位的时间范围。Specify timespan in seconds. ClusterManager 的 StandByReplicaKeepDuration。The StandByReplicaKeepDuration for ClusterManager.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 7Int, default is 7 不允许Not Allowed ClusterManager 的 TargetReplicaSetSize。The TargetReplicaSetSize for ClusterManager.
UpgradeHealthCheckIntervalUpgradeHealthCheckInterval 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 受监视应用程序升级期间的运行状况检查频率The frequency of health status checks during a monitored application upgrades
UpgradeStatusPollIntervalUpgradeStatusPollInterval 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 轮询应用程序升级状态的频率。The frequency of polling for application upgrade status. 此值确定任何 GetApplicationUpgradeProgress 调用的更新速率This value determines the rate of update for any GetApplicationUpgradeProgress call
CompleteClientRequestCompleteClientRequest Bool,默认值为 falseBool, default is false 动态Dynamic CM 接受即表示客户端请求已完成。Complete client request when accepted by CM.

通用Common

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PerfMonitorIntervalPerfMonitorInterval 以秒为单位的时间,默认值为 1Time in seconds, default is 1 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 性能监视时间间隔。Performance monitoring interval. 设置为 0 或负值,将禁用监视。Setting to 0 or negative value disables monitoring.

DefragmentationEmptyNodeDistributionPolicyDefragmentationEmptyNodeDistributionPolicy

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyIntegerValueMap,默认值为 NoneKeyIntegerValueMap, default is None 动态Dynamic 指定清空节点时遵循的策略碎片整理。Specifies the policy defragmentation follows when emptying nodes. 对于某个给定指标,0 表示 SF 应尝试在 UD 和 FD 间对节点进行均匀的碎片整理,1 仅指表示必须对节点进行碎片整理For a given metric 0 indicates that SF should try to defragment nodes evenly across UDs and FDs; 1 indicates only that the nodes must be defragmented

DefragmentationMetricsDefragmentationMetrics

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyBoolValueMap,默认值为 NoneKeyBoolValueMap, default is None 动态Dynamic 确定应用于碎片整理而不用于负载均衡的指标组。Determines the set of metrics that should be used for defragmentation and not for load balancing.

DefragmentationMetricsPercentOrNumberOfEmptyNodesTriggeringThresholdDefragmentationMetricsPercentOrNumberOfEmptyNodesTriggeringThreshold

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyDoubleValueMap,默认值为 NoneKeyDoubleValueMap, default is None 动态Dynamic 通过在范围 [0.0 - 1.0] 中指定百分比或将空节点数指定为 >= 1.0 的数字,确定需考虑进行群集碎片整理的可用节点数Determines the number of free nodes which are needed to consider cluster defragmented by specifying either percent in range [0.0 - 1.0] or number of empty nodes as number >= 1.0

诊断Diagnostics

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AdminOnlyHttpAuditAdminOnlyHttpAudit Bool,默认值为 trueBool, default is true 动态Dynamic 排除不会影响群集审核状态的 HTTP 请求。Exclude HTTP requests which do not impact the state of the cluster from auditing. 当前,仅排除“GET”类型的请求,但这可能会有变动。Currently; only requests of "GET" type are excluded; but this is subject to change.
AppDiagnosticStoreAccessRequiresImpersonationAppDiagnosticStoreAccessRequiresImpersonation Bool,默认值为 trueBool, default is true 动态Dynamic 代表应用程序访问诊断存储时是否需要模拟。Whether or not impersonation is required when accessing diagnostic stores on behalf of the application.
AppEtwTraceDeletionAgeInDaysAppEtwTraceDeletionAgeInDays Int,默认值为 3Int, default is 3 动态Dynamic 在多少天后删除包含应用程序 ETW 跟踪的旧 ETL 文件。Number of days after which we delete old ETL files containing application ETW traces.
ApplicationLogsFormatVersionApplicationLogsFormatVersion Int,默认值为 0Int, default is 0 动态Dynamic 用于应用程序日志格式的版本。Version for application logs format. 支持的值是 0 和 1.Supported values are 0 and 1. 版本 1 比版本 0 包含更多 ETW 事件记录的字段。Version 1 includes more fields from the ETW event record than version 0.
AuditHttpRequestsAuditHttpRequests Bool,默认值为 falseBool, default is false 动态Dynamic 启用或禁用 HTTP 审核。Turn HTTP auditing on or off. 审核的目的是查看已针对群集执行的活动,包括请求的发起者。The purpose of auditing is to see the activities that have been performed against the cluster; including who initiated the request. 请注意,会尽最大努力进行日志记录,但可能会发生跟踪丢失的情况。Note that this is a best attempt logging; and trace loss may occur. 不记录使用“用户”身份验证的 HTTP 请求。HTTP requests with "User" authentication is not recorded.
CaptureHttpTelemetryCaptureHttpTelemetry Bool,默认值为 trueBool, default is true 动态Dynamic 启用或禁用 HTTP 遥测。Turn HTTP telemetry on or off. 遥测的目的是使 Service Fabric 能够捕获遥测数据来帮助规划未来的工作并确定问题区域。The purpose of telemetry is for Service Fabric to be able to capture telemetry data to help plan future work and identify problem areas. 遥测不会记录任何个人数据和请求正文。Telemetry does not record any personal data or the request body. 遥测会捕获所有 HTTP 请求,除非另行配置。Telemetry captures all HTTP requests unless otherwise configured.
ClusterIdClusterId StringString 动态Dynamic 群集的唯一 ID。The unique id of the cluster. 于群集创建时生成。This is generated when the cluster is created.
ConsumerInstancesConsumerInstances StringString 动态Dynamic DCA 使用者实例列表。The list of DCA consumer instances.
DiskFullSafetySpaceInMBDiskFullSafetySpaceInMB Int,默认值为 1024Int, default is 1024 动态Dynamic 要避免被 DCA 使用的剩余磁盘空间(以 MB 为单位)。Remaining disk space in MB to protect from use by DCA.
EnableCircularTraceSessionEnableCircularTraceSession Bool,默认值为 falseBool, default is false 静态Static 标志指示是否应使用循环跟踪会话。Flag indicates whether circular trace sessions should be used.
EnablePlatformEventsFileSinkEnablePlatformEventsFileSink Bool,默认值为 falseBool, default is false 静态Static 启用/禁用将平台事件写入磁盘的操作Enable/Disable platform events being written to disk
EnableTelemetryEnableTelemetry Bool,默认值为 trueBool, default is true 动态Dynamic 这会启用或禁用遥测。This is going to enable or disable telemetry.
FailuresOnlyHttpTelemetryFailuresOnlyHttpTelemetry Bool,默认值为 falseBool, default is false 动态Dynamic 如果启用了 HTTP 遥测捕获,则仅捕获失败的请求。If HTTP telemetry capture is enabled; capture only failed requests. 这有助于减少针对遥测生成的事件数。This is to help cut down on the number of events generated for telemetry.
HttpTelemetryCapturePercentageHttpTelemetryCapturePercentage int,默认值为 50int, default is 50 动态Dynamic 如果启用了 HTTP 遥测捕获,则仅捕获随机百分比的请求。If HTTP telemetry capture is enabled; capture only a random percentage of requests. 这有助于减少为遥测生成的事件数。This is to help cut down on the number of events generated for telemetry.
MaxDiskQuotaInMBMaxDiskQuotaInMB Int,默认值为 65536Int, default is 65536 动态Dynamic Windows Fabric 日志文件的磁盘配额(以 MB 为单位)。Disk quota in MB for Windows Fabric log files.
ProducerInstancesProducerInstances StringString 动态Dynamic DCA 生成者实例列表。The list of DCA producer instances.

DnsServiceDnsService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
EnablePartitionedQueryEnablePartitionedQuery bool,默认值为 FALSEbool, default is FALSE 静态Static 用于启用对分区服务 DNS 查询的支持的标志。The flag to enable support for DNS queries for partitioned services. 默认情况下,此功能处于关闭状态。The feature is turned off by default. 有关详细信息,请参阅 Service Fabric DNS 服务For more information, see Service Fabric DNS Service.
InstanceCountInstanceCount int,默认值为 -1int, default is -1 静态Static 默认值为 -1,表示 DnsService 在每个节点上运行。Default value is -1 which means that DnsService is running on every node. OneBox 要求将此参数设置为 1,因为 DnsService 使用已知端口 53,因此不能在同一台计算机上存在多个实例。OneBox needs this to be set to 1 since DnsService uses well known port 53, so it cannot have multiple instances on the same machine.
IsEnabledIsEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 启用/禁用 DnsService。Enables/Disables DnsService. 默认已禁用 DnsService,需要设置此配置来启用它。DnsService is disabled by default and this config needs to be set to enable it.
PartitionPrefixPartitionPrefix string,默认值为“--”string, default is "--" 静态Static 控制对分区服务的 DNS 查询中的分区前缀字符串值。Controls the partition prefix string value in DNS queries for partitioned services. 值:The value :
  • 应符合 RFC,因为它将是 DNS 查询的一部分。Should be RFC-compliant as it will be part of a DNS query.
  • 不能包含句点“.”,因为句点会干扰 DNS 后缀行为。Should not contain a dot, '.', as dot interferes with DNS suffix behavior.
  • 长度不能超过 5 个字符。Should not be longer than 5 characters.
  • 不能为空字符串。Cannot be an empty string.
  • 如果重写 PartitionPrefix 设置,则必须重写 PartitionSuffix,反之亦然。If the PartitionPrefix setting is overridden, then PartitionSuffix must be overridden, and vice-versa.
有关详细信息,请参阅 Service Fabric DNS 服务For more information, see Service Fabric DNS Service..
PartitionSuffixPartitionSuffix string,默认值为“”string, default is "" 静态Static 控制对分区服务的 DNS 查询中的分区后缀字符串值。值:Controls the partition suffix string value in DNS queries for partitioned services.The value :
  • 应符合 RFC,因为它将是 DNS 查询的一部分。Should be RFC-compliant as it will be part of a DNS query.
  • 不能包含句点“.”,因为句点会干扰 DNS 后缀行为。Should not contain a dot, '.', as dot interferes with DNS suffix behavior.
  • 长度不能超过 5 个字符。Should not be longer than 5 characters.
  • 如果重写 PartitionPrefix 设置,则必须重写 PartitionSuffix,反之亦然。If the PartitionPrefix setting is overridden, then PartitionSuffix must be overridden, and vice-versa.
有关详细信息,请参阅 Service Fabric DNS 服务For more information, see Service Fabric DNS Service..

EventStoreServiceEventStoreService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
MinReplicaSetSizeMinReplicaSetSize int,默认值为 0int, default is 0 静态Static EventStore 服务的 MinReplicaSetSizeThe MinReplicaSetSize for EventStore service
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 静态Static EventStore 服务的 PlacementConstraintsThe PlacementConstraints for EventStore service
TargetReplicaSetSizeTargetReplicaSetSize int,默认值为 0int, default is 0 静态Static EventStore 服务的 TargetReplicaSetSizeThe TargetReplicaSetSize for EventStore service

FabricClientFabricClient

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ConnectionInitializationTimeoutConnectionInitializationTimeout 以秒为单位的时间,默认值为 2Time in seconds, default is 2 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 每次客户端尝试打开网关连接时的连接超时间隔。Connection timeout interval for each time client tries to open a connection to the gateway.
HealthOperationTimeoutHealthOperationTimeout 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 报告消息发送至运行状况管理器的超时时间。The timeout for a report message sent to Health Manager.
HealthReportRetrySendIntervalHealthReportRetrySendInterval 以秒为单位的时间,默认值为 30,最小值为 1Time in seconds, default is 30, minimum is 1 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 报告组件将累积的运行状况报告重新发送至运行状况管理器的时间间隔。The interval at which the reporting component resends accumulated health reports to Health Manager.
HealthReportSendIntervalHealthReportSendInterval 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 报告组件将累积的运行状况报告发送至运行状况管理器的时间间隔。The interval at which reporting component sends accumulated health reports to Health Manager.
KeepAliveIntervalInSecondsKeepAliveIntervalInSeconds Int,默认值为 20Int, default is 20 静态Static FabricClient 传输向网关发送保持连接消息的时间间隔。The interval at which the FabricClient transport sends keep-alive messages to the gateway. 值为 0,表示禁用 keepAlive。For 0; keepAlive is disabled. 必须是正值。Must be a positive value.
MaxFileSenderThreadsMaxFileSenderThreads Uint,默认值为 10Uint, default is 10 静态Static 并行传输的最大文件数。The max number of files that are transferred in parallel.
NodeAddressesNodeAddresses string,默认值为“”string, default is "" 静态Static 不同节点上的地址(连接字符串)的集合,可用于与命名服务通信。A collection of addresses (connection strings) on different nodes that can be used to communicate with the Naming Service. 最初,客户端随机选择一个地址进行连接。Initially the Client connects selecting one of the addresses randomly. 如果提供了多个连接字符串且因通信或超时错误导致连接失败,客户端按顺序切换为使用下一个地址。If more than one connection string is supplied and a connection fails because of a communication or timeout error; the Client switches to use the next address sequentially. 请参阅命名服务地址重试部分,了解有关重试语义的详细信息。See the Naming Service Address retry section for details on retries semantics.
PartitionLocationCacheLimitPartitionLocationCacheLimit Int,默认值为 100000Int, default is 100000 静态Static 为服务解析所缓存的分区数(设置为 0,表示无限制)。Number of partitions cached for service resolution (set to 0 for no limit).
RetryBackoffIntervalRetryBackoffInterval 以秒为单位的时间,默认值为 3Time in seconds, default is 3 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 重试操作之前的回退时间间隔。The back-off interval before retrying the operation.
ServiceChangePollIntervalServiceChangePollInterval 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 服务的连续轮询之间的间隔从客户端更改为用于注册服务更改通知回调的网关。The interval between consecutive polls for service changes from the client to the gateway for registered service change notifications callbacks.

FabricHostFabricHost

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ActivationMaxFailureCountActivationMaxFailureCount Int,默认值为 10Int, default is 10 动态Dynamic 这是系统在放弃前重试失败的激活的最大计数。This is the maximum count for which system will retry failed activation before giving up.
ActivationMaxRetryIntervalActivationMaxRetryInterval 以秒为单位的时间,默认值为 300Time in seconds, default is 300 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 激活的最大重试时间间隔。Max retry interval for Activation. 每次连续失败后,重试时间间隔的计算结果为 Min(ActivationMaxRetryInterval;连续失败计数 * ActivationRetryBackoffInterval)(即取括号中的最小值)。On every continuous failure the retry interval is calculated as Min( ActivationMaxRetryInterval; Continuous Failure Count * ActivationRetryBackoffInterval).
ActivationRetryBackoffIntervalActivationRetryBackoffInterval 以秒为单位的时间,默认值为 5Time in seconds, default is 5 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 每次激活失败的回退时间间隔;每次连续激活失败后,系统将重试激活最多 MaxActivationFailureCount 次。Backoff interval on every activation failure;On every continuous activation failure the system will retry the activation for up to the MaxActivationFailureCount. 每次尝试的重试间隔是连续激活失败与激活退让间隔的积。The retry interval on every try is a product of continuous activation failure and the activation back-off interval.
EnableRestartManagementEnableRestartManagement Bool,默认值为 falseBool, default is false 动态Dynamic 这会启用服务器重启。This is to enable server restart.
EnableServiceFabricAutomaticUpdatesEnableServiceFabricAutomaticUpdates Bool,默认值为 falseBool, default is false 动态Dynamic 这会通过 Windows 更新启用 Fabric 自动更新。This is to enable fabric automatic update via Windows Update.
EnableServiceFabricBaseUpgradeEnableServiceFabricBaseUpgrade Bool,默认值为 falseBool, default is false 动态Dynamic 这会启用服务器的基本更新。This is to enable base update for server.
FailureReportingExpeditedReportingIntervalEnabledFailureReportingExpeditedReportingIntervalEnabled Bool,默认值为 trueBool, default is true 静态Static 当 FabricHost 处于“故障报告”模式下时,可提高 DCA 中的上传速率。Enables faster uploading rates in DCA when FabricHost is in Failure Reporting mode.
FailureReportingTimeoutFailureReportingTimeout 时间跨度,默认值是 Common::TimeSpan::FromSeconds(60)TimeSpan, default is Common::TimeSpan::FromSeconds(60) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 当 FabricHost 遇到早期启动故障时 DCA 故障报告的超时。Timeout for DCA failure reporting in the case FabricHost encounters an early stage startup failure.
RunDCAOnStartupFailureRunDCAOnStartupFailure Bool,默认值为 trueBool, default is true 静态Static 确定在 FabricHost 中遇到启动问题时是否启动 DCA 来上传日志。Determines whether to launch DCA to upload logs when facing startup issues in FabricHost.
StartTimeoutStartTimeout 以秒为单位的时间,默认值为 300Time in seconds, default is 300 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. fabricactivationmanager 启动的超时时间。Time out for fabricactivationmanager startup.
StopTimeoutStopTimeout 以秒为单位的时间,默认值为 300Time in seconds, default is 300 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 托管服务激活、停用和升级的超时时间。The timeout for hosted service activation; deactivation and upgrade.

FabricNodeFabricNode

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ClientAuthX509FindTypeClientAuthX509FindType string,默认值为“FindByThumbprint”string, default is "FindByThumbprint" 动态Dynamic 指示如何在由 ClientAuthX509StoreName(FindByThumbprint;FindBySubjectName)支持的值指定的存储中搜索证书:。Indicates how to search for certificate in the store specified by ClientAuthX509StoreName Supported value: FindByThumbprint; FindBySubjectName.
ClientAuthX509FindValueClientAuthX509FindValue string,默认值为“”string, default is "" 动态Dynamic 用于查找默认管理员角色 FabricClient 的证书的搜索筛选器值。Search filter value used to locate certificate for default admin role FabricClient.
ClientAuthX509FindValueSecondaryClientAuthX509FindValueSecondary string,默认值为“”string, default is "" 动态Dynamic 用于查找默认管理员角色 FabricClient 的证书的搜索筛选器值。Search filter value used to locate certificate for default admin role FabricClient.
ClientAuthX509StoreNameClientAuthX509StoreName string,默认值为“My”string, default is "My" 动态Dynamic X.509 证书存储的名称,包含默认管理员角色 FabricClient 的证书。Name of the X.509 certificate store that contains certificate for default admin role FabricClient.
ClusterX509FindTypeClusterX509FindType string,默认值为“FindByThumbprint”string, default is "FindByThumbprint" 动态Dynamic 指示如何在由 ClusterX509StoreName 支持的值(“FindByThumbprint”;“FindBySubjectName”)指定的存储中搜索群集证书。使用“FindBySubjectName”时,如果有多个匹配项,使用到期时间最远的那一个。Indicates how to search for cluster certificate in the store specified by ClusterX509StoreName Supported values: "FindByThumbprint"; "FindBySubjectName" With "FindBySubjectName"; when there are multiple matches; the one with the furthest expiration is used.
ClusterX509FindValueClusterX509FindValue string,默认值为“”string, default is "" 动态Dynamic 用于查找群集证书的搜索筛选器值。Search filter value used to locate cluster certificate.
ClusterX509FindValueSecondaryClusterX509FindValueSecondary string,默认值为“”string, default is "" 动态Dynamic 用于查找群集证书的搜索筛选器值。Search filter value used to locate cluster certificate.
ClusterX509StoreNameClusterX509StoreName string,默认值为“My”string, default is "My" 动态Dynamic X.509 证书存储的名称,该存储包含用于保护群集内部通信的群集证书。Name of X.509 certificate store that contains cluster certificate for securing intra-cluster communication.
EndApplicationPortRangeEndApplicationPortRange Int,默认值为 0Int, default is 0 静态Static 由宿主子系统管理的应用程序端口的结束位置(不含)。End (no inclusive) of the application ports managed by hosting subsystem. 当托管中的 EndpointFilteringEnabled 为 true 时为必需。Required if EndpointFilteringEnabled is true in Hosting.
ServerAuthX509FindTypeServerAuthX509FindType string,默认值为“FindByThumbprint”string, default is "FindByThumbprint" 动态Dynamic 指示如何在由 ServerAuthX509StoreName 支持的值(FindByThumbprint;FindBySubjectName)指定的存储中搜索服务器证书。Indicates how to search for server certificate in the store specified by ServerAuthX509StoreName Supported value: FindByThumbprint; FindBySubjectName.
ServerAuthX509FindValueServerAuthX509FindValue string,默认值为“”string, default is "" 动态Dynamic 用于查找服务器证书的搜索筛选器值。Search filter value used to locate server certificate.
ServerAuthX509FindValueSecondaryServerAuthX509FindValueSecondary string,默认值为“”string, default is "" 动态Dynamic 用于查找服务器证书的搜索筛选器值。Search filter value used to locate server certificate.
ServerAuthX509StoreNameServerAuthX509StoreName string,默认值为“My”string, default is "My" 动态Dynamic X.509 证书存储的名称,包含用于准入服务的服务器证书。Name of X.509 certificate store that contains server certificate for entree service.
StartApplicationPortRangeStartApplicationPortRange Int,默认值为 0Int, default is 0 静态Static 由宿主子系统管理的应用程序端口的开始位置。Start of the application ports managed by hosting subsystem. 当托管中的 EndpointFilteringEnabled 为 true 时为必需。Required if EndpointFilteringEnabled is true in Hosting.
StateTraceIntervalStateTraceInterval 以秒为单位的时间,默认值为 300Time in seconds, default is 300 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. FM/FMM 上每个节点及以上节点的节点状态跟踪的时间间隔。The interval for tracing node status on each node and up nodes on FM/FMM.
UserRoleClientX509FindTypeUserRoleClientX509FindType string,默认值为“FindByThumbprint”string, default is "FindByThumbprint" 动态Dynamic 指示如何在由 UserRoleClientX509StoreName 支持的值(FindByThumbprint、FindBySubjectName)指定的存储中搜索证书。Indicates how to search for certificate in the store specified by UserRoleClientX509StoreName Supported value: FindByThumbprint; FindBySubjectName.
UserRoleClientX509FindValueUserRoleClientX509FindValue string,默认值为“”string, default is "" 动态Dynamic 用于查找默认用户角色 FabricClient 的证书的搜索筛选器值。Search filter value used to locate certificate for default user role FabricClient.
UserRoleClientX509FindValueSecondaryUserRoleClientX509FindValueSecondary string,默认值为“”string, default is "" 动态Dynamic 用于查找默认用户角色 FabricClient 的证书的搜索筛选器值。Search filter value used to locate certificate for default user role FabricClient.
UserRoleClientX509StoreNameUserRoleClientX509StoreName string,默认值为“My”string, default is "My" 动态Dynamic X.509 证书存储的名称,包含默认用户角色 FabricClient 的证书。Name of the X.509 certificate store that contains certificate for default user role FabricClient.

FailoverManagerFailoverManager

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AllowNodeStateRemovedForSeedNodeAllowNodeStateRemovedForSeedNode 布尔值,默认为 FALSEBool, default is FALSE 动态Dynamic 指示是否允许删除种子节点的状态的标志Flag to indicate if it's allowed to remove node state for a seed node
BuildReplicaTimeLimitBuildReplicaTimeLimit TimeSpan,默认值为 Common::TimeSpan::FromSeconds(3600)TimeSpan, default is Common::TimeSpan::FromSeconds(3600) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 生成监控状态副本的时间限制。超出该时间范围后会启动警告运行状况报告The time limit for building a stateful replica; after which a warning health report will be initiated
ClusterPauseThresholdClusterPauseThreshold int,默认值为 1int, default is 1 动态Dynamic 如果系统中的节点数低于此值,则会停止放置、负载均衡和故障转移。If the number of nodes in system go below this value then placement; load balancing; and failover is stopped.
CreateInstanceTimeLimitCreateInstanceTimeLimit TimeSpan,默认值为 Common::TimeSpan::FromSeconds(300)TimeSpan, default is Common::TimeSpan::FromSeconds(300) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 创建无状态实例的时间限制。超出该时间范围后会启动警告运行状况报告The time limit for creating a stateless instance; after which a warning health report will be initiated
ExpectedClusterSizeExpectedClusterSize int,默认值为 1int, default is 1 动态Dynamic 最初启动群集时;FM 会等待该数值个数的节点自行报告,然后开始放置其他服务(包括命名等系统服务)。When the cluster is initially started up; the FM will wait for this many nodes to report themselves up before it begins placing other services; including the system services like naming. 增加此值会增加群集启动所需的时间;但可防止早期节点重载,而且随着更多的节点进入联机状态,需要执行额外的移动操作。Increasing this value increases the time it takes a cluster to start up; but prevents the early nodes from becoming overloaded and also the additional moves that will be necessary as more nodes come online. 此值通常应设置为初始群集大小的一小部分。This value should generally be set to some small fraction of the initial cluster size.
ExpectedNodeDeactivationDurationExpectedNodeDeactivationDuration 时间范围,默认值为 Common::TimeSpan::FromSeconds(60.0 * 30)TimeSpan, default is Common::TimeSpan::FromSeconds(60.0 * 30) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是节点完成停用的预期工期。This is the expected duration for a node to complete deactivation in.
ExpectedNodeFabricUpgradeDurationExpectedNodeFabricUpgradeDuration 时间范围,默认值为 Common::TimeSpan::FromSeconds(60.0 * 30)TimeSpan, default is Common::TimeSpan::FromSeconds(60.0 * 30) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是 Windows Fabric 升级过程中升级节点的预期持续时间。This is the expected duration for a node to be upgraded during Windows Fabric upgrade.
ExpectedReplicaUpgradeDurationExpectedReplicaUpgradeDuration 时间范围,默认值为 Common::TimeSpan::FromSeconds(60.0 * 30)TimeSpan, default is Common::TimeSpan::FromSeconds(60.0 * 30) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 这是在应用程序升级过程中节点上要升级的所有副本的预期持续时间。This is the expected duration for all the replicas to be upgraded on a node during application upgrade.
IsSingletonReplicaMoveAllowedDuringUpgradeIsSingletonReplicaMoveAllowedDuringUpgrade bool,默认值为 TRUEbool, default is TRUE 动态Dynamic 如果设置为 true,则允许在升级过程中移动目标副本集大小为 1 的副本。If set to true; replicas with a target replica set size of 1 will be permitted to move during upgrade.
MinReplicaSetSizeMinReplicaSetSize int,默认值为 3int, default is 3 不允许Not Allowed 这是 FM 的最小副本集大小。This is the minimum replica set size for the FM. 如果活动 FM 副本数低于此值,FM 会拒绝对群集进行更改,直到至少恢复最小数目的副本If the number of active FM replicas drops below this value; the FM will reject changes to the cluster until at least the min number of replicas is recovered
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 不允许Not Allowed 故障转移管理器副本的任何放置约束Any placement constraints for the failover manager replicas
PlacementTimeLimitPlacementTimeLimit TimeSpan,默认值为 Common::TimeSpan::FromSeconds(600)TimeSpan, default is Common::TimeSpan::FromSeconds(600) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 达到目标副本计数的时间限制。超出该时间范围后会启动警告运行状况报告The time limit for reaching target replica count; after which a warning health report will be initiated
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是允许分区处于仲裁丢失状态的最长持续时间。This is the max duration for which we allow a partition to be in a state of quorum loss. 如果分区在此持续时间后仍然处于仲裁丢失状态,则通过将不可用副本视为已丢失,使分区从仲裁丢失状态中恢复。If the partition is still in quorum loss after this duration; the partition is recovered from quorum loss by considering the down replicas as lost. 请注意,这可能会导致数据丢失。Note that this can potentially incur data loss.
ReconfigurationTimeLimitReconfigurationTimeLimit TimeSpan,默认值为 Common::TimeSpan::FromSeconds(300)TimeSpan, default is Common::TimeSpan::FromSeconds(300) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 重新配置的时间限制。超出该时间范围后会启动警告运行状况报告The time limit for reconfiguration; after which a warning health report will be initiated
ReplicaRestartWaitDurationReplicaRestartWaitDuration 时间范围,默认值为 Common::TimeSpan::FromSeconds(60.0 * 30)TimeSpan, default is Common::TimeSpan::FromSeconds(60.0 * 30) 不允许Not Allowed 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是 FMService 的 ReplicaRestartWaitDurationThis is the ReplicaRestartWaitDuration for the FMService
SeedNodeQuorumAdditionalBufferNodesSeedNodeQuorumAdditionalBufferNodes int,默认值为 0int, default is 0 动态Dynamic 需启用的种子节点的缓冲区(及种子节点的仲裁)。故障转移管理器 (FM) 应允许最大数量的种子节点 (totalNumSeedNodes - (seedNodeQuorum + SeedNodeQuorumAdditionalBufferNodes)) 处于关闭状态。Buffer of seed nodes that is needed to be up (together with quorum of seed nodes) FM should allow a maximum of (totalNumSeedNodes - (seedNodeQuorum + SeedNodeQuorumAdditionalBufferNodes)) seed nodes to go down.
StandByReplicaKeepDurationStandByReplicaKeepDuration 时间范围,默认为 Common::TimeSpan::FromSeconds(3600.0 * 24 * 7)Timespan, default is Common::TimeSpan::FromSeconds(3600.0 * 24 * 7) 不允许Not Allowed 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是 FMService 的 StandByReplicaKeepDurationThis is the StandByReplicaKeepDuration for the FMService
TargetReplicaSetSizeTargetReplicaSetSize int,默认值为 7int, default is 7 不允许Not Allowed 这是 Windows Fabric 将维护的 FM 副本目标数量。This is the target number of FM replicas that Windows Fabric will maintain. 数字越大,FM 数据的可靠性越高;但会稍微降低其性能。A higher number results in higher reliability of the FM data; with a small performance tradeoff.
UserMaxStandByReplicaCountUserMaxStandByReplicaCount Int,默认值为 1Int, default is 1 动态Dynamic 系统为用户服务保留的默认最大备用副本数。The default max number of StandBy replicas that the system keeps for user services.
UserReplicaRestartWaitDurationUserReplicaRestartWaitDuration 以秒为单位的时间,默认值为 60.0 * 30Time in seconds, default is 60.0 * 30 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 当持久化副本不可用时,Windows Fabric 在创建新的替换副本(需要状态的副本)前,会等待该副本恢复正常,等待时间即为此持续时间。When a persisted replica goes down; Windows Fabric waits for this duration for the replica to come back up before creating new replacement replicas (which would require a copy of the state).
UserStandByReplicaKeepDurationUserStandByReplicaKeepDuration 以秒为单位的时间,默认值为 3600.0 * 24 * 7Time in seconds, default is 3600.0 * 24 * 7 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 持久化副本从不可用状态恢复时,可能已被替换为另一副本。When a persisted replica come back from a down state; it may have already been replaced. 此定时器决定 FM 在放弃备用副本之前保留其多长时间。This timer determines how long the FM will keep the standby replica before discarding it.

FaultAnalysisServiceFaultAnalysisService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
CompletedActionKeepDurationInSecondsCompletedActionKeepDurationInSeconds Int,默认值为 604800Int, default is 604800 静态Static 这是处于终态的操作的大致保留时长。This is approximately how long to keep actions that are in a terminal state. 这也取决于 StoredActionCleanupIntervalInSeconds,因为仅在此间隔时间内执行清理工作。This also depends on StoredActionCleanupIntervalInSeconds; since the work to cleanup is only done on that interval. 604800 秒是 7 天。604800 is 7 days.
DataLossCheckPollIntervalInSecondsDataLossCheckPollIntervalInSeconds int,默认值为 5int, default is 5 静态Static 这是系统在等待发生数据丢失时执行的检查之间的时间间隔。This is the time between the checks the system performs while waiting for data loss to happen. 每次内部迭代检查数据丢失数的次数是 DataLossCheckWaitDurationInSeconds/this。The number of times the data loss number will be checked per internal iteration is DataLossCheckWaitDurationInSeconds/this.
DataLossCheckWaitDurationInSecondsDataLossCheckWaitDurationInSeconds int,默认值为 25int, default is 25 静态Static 系统等待发生数据丢失的总时间量;以秒为单位。The total amount of time; in seconds; that the system will wait for data loss to happen. 这在调用 StartPartitionDataLossAsync() api 时在内部使用。This is internally used when the StartPartitionDataLossAsync() api is called.
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 0Int, default is 0 静态Static FaultAnalysisService 的 MinReplicaSetSize。The MinReplicaSetSize for FaultAnalysisService.
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 静态Static FaultAnalysisService 的 PlacementConstraints。The PlacementConstraints for FaultAnalysisService.
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. FaultAnalysisService 的 QuorumLossWaitDuration。The QuorumLossWaitDuration for FaultAnalysisService.
ReplicaDropWaitDurationInSecondsReplicaDropWaitDurationInSeconds int,默认值为 600int, default is 600 静态Static 调用数据丢失 api 时使用此参数。This parameter is used when the data loss api is called. 该参数控制内部调用删除副本后系统等待副本删除的时间。It controls how long the system will wait for a replica to get dropped after remove replica is internally invoked on it.
ReplicaRestartWaitDurationReplicaRestartWaitDuration 以秒为单位的时间,默认值为 60 分钟Time in seconds, default is 60 minutes 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. FaultAnalysisService 的 ReplicaRestartWaitDuration。The ReplicaRestartWaitDuration for FaultAnalysisService.
StandByReplicaKeepDurationStandByReplicaKeepDuration 以秒为单位的时间,默认值为 (60247) 分钟Time in seconds, default is (60247) minutes 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. FaultAnalysisService 的 StandByReplicaKeepDuration。The StandByReplicaKeepDuration for FaultAnalysisService.
StoredActionCleanupIntervalInSecondsStoredActionCleanupIntervalInSeconds Int,默认值为 3600Int, default is 3600 静态Static 这是清理存储的频率。This is how often the store will be cleaned up. 仅会删除处于终态和至少在 CompletedActionKeepDurationInSeconds 以前完成的操作。Only actions in a terminal state; and that completed at least CompletedActionKeepDurationInSeconds ago will be removed.
StoredChaosEventCleanupIntervalInSecondsStoredChaosEventCleanupIntervalInSeconds Int,默认值为 3600Int, default is 3600 静态Static 这是审核存储(以进行清理)的频率,如果事件数量超过 30000,则开始执行清理。This is how often the store will be audited for cleanup; if the number of events is more than 30000; the cleanup will kick in.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 0Int, default is 0 静态Static NOT_PLATFORM_UNIX_START,FaultAnalysisService 的 TargetReplicaSetSize。NOT_PLATFORM_UNIX_START The TargetReplicaSetSize for FaultAnalysisService.

联合Federation

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
LeaseDurationLeaseDuration 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 节点与其邻居之间的租约持续时间。Duration that a lease lasts between a node and its neighbors.
LeaseDurationAcrossFaultDomainLeaseDurationAcrossFaultDomain 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 所有容错域中的节点与其邻居之间的租约持续时间。Duration that a lease lasts between a node and its neighbors across fault domains.

FileStoreServiceFileStoreService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AcceptChunkUploadAcceptChunkUpload Bool,默认值为 TRUEBool, default is TRUE 动态Dynamic 用于确定文件存储服务在复制应用程序包期间是否接受基于区块的文件上传的配置。Config to determine whether file store service accepts chunk based file upload or not during copy application package.
AnonymousAccessEnabledAnonymousAccessEnabled Bool,默认值为 trueBool, default is true 静态Static 启用/禁用对 FileStoreService 共享的匿名访问。Enable/Disable anonymous access to the FileStoreService shares.
CommonName1Ntlmx509CommonNameCommonName1Ntlmx509CommonName string,默认值为“”string, default is "" 静态Static 使用 NTLM 身份验证时,用于在 CommonName1NtlmPasswordSecret 上生成 HMAC 的 X509 证书的公用名称The common name of the X509 certificate used to generate HMAC on the CommonName1NtlmPasswordSecret when using NTLM authentication
CommonName1Ntlmx509StoreLocationCommonName1Ntlmx509StoreLocation string,默认值为“LocalMachine”string, default is "LocalMachine" 静态Static 使用 NTLM 身份验证时,用于在 CommonName1NtlmPasswordSecret 上生成 HMAC 的 X509 证书的存储位置The store location of the X509 certificate used to generate HMAC on the CommonName1NtlmPasswordSecret when using NTLM authentication
CommonName1Ntlmx509StoreNameCommonName1Ntlmx509StoreName string,默认值为“MY”string, default is "MY" 静态Static 使用 NTLM 身份验证时,用于在 CommonName1NtlmPasswordSecret 上生成 HMAC 的 X509 证书的存储名称The store name of the X509 certificate used to generate HMAC on the CommonName1NtlmPasswordSecret when using NTLM authentication
CommonName2Ntlmx509CommonNameCommonName2Ntlmx509CommonName string,默认值为“”string, default is "" 静态Static 使用 NTLM 身份验证时,用于在 CommonName2NtlmPasswordSecret 上生成 HMAC 的 X509 证书的公用名称The common name of the X509 certificate used to generate HMAC on the CommonName2NtlmPasswordSecret when using NTLM authentication
CommonName2Ntlmx509StoreLocationCommonName2Ntlmx509StoreLocation string,默认值为“LocalMachine”string, default is "LocalMachine" 静态Static 使用 NTLM 身份验证时,用于在 CommonName2NtlmPasswordSecret 上生成 HMAC 的 X509 证书的存储位置The store location of the X509 certificate used to generate HMAC on the CommonName2NtlmPasswordSecret when using NTLM authentication
CommonName2Ntlmx509StoreNameCommonName2Ntlmx509StoreName string,默认值为“MY”string, default is "MY" 静态Static 使用 NTLM 身份验证时,用于在 CommonName2NtlmPasswordSecret 上生成 HMAC 的 X509 证书的存储名称The store name of the X509 certificate used to generate HMAC on the CommonName2NtlmPasswordSecret when using NTLM authentication
CommonNameNtlmPasswordSecretCommonNameNtlmPasswordSecret SecureString,默认值为 Common::SecureString("")SecureString, default is Common::SecureString("") 静态Static 密码,用于在使用 NTLM 身份验证时用作种子以生成相同密码The password secret which used as seed to generated same password when using NTLM authentication
DiskSpaceHealthReportingIntervalWhenCloseToOutOfDiskSpaceDiskSpaceHealthReportingIntervalWhenCloseToOutOfDiskSpace 时间范围,默认值为 Common::TimeSpan::FromMinutes(5)TimeSpan, default is Common::TimeSpan::FromMinutes(5) 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 当磁盘空间即将耗尽时,检查磁盘空间来报告运行状况事件之间的时间间隔。The time interval between checking of disk space for reporting health event when disk is close to out of space.
DiskSpaceHealthReportingIntervalWhenEnoughDiskSpaceDiskSpaceHealthReportingIntervalWhenEnoughDiskSpace TimeSpan,默认值为 Common::TimeSpan::FromMinutes(15)TimeSpan, default is Common::TimeSpan::FromMinutes(15) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 当磁盘空间充足时,检查磁盘空间来报告运行状况事件之间的时间间隔。The time interval between checking of disk space for reporting health event when there is enough space on disk.
EnableImageStoreHealthReportingEnableImageStoreHealthReporting bool,默认值为 TRUEbool, default is TRUE 静态Static 用于确定文件存储服务是否应报告其运行状况的配置。Config to determine whether file store service should report its health.
FreeDiskSpaceNotificationSizeInKBFreeDiskSpaceNotificationSizeInKB int64,默认值为 25*1024int64, default is 25*1024 动态Dynamic 可用磁盘空间的大小,低于此大小可能会出现运行状况警告。The size of free disk space below which health warning may occur. 此配置的最小值和 FreeDiskSpaceNotificationThresholdPercentage 配置用于确定是否发送运行状况警告。Minimum value of this config and FreeDiskSpaceNotificationThresholdPercentage config are used to determine sending of the health warning.
FreeDiskSpaceNotificationThresholdPercentageFreeDiskSpaceNotificationThresholdPercentage double,默认值为 0.02double, default is 0.02 动态Dynamic 可用磁盘空间的百分比,低于此值可能会出现运行状况警告。The percentage of free disk space below which health warning may occur. 此配置的最小值和 FreeDiskSpaceNotificationInMB 配置用于确定是否发送运行状况警告。Minimum value of this config and FreeDiskSpaceNotificationInMB config are used to determine sending of health warning.
GenerateV1CommonNameAccountGenerateV1CommonNameAccount bool,默认值为 TRUEbool, default is TRUE 静态Static 指定是否要使用用户名 V1 生成算法生成帐户。Specifies whether to generate an account with user name V1 generation algorithm. 从 Service Fabric 6.1 版开始,始终创建具有 v2 生成的帐户。Starting with Service Fabric version 6.1; an account with v2 generation is always created. 从/到不支持 V2 生成的版本升级需要 V1 帐户(6.1 版以前)。The V1 account is necessary for upgrades from/to versions that do not support V2 generation (prior to 6.1).
MaxCopyOperationThreadsMaxCopyOperationThreads Uint,默认值为 0Uint, default is 0 动态Dynamic 辅助节点可从主节点复制的最大并行文件数。The maximum number of parallel files that secondary can copy from primary. '0' == 核心数。'0' == number of cores.
MaxFileOperationThreadsMaxFileOperationThreads Uint,默认值为 100Uint, default is 100 静态Static 可在主节点中执行 FileOperations(复制/移动)操作的最大并行线程数。The maximum number of parallel threads allowed to perform FileOperations (Copy/Move) in the primary. '0' == 核心数。'0' == number of cores.
MaxRequestProcessingThreadsMaxRequestProcessingThreads Uint,默认值为 200Uint, default is 200 静态Static 可在主节点中处理请求的最大并行线程数。The maximum number of parallel threads allowed to process requests in the primary. '0' == 核心数。'0' == number of cores.
MaxSecondaryFileCopyFailureThresholdMaxSecondaryFileCopyFailureThreshold Uint,默认值为 25Uint, default is 25 动态Dynamic 放弃前,辅助节点上的最大文件副本重试次数。The maximum number of file copy retries on the secondary before giving up.
MaxStoreOperationsMaxStoreOperations Uint,默认值为 4096Uint, default is 4096 静态Static 主节点上可允许的最大并行存储事务操作数。The maximum number of parallel store transaction operations allowed on primary. ‘0’== 内核数。'0' == number of cores.
NamingOperationTimeoutNamingOperationTimeout 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 执行命名操作时的超时时间。The timeout for performing naming operation.
PrimaryAccountNTLMPasswordSecretPrimaryAccountNTLMPasswordSecret SecureString,默认值为空SecureString, default is empty 静态Static 密码,用于在使用 NTLM 身份验证时用作种子以生成相同密码。The password secret which used as seed to generated same password when using NTLM authentication.
PrimaryAccountNTLMX509StoreLocationPrimaryAccountNTLMX509StoreLocation string,默认值为“LocalMachine”string, default is "LocalMachine" 静态Static 使用 NTLM 身份验证时,用于在 PrimaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的存储位置。The store location of the X509 certificate used to generate HMAC on the PrimaryAccountNTLMPasswordSecret when using NTLM authentication.
PrimaryAccountNTLMX509StoreNamePrimaryAccountNTLMX509StoreName string,默认值为“MY”string, default is "MY" 静态Static 使用 NTLM 身份验证时,用于在 PrimaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的存储名称。The store name of the X509 certificate used to generate HMAC on the PrimaryAccountNTLMPasswordSecret when using NTLM authentication.
PrimaryAccountNTLMX509ThumbprintPrimaryAccountNTLMX509Thumbprint string,默认值为“”string, default is "" 静态Static 使用 NTLM 身份验证时,用于在 PrimaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的指纹。The thumbprint of the X509 certificate used to generate HMAC on the PrimaryAccountNTLMPasswordSecret when using NTLM authentication.
PrimaryAccountTypePrimaryAccountType string,默认值为“”string, default is "" 静态Static FileStoreService 共享的 ACL 主体的主帐户类型。The primary AccountType of the principal to ACL the FileStoreService shares.
PrimaryAccountUserNamePrimaryAccountUserName string,默认值为“”string, default is "" 静态Static FileStoreService 共享的 ACL 主体的主帐户用户名。The primary account Username of the principal to ACL the FileStoreService shares.
PrimaryAccountUserPasswordPrimaryAccountUserPassword SecureString,默认值为空SecureString, default is empty 静态Static FileStoreService 共享的 ACL 主体的主帐户密码。The primary account password of the principal to ACL the FileStoreService shares.
QueryOperationTimeoutQueryOperationTimeout 以秒为单位的时间,默认值为 60Time in seconds, default is 60 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 执行查询操作的超时时间。The timeout for performing query operation.
SecondaryAccountNTLMPasswordSecretSecondaryAccountNTLMPasswordSecret SecureString,默认值为空SecureString, default is empty 静态Static 密码,用于在使用 NTLM 身份验证时用作种子以生成相同密码。The password secret which used as seed to generated same password when using NTLM authentication.
SecondaryAccountNTLMX509StoreLocationSecondaryAccountNTLMX509StoreLocation string,默认值为“LocalMachine”string, default is "LocalMachine" 静态Static 使用 NTLM 身份验证时,用于在 SecondaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的存储位置。The store location of the X509 certificate used to generate HMAC on the SecondaryAccountNTLMPasswordSecret when using NTLM authentication.
SecondaryAccountNTLMX509StoreNameSecondaryAccountNTLMX509StoreName string,默认值为“MY”string, default is "MY" 静态Static 使用 NTLM 身份验证时,用于在 SecondaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的存储名称。The store name of the X509 certificate used to generate HMAC on the SecondaryAccountNTLMPasswordSecret when using NTLM authentication.
SecondaryAccountNTLMX509ThumbprintSecondaryAccountNTLMX509Thumbprint string,默认值为“”string, default is "" 静态Static 使用 NTLM 身份验证时,用于在 SecondaryAccountNTLMPasswordSecret 上生成 HMAC 的 X509 证书的指纹。The thumbprint of the X509 certificate used to generate HMAC on the SecondaryAccountNTLMPasswordSecret when using NTLM authentication.
SecondaryAccountTypeSecondaryAccountType string,默认值为“”string, default is "" 静态Static FileStoreService 共享的 ACL 主体的辅助帐户类型。The secondary AccountType of the principal to ACL the FileStoreService shares.
SecondaryAccountUserNameSecondaryAccountUserName string,默认值为“”string, default is "" 静态Static FileStoreService 共享的 ACL 主体的辅助帐户用户名。The secondary account Username of the principal to ACL the FileStoreService shares.
SecondaryAccountUserPasswordSecondaryAccountUserPassword SecureString,默认值为空SecureString, default is empty 静态Static FileStoreService 共享的 ACL 主体的辅助帐户密码。The secondary account password of the principal to ACL the FileStoreService shares.
SecondaryFileCopyRetryDelayMillisecondsSecondaryFileCopyRetryDelayMilliseconds uint,默认值为 500uint, default is 500 动态Dynamic 文件复制重试延迟(以毫秒为单位)。The file copy retry delay (in milliseconds).
UseChunkContentInTransportMessageUseChunkContentInTransportMessage bool,默认值为 TRUEbool, default is TRUE 动态Dynamic 一个标志,表示使用在 v6.4 中引入的新版上传协议。The flag for using the new version of the upload protocol introduced in v6.4. 此协议版本使用 Service Fabric 传输将文件上传到映像存储,其性能高于以前版本中使用的 SMB 协议。This protocol version uses service fabric transport to upload files to image store which provides better performance than SMB protocol used in previous versions.

HealthManagerHealthManager

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
EnableApplicationTypeHealthEvaluationEnableApplicationTypeHealthEvaluation Bool,默认值为 falseBool, default is false 静态Static 群集运行状况评估策略:启用按应用程序类型的运行状况评估。Cluster health evaluation policy: enable per application type health evaluation.
MaxSuggestedNumberOfEntityHealthReportsMaxSuggestedNumberOfEntityHealthReports Int,默认值为 100Int, default is 100 动态Dynamic 在引发关于监视程序运行状况报告逻辑的担忧之前,实体可具有的运行状况报告数量上限。The maximum number of health reports that an entity can have before raising concerns about the watchdog's health reporting logic. 每个运行状况实体都应具有相对较少的运行状况报告。Each health entity is supposed to have a relatively small number of health reports. 如果报告计数超过此数字,监视程序的实现可能会出现问题。If the report count goes above this number; there may be issues with the watchdog's implementation. 计算实体时,报告数太多的实体通过“警告”运行状况报告进行标记。An entity with too many reports is flagged through a Warning health report when the entity is evaluated.

HealthManager/ClusterHealthPolicyHealthManager/ClusterHealthPolicy

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ConsiderWarningAsErrorConsiderWarningAsError Bool,默认值为 falseBool, default is false 静态Static 群集运行状况评估策略:警告视为错误。Cluster health evaluation policy: warnings are treated as errors.
MaxPercentUnhealthyApplicationsMaxPercentUnhealthyApplications Int,默认值为 0Int, default is 0 静态Static 群集运行状况评估策略:为确保群集运行正常,可允许的最高运行不正常应用程序百分比。Cluster health evaluation policy: maximum percent of unhealthy applications allowed for the cluster to be healthy.
MaxPercentUnhealthyNodesMaxPercentUnhealthyNodes Int,默认值为 0Int, default is 0 静态Static 群集运行状况评估策略:为确保群集运行正常,可允许的最高运行不正常节点百分比。Cluster health evaluation policy: maximum percent of unhealthy nodes allowed for the cluster to be healthy.

HealthManager/ClusterUpgradeHealthPolicyHealthManager/ClusterUpgradeHealthPolicy

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
MaxPercentDeltaUnhealthyNodesMaxPercentDeltaUnhealthyNodes int,默认值为 10int, default is 10 静态Static 群集升级运行状况评估策略:为确保群集运行正常,可允许的最高增量不正常节点百分比Cluster upgrade health evaluation policy: maximum percent of delta unhealthy nodes allowed for the cluster to be healthy
MaxPercentUpgradeDomainDeltaUnhealthyNodesMaxPercentUpgradeDomainDeltaUnhealthyNodes int,默认值为 15int, default is 15 静态Static 群集升级运行状况评估策略:为确保群集运行正常,升级域中可允许的最高增量不正常节点百分比Cluster upgrade health evaluation policy: maximum percent of delta of unhealthy nodes in an upgrade domain allowed for the cluster to be healthy

HostingHosting

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ActivationMaxFailureCountActivationMaxFailureCount 整数,默认值为 10Whole number, default is 10 动态Dynamic 系统在放弃之前重试失败激活的次数Number of times system retries failed activation before giving up
ActivationMaxRetryIntervalActivationMaxRetryInterval 以秒为单位的时间,默认值为 300Time in seconds, default is 300 动态Dynamic 在每次连续激活失败时,系统将重试激活最多 ActivationMaxFailureCount 次。On every continuous activation failure, the system retries the activation for up to ActivationMaxFailureCount. ActivationMaxRetryInterval 指定每次激活失败之后、重试之前等待的时间间隔ActivationMaxRetryInterval specifies Wait time interval before retry after every activation failure
ActivationRetryBackoffIntervalActivationRetryBackoffInterval 以秒为单位的时间,默认值为 5Time in Seconds, default is 5 动态Dynamic 每次激活失败的退让间隔;在每次连续激活失败后,系统将重试激活最多 MaxActivationFailureCount 次。Backoff interval on every activation failure; On every continuous activation failure, the system retries the activation for up to the MaxActivationFailureCount. 每次尝试的重试间隔是连续激活失败与激活退让间隔的积。The retry interval on every try is a product of continuous activation failure and the activation back-off interval.
ActivationTimeoutActivationTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(180)TimeSpan, default is Common::TimeSpan::FromSeconds(180) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 应用程序激活、停用和升级的超时时间。The timeout for application activation; deactivation and upgrade.
ApplicationHostCloseTimeoutApplicationHostCloseTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在自助激活进程中检测到 Fabric 退出时;FabricRuntime 会关闭用户主机 (applicationhost) 进程中的所有副本。When Fabric exit is detected in a self activated processes; FabricRuntime closes all of the replicas in the user's host (applicationhost) process. 这是关闭操作的超时时间。This is the timeout for the close operation.
CnsNetworkPluginCnmUrlPortCnsNetworkPluginCnmUrlPort wstring,默认值为 L"48080"wstring, default is L"48080" 静态Static Azure cnm api url 端口Azure cnm api url port
CnsNetworkPluginCnsUrlPortCnsNetworkPluginCnsUrlPort wstring,默认值为 L"10090"wstring, default is L"10090" 静态Static Azure cns url 端口Azure cns url port
ContainerServiceArgumentsContainerServiceArguments string,默认为“-H localhost:2375 -H npipe://”string, default is "-H localhost:2375 -H npipe://" 静态Static Service Fabric (SF) 管理 docker 守护程序(在 Win10 等 windows 客户端计算机上除外)。Service Fabric (SF) manages docker daemon (except on windows client machines like Win10). 此配置允许用户指定启动时应传递到 Docker 守护程序的自定义参数。This configuration allows user to specify custom arguments that should be passed to docker daemon when starting it. 指定自定义参数时,Service Fabric 不会将“--pidfile”参数以外的任何其他参数传递给 Docker 引擎。When custom arguments are specified, Service Fabric do not pass any other argument to Docker engine except '--pidfile' argument. 因此,用户不应指定“--pidfile”参数作为其自定义参数的一部分。Hence users should not specify '--pidfile' argument as part of their customer arguments. 此外,自定义参数应确保 Docker 守护程序侦听 Windows 上的默认名称管道(或 Linux 上的 Unix 域套接字),以便 Service Fabric 可以与其通信。Also, the custom arguments should ensure that docker daemon listens on default name pipe on Windows (or Unix domain socket on Linux) for Service Fabric to be able to communicate with it.
ContainerServiceLogFileMaxSizeInKbContainerServiceLogFileMaxSizeInKb int,默认值为 32768int, default is 32768 静态Static docker 容器生成的日志文件的最大文件大小。Maximum file size of log file generated by docker containers. 仅限 Windows。Windows only.
ContainerImageDownloadTimeoutContainerImageDownloadTimeout int,秒数,默认为 1200(20 分钟)int, number of seconds, default is 1200 (20 mins) 动态Dynamic 下载映像操作超时需要经历的秒数。Number of seconds before download of image times out.
ContainerImagesToSkipContainerImagesToSkip 字符串,以竖线字符分隔的映像名称,默认值为 ""string, image names separated by vertical line character, default is "" 静态Static 不应删除的一个或多个容器映像的名称。Name of one or more container images that should not be deleted. 与 PruneContainerImages 参数一起使用。Used with the PruneContainerImages parameter.
ContainerServiceLogFileNamePrefixContainerServiceLogFileNamePrefix string,默认值是“sfcontainerlogs”string, default is "sfcontainerlogs" 静态Static docker 容器生成的日志文件的文件名前缀。File name prefix for log files generated by docker containers. 仅限 Windows。Windows only.
ContainerServiceLogFileRetentionCountContainerServiceLogFileRetentionCount int,默认值为 10int, default is 10 静态Static 在覆盖日志文件之前由 docker 容器生成的日志文件数。Number of log files generated by docker containers before log files are overwritten. 仅限 Windows。Windows only.
CreateFabricRuntimeTimeoutCreateFabricRuntimeTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 同步 FabricCreateRuntime 调用的超时时间The timeout value for the sync FabricCreateRuntime call
DefaultContainerRepositoryAccountNameDefaultContainerRepositoryAccountName string,默认值为“”string, default is "" 静态Static 使用默认凭据,而不是 ApplicationManifest.xml 中指定的凭据Default credentials used instead of credentials specified in ApplicationManifest.xml
DefaultContainerRepositoryPasswordDefaultContainerRepositoryPassword string,默认值为“”string, default is "" 静态Static 使用默认密码凭据,而不是 ApplicationManifest.xml 中指定的凭据Default password credentials used instead of credentials specified in ApplicationManifest.xml
DefaultContainerRepositoryPasswordTypeDefaultContainerRepositoryPasswordType string,默认值为“”string, default is "" 静态Static 不是空字符串时,此值可以是“Encrypted”或“SecretsStoreRef”。When not empty string, the value can be "Encrypted" or "SecretsStoreRef".
DefaultDnsSearchSuffixEmptyDefaultDnsSearchSuffixEmpty bool,默认值为 FALSEbool, default is FALSE 静态Static 默认情况下,服务名称会追加到容器服务的 SF DNS 名称后面。By default the service name is appended to the SF DNS name for container services. 此功能会停止此行为,以便默认情况下,解析路径中 SF DNS 名称的后面不追加任何内容。This feature stops this behavior so that nothing is appended to the SF DNS name by default in the resolution pathway.
DeploymentMaxFailureCountDeploymentMaxFailureCount int,默认值为 20int, default is 20 动态Dynamic 重试 DeploymentMaxFailureCount 次应用程序部署后,节点上该应用程序的部署才会失败。Application deployment will be retried for DeploymentMaxFailureCount times before failing the deployment of that application on the node.
DeploymentMaxRetryIntervalDeploymentMaxRetryInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(3600)TimeSpan, default is Common::TimeSpan::FromSeconds(3600) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 部署的最大重试时间间隔。Max retry interval for the deployment. 每次连续失败后,重试时间间隔的计算结果为 Min(DeploymentMaxRetryInterval;连续失败计数 * DeploymentRetryBackoffInterval)(即取括号中的最小值)On every continuous failure the retry interval is calculated as Min( DeploymentMaxRetryInterval; Continuous Failure Count * DeploymentRetryBackoffInterval)
DeploymentRetryBackoffIntervalDeploymentRetryBackoffInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(10)TimeSpan, default is Common::TimeSpan::FromSeconds(10) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 部署失败的回退时间间隔。Back-off interval for the deployment failure. 每次连续部署失败时,系统重试部署的次数会多达 MaxDeploymentFailureCount 次。On every continuous deployment failure the system will retry the deployment for up to the MaxDeploymentFailureCount. 重试时间间隔是连续部署失败的产物,为部署回退时间间隔。The retry interval is a product of continuous deployment failure and the deployment backoff interval.
DisableContainersDisableContainers bool,默认值为 FALSEbool, default is FALSE 静态Static 用于禁用容器的配置 - 使用此项而不是 DisableContainerServiceStartOnContainerActivatorOpen,后者是已弃用的配置Config for disabling containers - used instead of DisableContainerServiceStartOnContainerActivatorOpen which is deprecated config
DisableDockerRequestRetryDisableDockerRequestRetry bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 默认情况下,SF 与 DD(docker 守护程序)进行通信,对于发送到它的每个 http 请求,超时都是“DockerRequestTimeout”。By default SF communicates with DD (docker dameon) with a timeout of 'DockerRequestTimeout' for each http request sent to it. 如果 DD 在此时间段内没有响应,并且顶级操作仍然有剩余时间,则 SF 会重新发送请求。If DD does not responds within this time period; SF resends the request if top level operation still has remaining time. 对于 hyperv 容器,DD 有时候需要花费更多时间才能激活容器或停用容器。With hyperv container; DD sometimes take much more time to bring up the container or deactivate it. 在这种情况下,从 SF 的角度来看,DD 请求超时并且 SF 将重试操作。In such cases DD request times out from SF perspective and SF retries the operation. 有时,这好像给 DD 增加了更多压力。Sometimes this seems to adds more pressure on DD. 此配置允许禁用此重试并等待 DD 做出响应。This config allows to disable this retry and wait for DD to respond.
DnsServerListTwoIpsDnsServerListTwoIps 布尔值,默认为 FALSEBool, default is FALSE 静态Static 此标志会添加本地 DNS 服务器两次,以帮助缓解间歇性解析问题。This flags adds the local dns server twice to help alleviate intermittent resolve issues.
DoNotInjectLocalDnsServerDoNotInjectLocalDnsServer bool,默认值为 FALSEbool, default is FALSE 静态Static 阻止运行时注入本地 IP 来用作容器的 DNS 服务器。Prevents the runtime to injecting the local IP as DNS server for containers.
EnableActivateNoWindowEnableActivateNoWindow bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 激活进程是在不使用任何控制台的情况下在后台中创建的。The activated process is created in the background without any console.
EnableContainerServiceDebugModeEnableContainerServiceDebugMode bool,默认值为 TRUEbool, default is TRUE 静态Static 为 docker 容器启用/禁用日志记录。Enable/disable logging for docker containers. 仅限 Windows。Windows only.
EnableDockerHealthCheckIntegrationEnableDockerHealthCheckIntegration bool,默认值为 TRUEbool, default is TRUE 静态Static 实现 docker HEALTHCHECK 事件与 Service Fabric 系统运行状况报告的集成Enables integration of docker HEALTHCHECK events with Service Fabric system health report
EnableProcessDebuggingEnableProcessDebugging bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 在调试器下启用启动应用程序主机Enables launching application hosts under debugger
EndpointProviderEnabledEndpointProviderEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 启用 Fabric 终结点资源管理。Enables management of Endpoint resources by Fabric. 需要指定 FabricNode 中开始和结尾应用程序端口范围。Requires specification of start and end application port range in FabricNode.
FabricContainerAppsEnabledFabricContainerAppsEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static
FirewallPolicyEnabledFirewallPolicyEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 允许为具有 ServiceManifest 中指定的显式端口的终结点资源打开防火墙端口Enables opening firewall ports for Endpoint resources with explicit ports specified in ServiceManifest
GetCodePackageActivationContextTimeoutGetCodePackageActivationContextTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. CodePackageActivationContext 调用的超时时间。The timeout value for the CodePackageActivationContext calls. 这不适用于临时服务。This is not applicable to ad hoc services.
GovernOnlyMainMemoryForProcessesGovernOnlyMainMemoryForProcesses bool,默认值为 FALSEbool, default is FALSE 静态Static 资源治理的默认行为是对进程使用的总内存量(RAM + 交换)施加限制(在 MemoryInMB 中指定)。Default behavior of Resource Governance is to put limit specified in MemoryInMB on amount of total memory (RAM + swap) that process uses. 如果超出限制,进程会收到 OutOfMemory 异常。If the limit is exceeded; the process will receive OutOfMemory exception. 如果此参数设置为 true,则限制只会应用到进程将使用的 RAM 内存量。If this parameter is set to true; limit will be applied only to the amount of RAM memory that a process will use. 如果超出该限制,且此设置为 true,则 OS 会将主内存切换到磁盘。If this limit is exceeded; and if this setting is true; then OS will swap the main memory to disk.
IPProviderEnabledIPProviderEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 启用 IP 地址的管理。Enables management of IP addresses.
IsDefaultContainerRepositoryPasswordEncryptedIsDefaultContainerRepositoryPasswordEncrypted bool,默认值为 FALSEbool, default is FALSE 静态Static DefaultContainerRepositoryPassword 是否已加密。Whether the DefaultContainerRepositoryPassword is encrypted or not.
LinuxExternalExecutablePathLinuxExternalExecutablePath string,默认值为“/usr/bin/”string, default is "/usr/bin/" 静态Static 节点上外部可执行命令的主目录。The primary directory of external executable commands on the node.
NTLMAuthenticationEnabledNTLMAuthenticationEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 启用对以其他用户身份运行的代码包使用 NTLM 的支持,以跨计算机的进程进行安全通信。Enables support for using NTLM by the code packages that are running as other users so that the processes across machines can communicate securely.
NTLMAuthenticationPasswordSecretNTLMAuthenticationPasswordSecret SecureString,默认值为 Common::SecureString("")SecureString, default is Common::SecureString("") 静态Static 用于生成 NTLM 用户的密码的加密方式。Is an encrypted has that is used to generate the password for NTLM users. 如果 NTLMAuthenticationEnabled 为 true,则必须设置。Has to be set if NTLMAuthenticationEnabled is true. 由部署器进行验证。Validated by the deployer.
NTLMSecurityUsersByX509CommonNamesRefreshIntervalNTLMSecurityUsersByX509CommonNamesRefreshInterval TimeSpan,默认值为 Common::TimeSpan::FromMinutes(3)TimeSpan, default is Common::TimeSpan::FromMinutes(3) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 特定于环境的设置,主机以此定期时间间隔进行扫描,查找用于 FileStoreService NTLM 配置的新证书。Environment-specific settings The periodic interval at which Hosting scans for new certificates to be used for FileStoreService NTLM configuration.
NTLMSecurityUsersByX509CommonNamesRefreshTimeoutNTLMSecurityUsersByX509CommonNamesRefreshTimeout TimeSpan,默认值为 Common::TimeSpan::FromMinutes(4)TimeSpan, default is Common::TimeSpan::FromMinutes(4) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 使用证书公用名称配置 NTLM 用户的超时时间。The timeout for configuring NTLM users using certificate common names. FileStoreService 共享需要 NTLM 用户。The NTLM users are needed for FileStoreService shares.
PruneContainerImagesPruneContainerImages bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 从节点中删除未使用的应用程序容器映像。Remove unused application container images from nodes. 当从 Service Fabric 群集中注销某个 ApplicationType 时,此应用程序使用的容器映像将从 Service Fabric 将其下载到的节点上删除。When an ApplicationType is unregistered from the Service Fabric cluster, the container images that were used by this application will be removed on nodes where it was downloaded by Service Fabric. 此修剪每小时运行一次,因此,从群集中删除映像可能需要花费长达一小时(加上修剪映像的时间)。The pruning runs every hour, so it may take up to one hour (plus time to prune the image) for images to be removed from the cluster.
Service Fabric 从不下载或删除与应用程序无关的映像。Service Fabric will never download or remove images not related to an application. 必须显式删除已手动下载的或通过其他方式下载的无关映像。Unrelated images that were downloaded manually or otherwise must be removed explicitly.
可以在 ContainerImagesToSkip 参数中指定不应删除的映像。Images that should not be deleted can be specified in the ContainerImagesToSkip parameter.
RegisterCodePackageHostTimeoutRegisterCodePackageHostTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. FabricRegisterCodePackageHost 同步调用的超时时间值。The timeout value for the FabricRegisterCodePackageHost sync call. 这仅适用于多代码包应用程序主机,如 FWPThis is applicable for only multi code package application hosts like FWP
RequestTimeoutRequestTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(30)TimeSpan, default is Common::TimeSpan::FromSeconds(30) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 这表示用户的应用程序主机与各种托管的相关操作(如中心注册、运行时注册)的 Fabric 进程之间的通信超时时间。This represents the timeout for communication between the user's application host and Fabric process for various hosting related operations such as factory registration; runtime registration.
RunAsPolicyEnabledRunAsPolicyEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 允许以运行结构进程的用户以外的本地用户身份运行代码包。Enables running code packages as local user other than the user under which fabric process is running. 为启用此策略,必须以 SYSTEM 或具有 SeAssignPrimaryTokenPrivilege 的用户的身份运行 Fabric。In order to enable this policy Fabric must be running as SYSTEM or as user who has SeAssignPrimaryTokenPrivilege.
ServiceFactoryRegistrationTimeoutServiceFactoryRegistrationTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 同步 Register(Stateless/Stateful)ServiceFactory 调用的超时时间The timeout value for the sync Register(Stateless/Stateful)ServiceFactory call
ServiceTypeDisableFailureThresholdServiceTypeDisableFailureThreshold 整数,默认值为 1Whole number, default is 1 动态Dynamic 这是失败计数的阈值,超过此值后,将通知 FailoverManager (FM) 禁用该节点上的服务类型,并尝试在另一个节点上放置对象。This is the threshold for the failure count after which FailoverManager (FM) is notified to disable the service type on that node and try a different node for placement.
ServiceTypeDisableGraceIntervalServiceTypeDisableGraceInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(30)TimeSpan, default is Common::TimeSpan::FromSeconds(30) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 时间间隔,超过该时间间隔后禁用服务类型Time interval after which the service type can be disabled
ServiceTypeRegistrationTimeoutServiceTypeRegistrationTimeout 以秒为单位的时间,默认值为 300Time in Seconds, default is 300 动态Dynamic 允许在结构中注册 ServiceType 的最长时间Maximum time allowed for the ServiceType to be registered with fabric
UseContainerServiceArgumentsUseContainerServiceArguments bool,默认值为 TRUEbool, default is TRUE 静态Static 此配置指示承载跳过将参数(在配置 ContainerServiceArguments 中指定)传递到 Docker 守护程序。This config tells hosting to skip passing arguments (specified in config ContainerServiceArguments) to docker daemon.

HttpGatewayHttpGateway

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ActiveListenersActiveListeners Uint,默认值为 50Uint, default is 50 静态Static 要发布到 http 服务器队列的读取数。Number of reads to post to the http server queue. 此配置控制 HttpGateway 可以满足的并发请求数。This controls the number of concurrent requests that can be satisfied by the HttpGateway.
HttpGatewayHealthReportSendIntervalHttpGatewayHealthReportSendInterval 以秒为单位的时间,默认值为 30Time in seconds, default is 30 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. Http 网关将累积的运行状况报告发送至运行状况管理器的时间间隔。The interval at which the Http Gateway sends accumulated health reports to the Health Manager.
HttpStrictTransportSecurityHeaderHttpStrictTransportSecurityHeader 字符串,默认值为“”string,default is "" 动态Dynamic 指定“HTTP 严格传输安全性”标头值,该值将包括在 HttpGateway 发送的每个响应中。Specify the HTTP Strict Transport Security header value to be included in every response sent by the HttpGateway. 设置为空字符串时,该标头不会包含在网关响应中。When set to empty string; this header will not be included in the gateway response.
IsEnabledIsEnabled Bool,默认值为 falseBool, default is false 静态Static 启用/禁用 HttpGateway。Enables/Disables the HttpGateway. HttpGateway 默认已禁用。HttpGateway is disabled by default.
MaxEntityBodySizeMaxEntityBodySize Uint,默认值为 4194304Uint, default is 4194304 动态Dynamic 提供可预期的 http 请求正文的最大大小。Gives the maximum size of the body that can be expected from an http request. 默认值为 4MB。Default value is 4MB. 如果请求的正文大小大于此值,Httpgateway 将无法满足该请求。Httpgateway will fail a request if it has a body of size > this value. 最小读取块区大小为 4096 个字节。Minimum read chunk size is 4096 bytes. 因此,该值必须 > = 4096。So this has to be >= 4096.

ImageStoreServiceImageStoreService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
已启用Enabled Bool,默认值为 falseBool, default is false 静态Static ImageStoreService 的已启用标志。The Enabled flag for ImageStoreService. 默认值:falseDefault: false
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 3Int, default is 3 静态Static ImageStoreService 的 MinReplicaSetSize。The MinReplicaSetSize for ImageStoreService.
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 静态Static ImageStoreService 的 PlacementConstraints。The PlacementConstraints for ImageStoreService.
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. ImageStoreService 的 QuorumLossWaitDuration。The QuorumLossWaitDuration for ImageStoreService.
ReplicaRestartWaitDurationReplicaRestartWaitDuration 以秒为单位的时间,默认值为 60.0 * 30Time in seconds, default is 60.0 * 30 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. ImageStoreService 的 ReplicaRestartWaitDuration。The ReplicaRestartWaitDuration for ImageStoreService.
StandByReplicaKeepDurationStandByReplicaKeepDuration 以秒为单位的时间,默认值为 3600.0 * 2Time in seconds, default is 3600.0 * 2 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. ImageStoreService 的 StandByReplicaKeepDuration。The StandByReplicaKeepDuration for ImageStoreService.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 7Int, default is 7 静态Static ImageStoreService 的 TargetReplicaSetSize。The TargetReplicaSetSize for ImageStoreService.

KtlLoggerKtlLogger

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AutomaticMemoryConfigurationAutomaticMemoryConfiguration Int,默认值为 1Int, default is 1 动态Dynamic 该标志指示是否应自动且动态地配置内存设置。Flag that indicates if the memory settings should be automatically and dynamically configured. 如果设置为 0,则根据系统条件直接使用内存配置设置而不进行任何更改。If zero then the memory configuration settings are used directly and do not change based on system conditions. 如果设置为 1,则自动配置内存设置,并可根据系统条件进行更改。If one then the memory settings are configured automatically and may change based on system conditions.
MaximumDestagingWriteOutstandingInKBMaximumDestagingWriteOutstandingInKB Int,默认值为 0Int, default is 0 动态Dynamic 共享日志可位于专用日志之前的 KB 数。The number of KB to allow the shared log to advance ahead of the dedicated log. 使用 0 表示没有限制。Use 0 to indicate no limit.
SharedLogIdSharedLogId string,默认值为“”string, default is "" 静态Static 共享日志容器的唯一 guid。Unique guid for shared log container. 若要使用 Fabric 数据根目录下的默认路径,请设置为“”。Use "" if using default path under fabric data root.
SharedLogPathSharedLogPath string,默认值为“”string, default is "" 静态Static 要放置共享日志容器的位置的路径和文件名。Path and file name to location to place shared log container. 设置为“”表示使用 Fabric 数据根目录下的默认路径。Use "" for using default path under fabric data root.
SharedLogSizeInMBSharedLogSizeInMB Int,默认值为 8192Int, default is 8192 静态Static 共享日志容器中要分配的 MB 数。The number of MB to allocate in the shared log container.
SharedLogThrottleLimitInPercentUsedSharedLogThrottleLimitInPercentUsed int,默认值为 0int, default is 0 静态Static 将引发限制的共享日志使用百分比。The percentage of usage of the shared log that will induce throttling. 值应当介于 0 和 100 之间。Value should be between 0 and 100. 值为 0 表示使用默认百分比值。A value of 0 implies using the default percentage value. 值为 100 表示根本不进行限制。A value of 100 implies no throttling at all. 如果值介于 1 到 99 之间,则指定日志使用百分比,高于该百分比将进行限制;例如如果共享日志为 10 GB,而该值为 90,则将在使用 9 GB 后进行限制。A value between 1 and 99 specifies the percentage of log usage above which throttling will occur; for example if the shared log is 10GB and the value is 90 then throttling will occur once 9GB is in use. 建议使用默认值。Using the default value is recommended.
WriteBufferMemoryPoolMaximumInKBWriteBufferMemoryPoolMaximumInKB Int,默认值为 0Int, default is 0 动态Dynamic 允许写入缓冲区内存池增长到的 KB 数。The number of KB to allow the write buffer memory pool to grow up to. 使用 0 表示没有限制。Use 0 to indicate no limit.
WriteBufferMemoryPoolMinimumInKBWriteBufferMemoryPoolMinimumInKB Int,默认值为 8388608Int, default is 8388608 动态Dynamic 最初为写入缓冲区内存池分配的 KB 数。The number of KB to initially allocate for the write buffer memory pool. 设置为 0 表示没有限制,默认值应与以下 SharedLogSizeInMB 值保持一致。Use 0 to indicate no limit Default should be consistent with SharedLogSizeInMB below.

ManagedIdentityTokenServiceManagedIdentityTokenService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
IsEnabledIsEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 用于控制群集中托管标识令牌服务的状态的标志;必须具有此标志才能使用 Service Fabric 应用程序的托管标识功能。Flag controlling the presence and status of the Managed Identity Token Service in the cluster;this is a prerequisite for using the managed identity functionality of Service Fabric applications.

管理Management

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AutomaticUnprovisionIntervalAutomaticUnprovisionInterval 时间范围,默认值为 Common::TimeSpan::FromMinutes(5)TimeSpan, default is Common::TimeSpan::FromMinutes(5) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在自动应用程序类型清理期间,允许取消注册应用程序类型的清理时间间隔。The cleanup interval for allowed for unregister application type during automatic application type cleanup.
AzureStorageMaxConnectionsAzureStorageMaxConnections Int,默认值为 5000Int, default is 5000 动态Dynamic 最大并发 Azure 存储连接数。The maximum number of concurrent connections to azure storage.
AzureStorageMaxWorkerThreadsAzureStorageMaxWorkerThreads Int,默认值为 25Int, default is 25 动态Dynamic 最大并行工作线程数。The maximum number of worker threads in parallel.
AzureStorageOperationTimeoutAzureStorageOperationTimeout 以秒为单位的时间,默认值为 6000Time in seconds, default is 6000 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 完成 xstore 操作的超时时间。Time out for xstore operation to complete.
CleanupApplicationPackageOnProvisionSuccessCleanupApplicationPackageOnProvisionSuccess bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 成功预配时启用或禁用应用程序包的自动清理。Enables or disables the automatic cleanup of application package on successful provision.
CleanupUnusedApplicationTypesCleanupUnusedApplicationTypes 布尔值,默认为 FALSEBool, default is FALSE 动态Dynamic 如果启用,此配置允许自动取消注册未使用的应用程序类型版本,这会跳过最近三个未使用的版本,然后释放图像存储所占用的磁盘空间。This configuration if enabled, allows to automatically unregister unused application type versions skipping the latest three unused versions, thereby trimming the disk space occupied by image store. 将在成功预配结束时对该特定应用类型触发自动清理;对于所有应用程序类型,自动清理也将每天定期运行一次。The automatic cleanup will be triggered at the end of successful provision for that specific app type and also runs periodically once a day for all the application types. 要跳过的未使用的版本数量可通过参数“MaxUnusedAppTypeVersionsToKeep”进行配置。Number of unused versions to skip is configurable using parameter "MaxUnusedAppTypeVersionsToKeep".
DisableChecksumValidationDisableChecksumValidation Bool,默认值为 falseBool, default is false 静态Static 通过此配置可在应用程序预配过程中启用或禁用校验和验证。This configuration allows us to enable or disable checksum validation during application provisioning.
DisableServerSideCopyDisableServerSideCopy Bool,默认值为 falseBool, default is false 静态Static 此配置可以在应用程序预配过程中启用或禁用 ImageStore 上应用程序包的服务器端副本。This configuration enables or disables server-side copy of application package on the ImageStore during application provisioning.
ImageCachingEnabledImageCachingEnabled Bool,默认值为 trueBool, default is true 静态Static 通过此配置可启用或禁用缓存。This configuration allows us to enable or disable caching.
ImageStoreConnectionStringImageStoreConnectionString SecureStringSecureString 静态Static ImageStore 的根的连接字符串。Connection string to the Root for ImageStore.
ImageStoreMinimumTransferBPSImageStoreMinimumTransferBPS Int,默认值为 1024Int, default is 1024 动态Dynamic 群集和 ImageStore 之间的最小传输速率。The minimum transfer rate between the cluster and ImageStore. 此值用于确定访问外部 ImageStore 时的超时时间。This value is used to determine the timeout when accessing the external ImageStore. 仅当群集和 ImageStore 之间的延迟较高时可更改此值,以允许群集获得更多的时间从外部 ImageStore 进行下载。Change this value only if the latency between the cluster and ImageStore is high to allow more time for the cluster to download from the external ImageStore.
MaxUnusedAppTypeVersionsToKeepMaxUnusedAppTypeVersionsToKeep Int,默认值为 3Int, default is 3 动态Dynamic 此配置会定义要跳过清理操作的未使用的应用程序类型版本数。This configuration defines the number of unused application type versions to be skipped for cleanup. 仅在启用参数 CleanupUnusedApplicationTypes 后,此参数才适用。This parameter is applicable only if parameter CleanupUnusedApplicationTypes is enabled.

MetricActivityThresholdsMetricActivityThresholds

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyIntegerValueMap,默认值为 NoneKeyIntegerValueMap, default is None 动态Dynamic 为群集中的指标确定一组 MetricActivityThresholds。Determines the set of MetricActivityThresholds for the metrics in the cluster. 如果 maxNodeLoad 大于 MetricActivityThresholds,则均衡会起作用。Balancing will work if maxNodeLoad is greater than MetricActivityThresholds. 定义碎片整理指标的负载量,等于或低于此值 Service Fabric 将认为该节点为空For defrag metrics it defines the amount of load equal to or below which Service Fabric will consider the node empty

MetricBalancingThresholdsMetricBalancingThresholds

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyDoubleValueMap,默认值为 NoneKeyDoubleValueMap, default is None 动态Dynamic 为群集中的指标确定一组 MetricBalancingThresholds。Determines the set of MetricBalancingThresholds for the metrics in the cluster. 如果 maxNodeLoad/minNodeLoad 大于 MetricBalancingThresholds,则均衡会起作用。Balancing will work if maxNodeLoad/minNodeLoad is greater than MetricBalancingThresholds. 如果至少一个 FD 或 UD 上的 maxNodeLoad/minNodeLoad 小于 MetricBalancingThresholds,则碎片整理会起作用。Defragmentation will work if maxNodeLoad/minNodeLoad in at least one FD or UD is smaller than MetricBalancingThresholds.

MetricLoadStickinessForSwapMetricLoadStickinessForSwap

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyDoubleValueMap,默认值为 NoneKeyDoubleValueMap, default is None 动态Dynamic 确定在交换后继续使用副本的负载部分。它采用介于 0(负载不继续使用副本)到 1(负载要继续使用副本)之间的值Determines the part of the load that sticks with replica when swapped It takes value between 0 (load doesn't stick with replica) and 1 (load sticks with replica - default)

NamingServiceNamingService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
GatewayServiceDescriptionCacheLimitGatewayServiceDescriptionCacheLimit Int,默认值为 0Int, default is 0 静态Static 命名网关处的 LRU 服务说明缓存中可维持的最大条目数(设置为 0 表示无限制)。The maximum number of entries maintained in the LRU service description cache at the Naming Gateway (set to 0 for no limit).
MaxClientConnectionsMaxClientConnections Int,默认值为 1000Int, default is 1000 动态Dynamic 每个网关允许的最大客户端连接数。The maximum allowed number of client connections per gateway.
MaxFileOperationTimeoutMaxFileOperationTimeout 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 文件存储服务操作的最大超时时间。The maximum timeout allowed for file store service operation. 会拒绝指定更长超时时间的请求。Requests specifying a larger timeout will be rejected.
MaxIndexedEmptyPartitionsMaxIndexedEmptyPartitions Int,默认值为 1000Int, default is 1000 动态Dynamic 将在通知缓存中保持被索引状态以同步重新连接的客户端的最大空分区数。The maximum number of empty partitions that will remain indexed in the notification cache for synchronizing reconnecting clients. 将按查找版本的升序顺序,从索引中删除超出此数目的所有空分区。Any empty partitions above this number will be removed from the index in ascending lookup version order. 重新连接的客户端仍然可以同步并接收错过的空分区更新,但是同步协议会更昂贵。Reconnecting clients can still synchronize and receive missed empty partition updates; but the synchronization protocol becomes more expensive.
MaxMessageSizeMaxMessageSize Int,默认值为 4*1024*1024Int, default is 4*1024*1024 静态Static 使用命名时客户端节点通信的最大消息大小。The maximum message size for client node communication when using naming. DOS 攻击缓解,默认值为 4MB。DOS attack alleviation; default value is 4MB.
MaxNamingServiceHealthReportsMaxNamingServiceHealthReports Int,默认值为 10Int, default is 10 动态Dynamic 命名存储服务一次所报告的运行不正常的最大慢速操作数量。The maximum number of slow operations that Naming store service reports unhealthy at one time. 如果设置为 0,则将发送所有慢速操作。If 0; all slow operations are sent.
MaxOperationTimeoutMaxOperationTimeout 以秒为单位的时间,默认值为 600Time in seconds, default is 600 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 所允许的客户端操作的最大超时时间。The maximum timeout allowed for client operations. 将拒绝指定更长超时时间的请求。Requests specifying a larger timeout will be rejected.
MaxOutstandingNotificationsPerClientMaxOutstandingNotificationsPerClient Int,默认值为 1000Int, default is 1000 动态Dynamic 网关强行关闭客户端注册前的最大未完成通知数。The maximum number of outstanding notifications before a client registration is forcibly closed by the gateway.
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 3Int, default is 3 不允许Not Allowed 完成更新所需写入的最少命名服务副本数。The minimum number of Naming Service replicas required to write into to complete an update. 如果系统中的活动副本数少于此数量,则可靠性系统将拒绝对命名服务存储执行的更新操作,直到副本还原为止。If there are fewer replicas than this active in the system the Reliability System denies updates to the Naming Service Store until replicas are restored. 此值不应超过 TargetReplicaSetSize 的值。This value should never be more than the TargetReplicaSetSize.
PartitionCountPartitionCount Int,默认值为 3Int, default is 3 不允许Not Allowed 要创建的命名服务存储的分区数。The number of partitions of the Naming Service store to be created. 每个分区都拥有与其索引相对应的一个分区键,因此存在分区键 [0; PartitionCount]。Each partition owns a single partition key that corresponds to its index; so partition keys [0; PartitionCount] exist. 增加命名服务分区数可减少由任何备份副本集保持的数据的平均量,从而增加命名服务可以执行的规模;其代价是增加资源的利用(因为必须维护 PartitionCount*ReplicaSetSize 服务副本)。Increasing the number of Naming Service partitions increases the scale that the Naming Service can perform at by decreasing the average amount of data held by any backing replica set; at a cost of increased utilization of resources (since PartitionCount*ReplicaSetSize service replicas must be maintained).
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 不允许Not Allowed 命名服务的放置约束。Placement constraint for the Naming Service.
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 不允许Not Allowed 指定以秒为单位的时间跨度。Specify timespan in seconds. 命名服务进入仲裁丢失状态时,此计时器将启动。When a Naming Service gets into quorum loss; this timer starts. 指定时间到期后,FM 将不可用副本视为丢失,并尝试恢复仲裁。When it expires the FM will consider the down replicas as lost; and attempt to recover quorum. 请注意,这可能导致数据丢失。Not that this may result in data loss.
RepairIntervalRepairInterval 以秒为单位的时间,默认值为 5Time in seconds, default is 5 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 针对授权所有者和名称所有者之间命名不一致情况的修复操作的时间间隔。Interval in which the naming inconsistency repair between the authority owner and name owner will start.
ReplicaRestartWaitDurationReplicaRestartWaitDuration 以秒为单位的时间,默认值为 (60.0 * 30)Time in seconds, default is (60.0 * 30) 不允许Not Allowed 指定以秒为单位的时间范围。Specify timespan in seconds. 命名服务副本不可用时,此定时器将启动。When a Naming Service replica goes down; this timer starts. 当它到期时,FM 将开始替换已经关闭的副本(暂不将其视为丢失)。When it expires the FM will begin to replace the replicas which are down (it does not yet consider them lost).
ServiceDescriptionCacheLimitServiceDescriptionCacheLimit Int,默认值为 0Int, default is 0 静态Static 命名存储服务处的 LRU 服务说明缓存中可维持的最大条目数(设置为 0 表示无限制)。The maximum number of entries maintained in the LRU service description cache at the Naming Store Service (set to 0 for no limit).
ServiceNotificationTimeoutServiceNotificationTimeout 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 将服务通知传送到客户端时使用的超时时间。The timeout used when delivering service notifications to the client.
StandByReplicaKeepDurationStandByReplicaKeepDuration 以秒为单位的时间,默认值为 3600.0 * 2Time in seconds, default is 3600.0 * 2 不允许Not Allowed 指定以秒为单位的时间范围。Specify timespan in seconds. 命名服务副本从不可用状态恢复时,可能已被替换为另一副本。When a Naming Service replica come back from a down state; it may have already been replaced. 此定时器决定 FM 在放弃备用副本之前保留其多长时间。This timer determines how long the FM will keep the standby replica before discarding it.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 7Int, default is 7 不允许Not Allowed 命名服务存储的每个分区的副本集数量。The number of replica sets for each partition of the Naming Service store. 增加副本集的数量会增加命名服务存储中信息的可靠性水平;减少此更改将导致信息由于节点故障而丢失;其代价是增加 Windows Fabric 上的负载以及对命名数据执行更新所花费的时间。Increasing the number of replica sets increases the level of reliability for the information in the Naming Service Store; decreasing the change that the information will be lost as a result of node failures; at a cost of increased load on Windows Fabric and the amount of time it takes to perform updates to the naming data.

NodeBufferPercentageNodeBufferPercentage

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup KeyDoubleValueMap,默认值为 NoneKeyDoubleValueMap, default is None 动态Dynamic 每个指标名称的节点容量百分比;用作缓冲区,以便在故障转移的情况下在节点上保留一些可用空间。Node capacity percentage per metric name; used as a buffer in order to keep some free place on a node for the failover case.

NodeCapacitiesNodeCapacities

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup NodeCapacityCollectionMapNodeCapacityCollectionMap 静态Static 不同指标的节点容量集合。A collection of node capacities for different metrics.

NodeDomainIdsNodeDomainIds

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup NodeFaultDomainIdCollectionNodeFaultDomainIdCollection 静态Static 描述节点所属的容错域。Describes the fault domains a node belongs to. 通过用于描述数据中心中节点所在位置的 URI 定义容错域。The fault domain is defined through a URI that describes the location of the node in the datacenter. 容错域 URI 的格式是 fd:/fd/,后跟 URI 路径段。Fault Domain URIs are of the format fd:/fd/ followed by a URI path segment.
UpgradeDomainIdUpgradeDomainId string,默认值为“”string, default is "" 静态Static 描述节点所属的升级域。Describes the upgrade domain a node belongs to.

NodePropertiesNodeProperties

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup NodePropertyCollectionMapNodePropertyCollectionMap 静态Static 节点属性的字符串键值对的集合。A collection of string key-value pairs for node properties.

PaasPaas

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ClusterIdClusterId string,默认值为“”string, default is "" 不允许Not Allowed 由 Fabric 用于配置保护的 X509 证书存储。X509 certificate store used by fabric for configuration protection.

PerformanceCounterLocalStorePerformanceCounterLocalStore

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
计数器Counters StringString 动态Dynamic 要收集的性能计数器的逗号分隔列表。Comma-separated list of performance counters to collect.
IsEnabledIsEnabled Bool,默认值为 trueBool, default is true 动态Dynamic 标志指示是否启用本地节点上的性能计数器集合。Flag indicates whether performance counter collection on local node is enabled.
MaxCounterBinaryFileSizeInMBMaxCounterBinaryFileSizeInMB Int,默认值为 1Int, default is 1 动态Dynamic 每个性能计数器二进制文件的最大大小(以 MB 为单位)。Maximum size (in MB) for each performance counter binary file.
NewCounterBinaryFileCreationIntervalInMinutesNewCounterBinaryFileCreationIntervalInMinutes Int,默认值为 10Int, default is 10 动态Dynamic 在其之后创建新的性能计数器二进制文件的最大间隔(以秒为单位)。Maximum interval (in seconds) after which a new performance counter binary file is created.
SamplingIntervalInSecondsSamplingIntervalInSeconds Int,默认值为 60Int, default is 60 动态Dynamic 正在收集的性能计数器的采样间隔。Sampling interval for performance counters being collected.

PlacementAndLoadBalancingPlacementAndLoadBalancing

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AffinityConstraintPriorityAffinityConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定相关性约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of affinity constraint: 0: Hard; 1: Soft; negative: Ignore.
ApplicationCapacityConstraintPriorityApplicationCapacityConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定容量约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of capacity constraint: 0: Hard; 1: Soft; negative: Ignore.
AutoDetectAvailableResourcesAutoDetectAvailableResources bool,默认值为 TRUEbool, default is TRUE 静态Static 此配置会触发对节点上可用资源(CPU 和内存)的自动检测,将此配置设置为 true 时,如果用户已指定错误的节点容量或者根本未定义它们,我们会读取实际容量并更正它们,将此配置设置为 false 时 - 我们会跟踪用户已指定错误的节点容量这一警告;但不会纠正它们;这意味着用户希望将容量指定为 > 节点实际拥有的容量,或者如果未指定容量;它会认为是不限的容量This config will trigger auto detection of available resources on node (CPU and Memory) When this config is set to true - we will read real capacities and correct them if user specified bad node capacities or didn't define them at all If this config is set to false - we will trace a warning that user specified bad node capacities; but we will not correct them; meaning that user wants to have the capacities specified as > than the node really has or if capacities are undefined; it will assume unlimited capacity
BalancingDelayAfterNewNodeBalancingDelayAfterNewNode 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 添加新节点后,不在此时间段内启动平衡活动。Do not start balancing activities within this period after adding a new node.
BalancingDelayAfterNodeDownBalancingDelayAfterNodeDown 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 发生节点关闭事件后,不在此时间段内启动平衡活动。Do not start balancing activities within this period after a node down event.
CapacityConstraintPriorityCapacityConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定容量约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of capacity constraint: 0: Hard; 1: Soft; negative: Ignore.
ConsecutiveDroppedMovementsHealthReportLimitConsecutiveDroppedMovementsHealthReportLimit Int,默认值为 20Int, default is 20 动态Dynamic 定义在进行诊断并发出运行状况警告之前,删除 ResourceBalancer 发出的移动的连续次数。Defines the number of consecutive times that ResourceBalancer-issued Movements are dropped before diagnostics are conducted and health warnings are emitted. 负值:在此情况下没有发出任何警告。Negative: No Warnings Emitted under this condition.
ConstraintFixPartialDelayAfterNewNodeConstraintFixPartialDelayAfterNewNode 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 添加新节点后,不在此时间段内修复 FaultDomain 和 UpgradeDomain 约束冲突。DDo not Fix FaultDomain and UpgradeDomain constraint violations within this period after adding a new node.
ConstraintFixPartialDelayAfterNodeDownConstraintFixPartialDelayAfterNodeDown 以秒为单位的时间,默认值为 120Time in seconds, default is 120 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 发生节点关闭事件后,不在此时间段内修复 FaultDomain 和 UpgradeDomain 约束冲突。Do not Fix FaultDomain and UpgradeDomain constraint violations within this period after a node down event.
ConstraintViolationHealthReportLimitConstraintViolationHealthReportLimit Int,默认值为 50Int, default is 50 动态Dynamic 定义在进行诊断并发出运行状况报告之前,违反约束的副本持续处于未解决状态的次数。Defines the number of times constraint violating replica has to be persistently unfixed before diagnostics are conducted and health reports are emitted.
DetailedConstraintViolationHealthReportLimitDetailedConstraintViolationHealthReportLimit Int,默认值为 200Int, default is 200 动态Dynamic 定义在进行诊断并发出详细运行状况报告之前,违反约束的副本持续未固定的次数。Defines the number of times constraint violating replica has to be persistently unfixed before diagnostics are conducted and detailed health reports are emitted.
DetailedDiagnosticsInfoListLimitDetailedDiagnosticsInfoListLimit Int,默认值为 15Int, default is 15 动态Dynamic 定义诊断中每个约束在截断前要包含的诊断条目数(附带详细信息)。Defines the number of diagnostic entries (with detailed information) per constraint to include before truncation in Diagnostics.
DetailedNodeListLimitDetailedNodeListLimit Int,默认值为 15Int, default is 15 动态Dynamic 定义在未放置副本报告中每个约束在截断前要包含的节点数。Defines the number of nodes per constraint to include before truncation in the Unplaced Replica reports.
DetailedPartitionListLimitDetailedPartitionListLimit Int,默认值为 15Int, default is 15 动态Dynamic 定义诊断中一个约束在截断前要包含的分区数(按诊断条目)。Defines the number of partitions per diagnostic entry for a constraint to include before truncation in Diagnostics.
DetailedVerboseHealthReportLimitDetailedVerboseHealthReportLimit Int,默认值为 200Int, default is 200 动态Dynamic 定义在发出详细运行状况报告之前,未放置的副本必须持续处于未放置状态的次数。Defines the number of times an unplaced replica has to be persistently unplaced before detailed health reports are emitted.
EnforceUserServiceMetricCapacitiesEnforceUserServiceMetricCapacities bool,默认值为 FALSEbool, default is FALSE 静态Static 启用 Fabric 服务保护。Enables fabric services protection. 所有用户服务都在一个作业对象/cgroup 下,且仅限指定的资源量使用。All user services are under one job object/cgroup and limited to specified amount of resources. 这必须是静态的(需要重启 FabricHost),因为用户作业对象和设置限制的创建/删除是在 Fabric 主机打开期间完成的。This needs to be static (requires restart of FabricHost) as creation/removal of user job object and setting limits in done during open of Fabric Host.
FaultDomainConstraintPriorityFaultDomainConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定容错域约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of fault domain constraint: 0: Hard; 1: Soft; negative: Ignore.
GlobalMovementThrottleCountingIntervalGlobalMovementThrottleCountingInterval 以秒为单位的时间,默认值为 600Time in seconds, default is 600 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 指示刚过去的用于跟踪每个域副本移动的时间间隔的长度(与 GlobalMovementThrottleThreshold 配合使用)。Indicate the length of the past interval for which to track per domain replica movements (used along with GlobalMovementThrottleThreshold). 要完全忽略全局限制,可以将其设置为 0。Can be set to 0 to ignore global throttling altogether.
GlobalMovementThrottleThresholdGlobalMovementThrottleThreshold Uint,默认值为 1000Uint, default is 1000 动态Dynamic GlobalMovementThrottleCountingInterval 所指示的刚过去的时间间隔中的平衡阶段中所允许的最大移动数。Maximum number of movements allowed in the Balancing Phase in the past interval indicated by GlobalMovementThrottleCountingInterval.
GlobalMovementThrottleThresholdForBalancingGlobalMovementThrottleThresholdForBalancing Uint,默认值为 0Uint, default is 0 动态Dynamic GlobalMovementThrottleCountingInterval 所指示的刚过去的时间间隔中的平衡阶段中所允许的最大移动数。Maximum number of movements allowed in Balancing Phase in the past interval indicated by GlobalMovementThrottleCountingInterval. 0 表示没有限制。0 indicates no limit.
GlobalMovementThrottleThresholdForPlacementGlobalMovementThrottleThresholdForPlacement Uint,默认值为 0Uint, default is 0 动态Dynamic GlobalMovementThrottleCountingInterval 所指示的刚过去的时间间隔中的放置阶段中所允许的最大移动数。0 表示没有限制。Maximum number of movements allowed in Placement Phase in the past interval indicated by GlobalMovementThrottleCountingInterval.0 indicates no limit.
GlobalMovementThrottleThresholdPercentageGlobalMovementThrottleThresholdPercentage double,默认值为 0double, default is 0 动态Dynamic 在刚过去的、由 GlobalMovementThrottleCountingInterval 指定的时间间隔中,均衡和放置阶段所允许的最大总移动数(以群集中副本总数的百分比表示)。Maximum number of total movements allowed in Balancing and Placement phases (expressed as percentage of total number of replicas in the cluster) in the past interval indicated by GlobalMovementThrottleCountingInterval. 0 表示没有限制。0 indicates no limit. 如果指定了此参数和 GlobalMovementThrottleThreshold,则使用更保守的限制。If both this and GlobalMovementThrottleThreshold are specified; then more conservative limit is used.
GlobalMovementThrottleThresholdPercentageForBalancingGlobalMovementThrottleThresholdPercentageForBalancing double,默认值为 0double, default is 0 动态Dynamic 在刚过去的、由 GlobalMovementThrottleCountingInterval 指定的时间间隔中,均衡阶段所允许的最大移动数(以 PLB 中副本总数的百分比表示)。Maximum number of movements allowed in Balancing Phase (expressed as percentage of total number of replicas in PLB) in the past interval indicated by GlobalMovementThrottleCountingInterval. 0 表示没有限制。0 indicates no limit. 如果指定了此参数和 GlobalMovementThrottleThresholdForBalancing,则使用更保守的限制。If both this and GlobalMovementThrottleThresholdForBalancing are specified; then more conservative limit is used.
InBuildThrottlingAssociatedMetricInBuildThrottlingAssociatedMetric string,默认值为“”string, default is "" 静态Static 此限制的关联指标名称。The associated metric name for this throttling.
InBuildThrottlingEnabledInBuildThrottlingEnabled Bool,默认值为 falseBool, default is false 动态Dynamic 决定是否启用内置限制。Determine whether the in-build throttling is enabled.
InBuildThrottlingGlobalMaxValueInBuildThrottlingGlobalMaxValue Int,默认值为 0Int, default is 0 动态Dynamic 全局范围内所允许的最大内置副本数。The maximal number of in-build replicas allowed globally.
InterruptBalancingForAllFailoverUnitUpdatesInterruptBalancingForAllFailoverUnitUpdates Bool,默认值为 falseBool, default is false 动态Dynamic 决定是否有任何类型的故障转移单元更新应中断快速或慢速平衡运行。Determines if any type of failover unit update should interrupt fast or slow balancing run. 如果指定为“false”,则会在 FailoverUnit 出现以下情况时中断均衡运行:被创建/删除、缺少副本、更改了主副本位置或更改了副本数量。With specified "false" balancing run will be interrupted if FailoverUnit: is created/deleted; has missing replicas; changed primary replica location or changed number of replicas. 在其他情况下不会中断均衡运行,包括 FailoverUnit 具有额外副本、更改了任何副本标志、仅更改了分区版本或任何其他情况。Balancing run will NOT be interrupted in other cases - if FailoverUnit: has extra replicas; changed any replica flag; changed only partition version or any other case.
MinConstraintCheckIntervalMinConstraintCheckInterval 以秒为单位的时间,默认值为 1Time in seconds, default is 1 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 定义在两个连续的约束检查循环之前必须经过的最短时间。Defines the minimum amount of time that must pass before two consecutive constraint check rounds.
MinLoadBalancingIntervalMinLoadBalancingInterval 以秒为单位的时间,默认值为 5Time in seconds, default is 5 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 定义在两个连续平衡循环之前必须经过的最短时间。Defines the minimum amount of time that must pass before two consecutive balancing rounds.
MinPlacementIntervalMinPlacementInterval 以秒为单位的时间,默认值为 1Time in seconds, default is 1 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 定义在两个连续的放置循环之前必须经过的最短时间。Defines the minimum amount of time that must pass before two consecutive placement rounds.
MoveExistingReplicaForPlacementMoveExistingReplicaForPlacement Bool,默认值为 trueBool, default is true 动态Dynamic 该设置决定放置过程中是否移动现有副本。Setting which determines if to move existing replica during placement.
MovementPerPartitionThrottleCountingIntervalMovementPerPartitionThrottleCountingInterval 以秒为单位的时间,默认值为 600Time in seconds, default is 600 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 指示刚过去的用于跟踪每个分区的副本移动的时间间隔的长度(与 MovementPerPartitionThrottleThreshold 配合使用)。Indicate the length of the past interval for which to track replica movements for each partition (used along with MovementPerPartitionThrottleThreshold).
MovementPerPartitionThrottleThresholdMovementPerPartitionThrottleThreshold Uint,默认值为 50Uint, default is 50 动态Dynamic 如果分区副本的均衡相关移动数量在刚过去的、由 MovementPerPartitionThrottleCountingInterval 指示的间隔时间中已达到或超过 MovementPerFailoverUnitThrottleThreshold,则该分区不会发生均衡相关的移动。No balancing-related movement will occur for a partition if the number of balancing related movements for replicas of that partition has reached or exceeded MovementPerFailoverUnitThrottleThreshold in the past interval indicated by MovementPerPartitionThrottleCountingInterval.
MoveParentToFixAffinityViolationMoveParentToFixAffinityViolation Bool,默认值为 falseBool, default is false 动态Dynamic 该设置决定是否可通过移动父副本来修复相关性约束。Setting which determines if parent replicas can be moved to fix affinity constraints.
PartiallyPlaceServicesPartiallyPlaceServices Bool,默认值为 trueBool, default is true 动态Dynamic 决定在给定有限的适当节点的情况下,是否“全部或完全不”放置群集中的所有服务副本。Determines if all service replicas in cluster will be placed "all or nothing" given limited suitable nodes for them.
PlaceChildWithoutParentPlaceChildWithoutParent Bool,默认值为 trueBool, default is true 动态Dynamic 该设置决定如果没启用父副本,是否可以放置子服务副本。Setting which determines if child service replica can be placed if no parent replica is up.
PlacementConstraintPriorityPlacementConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定放置约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of placement constraint: 0: Hard; 1: Soft; negative: Ignore.
PlacementConstraintValidationCacheSizePlacementConstraintValidationCacheSize Int,默认值为 10000Int, default is 10000 动态Dynamic 限制用于快速验证和缓存放置约束表达式的表格的大小。Limits the size of the table used for quick validation and caching of Placement Constraint Expressions.
PlacementSearchTimeoutPlacementSearchTimeout 以秒为单位的时间,默认值为 0.5Time in seconds, default is 0.5 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. 这是放置服务时,返回结果之前可搜索的最长时间。When placing services; search for at most this long before returning a result.
PLBRefreshGapPLBRefreshGap 以秒为单位的时间,默认值为 1Time in seconds, default is 1 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 定义 PLB 再次刷新状态之前必须经过的最短时间。Defines the minimum amount of time that must pass before PLB refreshes state again.
PreferredLocationConstraintPriorityPreferredLocationConstraintPriority Int,默认值为 2Int, default is 2 动态Dynamic 确定首选位置约束的优先级:0:硬;1:软;2:最佳;负值:忽略Determines the priority of preferred location constraint: 0: Hard; 1: Soft; 2: Optimization; negative: Ignore
PreferredPrimaryDomainsConstraintPriorityPreferredPrimaryDomainsConstraintPriority Int,默认值为 1Int, default is 1 动态Dynamic 确定首选主域约束的优先级:0:硬;1:软;负值:忽略Determines the priority of preferred primary domain constraint: 0: Hard; 1: Soft; negative: Ignore
PreferUpgradedUDsPreferUpgradedUDs bool,默认值为 FALSEbool,default is FALSE 动态Dynamic 启用和禁用首选移动到已升级 UD 的逻辑。Turns on and off logic which prefers moving to already upgraded UDs. 自 SF 7.0 开始,此参数的默认值从 TRUE 更改为 FALSE。Starting with SF 7.0, the default value for this parameter is changed from TRUE to FALSE.
PreventTransientOvercommitPreventTransientOvercommit Bool,默认值为 falseBool, default is false 动态Dynamic 确定 PLB 是否应该立即对由启动的移动所释放的资源进行计数。Determines should PLB immediately count on resources that will be freed up by the initiated moves. 默认情况下,PLB 可以在同一节点上发起移出和移入操作,这会造成暂时性过载。By default; PLB can initiate move out and move in on the same node which can create transient overcommit. 将此参数设置为 true 可防止这种过载,并可禁用按需碎片整理(也称为 placementWithMove)。Setting this parameter to true will prevent those kinds of overcommits and on-demand defrag (aka placementWithMove) will be disabled.
ScaleoutCountConstraintPriorityScaleoutCountConstraintPriority Int,默认值为 0Int, default is 0 动态Dynamic 确定横向扩展计数约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of scaleout count constraint: 0: Hard; 1: Soft; negative: Ignore.
SubclusteringEnabledSubclusteringEnabled 布尔值,默认为 FALSEBool, default is FALSE 动态Dynamic 在计算均衡的标准偏差时确认子群集Acknowledge subclustering when calculating standard deviation for balancing
SubclusteringReportingPolicySubclusteringReportingPolicy Int,默认值为 1Int, default is 1 动态Dynamic 定义是否应发送且如何发送子群集运行状况报告:0:不报告;1:警告;2:OKDefines how and if the subclustering health reports are sent: 0: Do not report; 1: Warning; 2: OK
SwapPrimaryThrottlingAssociatedMetricSwapPrimaryThrottlingAssociatedMetric string,默认值为“”string, default is "" 静态Static 此限制的关联指标名称。The associated metric name for this throttling.
SwapPrimaryThrottlingEnabledSwapPrimaryThrottlingEnabled Bool,默认值为 falseBool, default is false 动态Dynamic 确定是否启用交换主限制。Determine whether the swap-primary throttling is enabled.
SwapPrimaryThrottlingGlobalMaxValueSwapPrimaryThrottlingGlobalMaxValue Int,默认值为 0Int, default is 0 动态Dynamic 全局范围内所允许的最大交换主副本数。The maximal number of swap-primary replicas allowed globally.
TraceCRMReasonsTraceCRMReasons Bool,默认值为 trueBool, default is true 动态Dynamic 指定是否要寻找向操作事件通道移动(CRM 发出的移动)的原因。Specifies whether to trace reasons for CRM issued movements to the operational events channel.
UpgradeDomainConstraintPriorityUpgradeDomainConstraintPriority Int,默认值为 1Int, default is 1 动态Dynamic 确定升级域约束的优先级:0:硬;1:软;负值:忽略。Determines the priority of upgrade domain constraint: 0: Hard; 1: Soft; negative: Ignore.
UseMoveCostReportsUseMoveCostReports Bool,默认值为 falseBool, default is false 动态Dynamic 指示 LB 忽略评分函数的成本元素,从而可能产生大量可优化平衡放置的移动。Instructs the LB to ignore the cost element of the scoring function; resulting potentially large number of moves for better balanced placement.
UseSeparateSecondaryLoadUseSeparateSecondaryLoad Bool,默认值为 trueBool, default is true 动态Dynamic 确定是否应对次要副本使用单独负载的设置。Setting which determines if separate load should be used for secondary replicas.
UseSeparateSecondaryMoveCostUseSeparateSecondaryMoveCost Bool,默认值为 falseBool, default is false 动态Dynamic 确定是否应对次要副本使用单独的迁移成本的设置。Setting which determines if separate move cost should be used for secondary replicas.
ValidatePlacementConstraintValidatePlacementConstraint Bool,默认值为 trueBool, default is true 动态Dynamic 指定更新服务的服务说明时,是否验证服务的放置约束表达式。Specifies whether or not the PlacementConstraint expression for a service is validated when a service's ServiceDescription is updated.
ValidatePrimaryPlacementConstraintOnPromoteValidatePrimaryPlacementConstraintOnPromote Bool,默认值为 TRUEBool, default is TRUE 动态Dynamic 指定是否在故障转移时评估服务的 PlacementConstraint 表达式的主要首选项。Specifies whether or not the PlacementConstraint expression for a service is evaluated for primary preference on failover.
VerboseHealthReportLimitVerboseHealthReportLimit Int,默认值为 20Int, default is 20 动态Dynamic 定义副本进入未放置状态的次数超过多少次后,便报告副本运行状况警告(如果已启用详细运行状况报告)。Defines the number of times a replica has to go unplaced before a health warning is reported for it (if verbose health reporting is enabled).
NodeLoadsOperationalTracingEnabledNodeLoadsOperationalTracingEnabled Bool,默认值为 trueBool, default is true 动态Dynamic 在事件存储中启用“节点加载”操作结构跟踪的配置。Config that enables Node Load operational structural trace in the event store.
NodeLoadsOperationalTracingIntervalNodeLoadsOperationalTracingInterval 时间范围,默认值为 Common::TimeSpan::FromSeconds(20)TimeSpan, default is Common::TimeSpan::FromSeconds(20) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 对每个服务域跟踪将节点加载到事件存储的操作的时间间隔。The interval with which to trace node loads to event store for each service domain.

ReconfigurationAgentReconfigurationAgent

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ApplicationUpgradeMaxReplicaCloseDurationApplicationUpgradeMaxReplicaCloseDuration 以秒为单位的时间,默认值为 900Time in seconds, default is 900 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在应用程序升级过程中,如果服务主机具有进入关闭状态的副本,系统在终止这类服务主机前所等待的时间。The duration for which the system will wait before terminating service hosts that have replicas that are stuck in close during Application Upgrade.
FabricUpgradeMaxReplicaCloseDurationFabricUpgradeMaxReplicaCloseDuration 以秒为单位的时间,默认值为 900Time in seconds, default is 900 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在结构升级过程中,如果服务主机具有进入关闭状态的副本,系统在终止这类服务主机前所等待的时间。The duration for which the system will wait before terminating service hosts that have replicas that are stuck in close during fabric upgrade.
GracefulReplicaShutdownMaxDurationGracefulReplicaShutdownMaxDuration TimeSpan,默认值为 Common::TimeSpan::FromSeconds(120)TimeSpan, default is Common::TimeSpan::FromSeconds(120) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 如果服务主机具有进入关闭状态的副本,则该配置决定系统在终止这类服务主机前所等待的时间。The duration for which the system will wait before terminating service hosts that have replicas that are stuck in close. 如果此值设置为 0,则不会指示关闭副本。If this value is set to 0, replicas will not be instructed to close.
NodeDeactivationMaxReplicaCloseDurationNodeDeactivationMaxReplicaCloseDuration 以秒为单位的时间,默认值为 900Time in seconds, default is 900 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 在节点停用过程中,如果服务主机具有进入关闭状态的副本,系统在终止这类服务主机前所等待的时间。The duration for which the system will wait before terminating service hosts that have replicas that are stuck in close during node deactivation.
PeriodicApiSlowTraceIntervalPeriodicApiSlowTraceInterval 以秒为单位的时间,默认值为 5 分钟Time in seconds, default is 5 minutes 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. PeriodicApiSlowTraceInterval 定义 API 监视器追溯慢速 API 调用的时间间隔。PeriodicApiSlowTraceInterval defines the interval over which slow API calls will be retraced by the API monitor.
ReplicaChangeRoleFailureRestartThresholdReplicaChangeRoleFailureRestartThreshold int,默认值为 10int, default is 10 动态Dynamic 整数。Integer. 指定在主升级过程中 API 故障的次数,超过此次数后应用自动缓解措施(副本重启)。Specify the number of API failures during primary promotion after which auto-mitigation action (replica restart) will be applied.
ReplicaChangeRoleFailureWarningReportThresholdReplicaChangeRoleFailureWarningReportThreshold int,默认值为 2147483647int, default is 2147483647 动态Dynamic 整数。Integer. 指定在主升级过程中 API 故障的次数,超过此次数后引发警告运行状况报告。Specify the number of API failures during primary promotion after which warning health report will be raised.
ServiceApiHealthDurationServiceApiHealthDuration 以秒为单位的时间,默认值为 30 分钟Time in seconds, default is 30 minutes 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. ServiceApiHealthDuration 定义在等待多少时间后 API 仍未运行的话就报告其运行不正常。ServiceApiHealthDuration defines how long do we wait for a service API to run before we report it unhealthy.
ServiceReconfigurationApiHealthDurationServiceReconfigurationApiHealthDuration 以秒为单位的时间,默认值为 30Time in seconds, default is 30 动态Dynamic 指定以秒为单位的时间跨度。Specify timespan in seconds. ServiceReconfigurationApiHealthDuration 定义在报告运行不正常之前,等待服务 API 运行的时间。ServiceReconfigurationApiHealthDuration defines how long do we wait for a service API to run before we report unhealthy. 这适用于影响可用性的 API 调用。This applies to API calls that impact availability.

复制Replication

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
BatchAcknowledgementIntervalBatchAcknowledgementInterval TimeSpan,默认值为 Common::TimeSpan::FromMilliseconds(15)TimeSpan, default is Common::TimeSpan::FromMilliseconds(15) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 确定接收到操作后到发送回确认之前,复制器等待的时间。Determines the amount of time that the replicator waits after receiving an operation before sending back an acknowledgement. 在该时间段期间接收的其他操作将通过一条消息发送回其确认 -> 减少网络流量,同时可能降低复制器的吞吐量。Other operations received during this time period will have their acknowledgements sent back in a single message-> reducing network traffic but potentially reducing the throughput of the replicator.
MaxCopyQueueSizeMaxCopyQueueSize uint,默认值为 1024uint, default is 1024 静态Static 这是用于定义队列初始大小的最大值,该队列用于维护复制操作。This is the maximum value defines the initial size for the queue which maintains replication operations. 请注意,它必须是 2 的幂。Note that it must be a power of 2. 如果在运行时该队列增长到此大小,将限制主复制器和辅助复制器之间的操作。If during runtime the queue grows to this size operation will be throttled between the primary and secondary replicators.
MaxPrimaryReplicationQueueMemorySizeMaxPrimaryReplicationQueueMemorySize uint,默认值为 0uint, default is 0 静态Static 这是主复制队列的最大值(以字节为单位)。This is the maximum value of the primary replication queue in bytes.
MaxPrimaryReplicationQueueSizeMaxPrimaryReplicationQueueSize uint,默认值为 1024uint, default is 1024 静态Static 这是主复制队列中可以存在的最大操作数量。This is the maximum number of operations that could exist in the primary replication queue. 请注意,它必须是 2 的幂。Note that it must be a power of 2.
MaxReplicationMessageSizeMaxReplicationMessageSize uint,默认值为 52428800uint, default is 52428800 静态Static 复制操作的最大消息大小。Maximum message size of replication operations. 默认值为 50MB。Default is 50MB.
MaxSecondaryReplicationQueueMemorySizeMaxSecondaryReplicationQueueMemorySize uint,默认值为 0uint, default is 0 静态Static 这是辅助复制队列的最大值(以字节为单位)。This is the maximum value of the secondary replication queue in bytes.
MaxSecondaryReplicationQueueSizeMaxSecondaryReplicationQueueSize uint,默认值为 2048uint, default is 2048 静态Static 这是辅助复制队列中可以存在的最大操作数量。This is the maximum number of operations that could exist in the secondary replication queue. 请注意,它必须是 2 的幂。Note that it must be a power of 2.
QueueHealthMonitoringIntervalQueueHealthMonitoringInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(30)TimeSpan, default is Common::TimeSpan::FromSeconds(30) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 此值确定复制器用于监视复制操作队列中任何警告/错误运行状况事件的时间段。This value determines the time period used by the Replicator to monitor any warning/error health events in the replication operation queues. 如果值为 "0",则禁用运行状况监视A value of '0' disables health monitoring
QueueHealthWarningAtUsagePercentQueueHealthWarningAtUsagePercent uint,默认值为 80uint, default is 80 静态Static 此值确定复制队列使用率(以百分比表示),超过该值后我们会报告高队列使用率警告。This value determines the replication queue usage(in percentage) after which we report warning about high queue usage. QueueHealthMonitoringInterval 的宽限期间隔后,进行此操作。We do so after a grace interval of QueueHealthMonitoringInterval. 如果宽限期时间间隔内队列使用率低于此百分比If the queue usage falls below this percentage in the grace interval
ReplicatorAddressReplicatorAddress string,默认值为“localhost:0”string, default is "localhost:0" 静态Static 采用字符串形式 -'IP:Port' 的终结点,Windows Fabric 复制器将其用于与其他副本建立连接以发送/接收操作。The endpoint in form of a string -'IP:Port' which is used by the Windows Fabric Replicator to establish connections with other replicas in order to send/receive operations.
ReplicatorListenAddressReplicatorListenAddress string,默认值为“localhost:0”string, default is "localhost:0" 静态Static 采用字符串形式 -'IP:Port' 的终结点,Windows Fabric 复制器将其用于从其他副本接收操作。The endpoint in form of a string -'IP:Port' which is used by the Windows Fabric Replicator to receive operations from other replicas.
ReplicatorPublishAddressReplicatorPublishAddress string,默认值为“localhost:0”string, default is "localhost:0" 静态Static 采用字符串形式 -'IP:Port' 的终结点,Windows Fabric 复制器将其用于向其他副本发送操作。The endpoint in form of a string -'IP:Port' which is used by the Windows Fabric Replicator to send operations to other replicas.
RetryIntervalRetryInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(5)TimeSpan, default is Common::TimeSpan::FromSeconds(5) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 丢失或拒绝操作时,此计时器确定复制器重试发送该操作的频率。When an operation is lost or rejected this timer determines how often the replicator will retry sending the operation.

ResourceMonitorServiceResourceMonitorService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
IsEnabledIsEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 控制是否在群集中启用该服务。Controls if the service is enabled in the cluster or not.

RunAsRunAs

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
RunAsAccountNameRunAsAccountName string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户名称。Indicates the RunAs account name. 仅需用于“DomainUser”或“ManagedServiceAccount”帐户类型。This is only needed for "DomainUser" or "ManagedServiceAccount" account type. 有效值为“domain\user”或“user@domain”。Valid values are "domain\user" or "user@domain".
RunAsAccountTypeRunAsAccountType string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户类型。Indicates the RunAs account type. 需用于任何 RunAs 部分,有效值为“DomainUser/NetworkService/ManagedServiceAccount/LocalSystem”。This is needed for any RunAs section Valid values are "DomainUser/NetworkService/ManagedServiceAccount/LocalSystem".
RunAsPasswordRunAsPassword string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户密码。Indicates the RunAs account password. 仅需用于“DomainUser”帐户类型。This is only needed for "DomainUser" account type.

RunAs_DCARunAs_DCA

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
RunAsAccountNameRunAsAccountName string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户名称。Indicates the RunAs account name. 仅需用于“DomainUser”或“ManagedServiceAccount”帐户类型。This is only needed for "DomainUser" or "ManagedServiceAccount" account type. 有效值为“domain\user”或“user@domain”。Valid values are "domain\user" or "user@domain".
RunAsAccountTypeRunAsAccountType string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户类型。Indicates the RunAs account type. 需用于任何 RunAs 部分,有效值为“LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem”。This is needed for any RunAs section Valid values are "LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem".
RunAsPasswordRunAsPassword string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户密码。Indicates the RunAs account password. 仅需用于“DomainUser”帐户类型。This is only needed for "DomainUser" account type.

RunAs_FabricRunAs_Fabric

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
RunAsAccountNameRunAsAccountName string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户名称。Indicates the RunAs account name. 仅需用于“DomainUser”或“ManagedServiceAccount”帐户类型。This is only needed for "DomainUser" or "ManagedServiceAccount" account type. 有效值为“domain\user”或“user@domain”。Valid values are "domain\user" or "user@domain".
RunAsAccountTypeRunAsAccountType string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户类型。Indicates the RunAs account type. 需用于任何 RunAs 部分,有效值为“LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem”。This is needed for any RunAs section Valid values are "LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem".
RunAsPasswordRunAsPassword string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户密码。Indicates the RunAs account password. 仅需用于“DomainUser”帐户类型。This is only needed for "DomainUser" account type.

RunAs_HttpGatewayRunAs_HttpGateway

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
RunAsAccountNameRunAsAccountName string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户名称。Indicates the RunAs account name. 仅需用于“DomainUser”或“ManagedServiceAccount”帐户类型。This is only needed for "DomainUser" or "ManagedServiceAccount" account type. 有效值为“domain\user”或“user@domain”。Valid values are "domain\user" or "user@domain".
RunAsAccountTypeRunAsAccountType string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户类型。Indicates the RunAs account type. 需用于任何 RunAs 部分,有效值为“LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem”。This is needed for any RunAs section Valid values are "LocalUser/DomainUser/NetworkService/ManagedServiceAccount/LocalSystem".
RunAsPasswordRunAsPassword string,默认值为“”string, default is "" 动态Dynamic 指示 RunAs 帐户密码。Indicates the RunAs account password. 仅需用于“DomainUser”帐户类型。This is only needed for "DomainUser" account type.

安全性Security

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AADCertEndpointFormatAADCertEndpointFormat string,默认值为“”string, default is "" 静态Static AAD 证书终结点格式,默认为“Azure 商业版”,为非默认环境指定,例如 Azure 中国云“https://portal.partner.microsoftonline.cn/{0}/federationmetadata/2007-06/federationmetadata.xml”AAD Cert Endpoint Format, default Azure Commercial, specified for non-default environment such as Azure China Cloud "https://portal.partner.microsoftonline.cn/{0}/federationmetadata/2007-06/federationmetadata.xml"
AADClientApplicationAADClientApplication string,默认值为“”string, default is "" 静态Static 表示 Fabric 客户端的本机客户端应用程序名称或 IDNative Client application name or ID representing Fabric Clients
AADClusterApplicationAADClusterApplication string,默认值为“”string, default is "" 静态Static 表示群集的 Web API 应用程序名称或 IDWeb API application name or ID representing the cluster
AADLoginEndpointAADLoginEndpoint string,默认值为“”string, default is "" 静态Static AAD 登录终结点,默认为“Azure 商业版”,为非默认环境指定,例如 Azure 中国云“https://portal.partner.microsoftonline.cn”AAD Login Endpoint, default Azure Commercial, specified for non-default environment such as Azure China Cloud "https://portal.partner.microsoftonline.cn"
AADTenantIdAADTenantId string,默认值为“”string, default is "" 静态Static 租户 ID (GUID)Tenant ID (GUID)
AcceptExpiredPinnedClusterCertificateAcceptExpiredPinnedClusterCertificate bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 指示是否接受由指纹声明但已过期的群集证书。仅适用于群集证书,因此请将群集保持在活动状态。Flag indicating whether to accept expired cluster certificates declared by thumbprint Applies only to cluster certificates; so as to keep the cluster alive.
AdminClientCertThumbprintsAdminClientCertThumbprints string,默认值为“”string, default is "" 动态Dynamic 管理员角色客户端使用的证书的指纹。Thumbprints of certificates used by clients in admin role. 该参数是以逗号分隔的名称列表。It is a comma-separated name list.
AADTokenEndpointFormatAADTokenEndpointFormat string,默认值为“”string, default is "" 静态Static AAD 令牌终结点,默认为“Azure 商业版”,为非默认环境指定,例如 Azure 中国云“https://portal.partner.microsoftonline.cn/{0}”AAD Token Endpoint, default Azure Commercial, specified for non-default environment such as Azure China Cloud "https://portal.partner.microsoftonline.cn/{0}"
AdminClientClaimsAdminClientClaims string,默认值为“”string, default is "" 动态Dynamic 管理员客户端所需的所有可能的声明;其格式与 ClientClaims 相同;此列表会从内部添加到 ClientClaims;所以不需要另外将相同的条目添加到 ClientClaims。All possible claims expected from admin clients; the same format as ClientClaims; this list internally gets added to ClientClaims; so no need to also add the same entries to ClientClaims.
AdminClientIdentitiesAdminClientIdentities string,默认值为“”string, default is "" 动态Dynamic 管理员角色结构客户端的 Windows 标识;用于授权特权结构操作。Windows identities of fabric clients in admin role; used to authorize privileged fabric operations. 该参数是以逗号分隔的列表;每个条目是域帐户名或组名称。It is a comma-separated list; each entry is a domain account name or group name. 为方便起见,自动向运行 fabric.exe 的帐户分配管理员角色;组 ServiceFabricAdministrators 也是这样。For convenience; the account that runs fabric.exe is automatically assigned admin role; so is group ServiceFabricAdministrators.
AppRunAsAccountGroupX509FolderAppRunAsAccountGroupX509Folder 字符串,默认为 /home/sfuser/sfusercertsstring, default is /home/sfuser/sfusercerts 静态Static AppRunAsAccountGroup X509 证书和私钥所在的文件夹Folder where AppRunAsAccountGroup X509 certificates and private keys are located
CertificateExpirySafetyMarginCertificateExpirySafetyMargin TimeSpan,默认值为 Common::TimeSpan::FromMinutes(43200)TimeSpan, default is Common::TimeSpan::FromMinutes(43200) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 证书过期的安全宽限期;过期日期接近此时间值时证书运行状况报告状态从“OK”更改为“警告”。Safety margin for certificate expiration; certificate health report status changes from OK to Warning when expiration is closer than this. 默认值为 30 天。Default is 30 days.
CertificateHealthReportingIntervalCertificateHealthReportingInterval TimeSpan,默认值为 Common::TimeSpan::FromSeconds(3600 * 8)TimeSpan, default is Common::TimeSpan::FromSeconds(3600 * 8) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 指定证书运行状况报告的时间间隔;默认值为 8 小时;设置为 0 会禁用证书运行状况报告Specify interval for certificate health reporting; default to 8 hours; setting to 0 disables certificate health reporting
ClientCertThumbprintsClientCertThumbprints string,默认值为“”string, default is "" 动态Dynamic 客户端用于与群集通信的证书的指纹;群集使用此指纹授权传入连接。Thumbprints of certificates used by clients to talk to the cluster; cluster uses this authorize incoming connection. 该参数是以逗号分隔的名称列表。It is a comma-separated name list.
ClientClaimAuthEnabledClientClaimAuthEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 指示是否对客户端启用基于声明的身份验证;将此参数设置 true 会隐式设置 ClientRoleEnabled。Indicates if claim-based authentication is enabled on clients; setting this true implicitly sets ClientRoleEnabled.
ClientClaimsClientClaims string,默认值为“”string, default is "" 动态Dynamic 客户端连接到网关所需的所有可能的声明。All possible claims expected from clients for connecting to gateway. 这是一个 'OR' 列表:ClaimsEntry || ClaimsEntry || ClaimsEntry ... 每个 ClaimsEntry 均为 "AND" 列表:ClaimType=ClaimValue && ClaimType=ClaimValue && ClaimType=ClaimValue ...This is a 'OR' list: ClaimsEntry || ClaimsEntry || ClaimsEntry ... each ClaimsEntry is a "AND" list: ClaimType=ClaimValue && ClaimType=ClaimValue && ClaimType=ClaimValue ...
ClientIdentitiesClientIdentities string,默认值为“”string, default is "" 动态Dynamic FabricClient 的 Windows 标识;命名网关使用该参数授权传入连接。Windows identities of FabricClient; naming gateway uses this to authorize incoming connections. 该参数是以逗号分隔的列表;每个条目是域帐户名或组名称。It is a comma-separated list; each entry is a domain account name or group name. 为方便起见,自动允许运行 fabric.exe 的帐户;组 ServiceFabricAllowedUsers 和 ServiceFabricAdministrators 也是这样。For convenience; the account that runs fabric.exe is automatically allowed; so are group ServiceFabricAllowedUsers and ServiceFabricAdministrators.
ClientRoleEnabledClientRoleEnabled bool,默认值为 FALSEbool, default is FALSE 静态Static 指示是否启用了客户端角色。设置为 true,则基于客户端标识为其分配角色。Indicates if client role is enabled; when set to true; clients are assigned roles based on their identities. 对于 V2;启用该角色意味着不在 AdminClientCommonNames/AdminClientIdentities 中的客户端只能执行只读操作。For V2; enabling this means client not in AdminClientCommonNames/AdminClientIdentities can only execute read-only operations.
ClusterCertThumbprintsClusterCertThumbprints string,默认值为“”string, default is "" 动态Dynamic 允许加入群集的证书的指纹;以逗号分隔的名称列表。Thumbprints of certificates allowed to join the cluster; a comma-separated name list.
ClusterCredentialTypeClusterCredentialType string,默认值为“None”string, default is "None" 不允许Not Allowed 指示为保护群集安全而使用的安全凭据的类型。Indicates the type of security credentials to use in order to secure the cluster. 有效值为 "None/X509/Windows"Valid values are "None/X509/Windows"
ClusterIdentitiesClusterIdentities string,默认值为“”string, default is "" 动态Dynamic 群集节点的 Windows 标识;用于群集成员身份授权。Windows identities of cluster nodes; used for cluster membership authorization. 该参数是以逗号分隔的列表;每个条目是域帐户名或组名称It is a comma-separated list; each entry is a domain account name or group name
ClusterSpnClusterSpn string,默认值为“”string, default is "" 不允许Not Allowed 群集的服务主体名称;结构以单个域用户(gMSA/域用户帐户)运行。Service principal name of the cluster; when fabric runs as a single domain user (gMSA/domain user account). 该参数是租用侦听器和 fabric.exe 中的侦听器的 SPN:联合侦听器、内部复制侦听器、运行时服务侦听器和命名网关侦听器。It is the SPN of lease listeners and listeners in fabric.exe: federation listeners; internal replication listeners; runtime service listener and naming gateway listener. 结构以计算机帐户运行时该参数应留空;在这种情况下从侦听器传输地址连接端计算侦听器 SPN。This should be left empty when fabric runs as machine accounts; in which case connecting side compute listener SPN from listener transport address.
CrlCheckingFlagCrlCheckingFlag uint,默认值为 0x40000000uint, default is 0x40000000 动态Dynamic 默认证书链验证标记;可能被特定于组件的标记重写;例如 Federation/X509CertChainFlags 0x10000000 CERT_CHAIN_REVOCATION_CHECK_END_CERT 0x20000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN 0x40000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN_EXCLUDE_ROOT 0x80000000 CERT_CHAIN_REVOCATION_CHECK_CACHE_ONLY 设置为 0 会禁用 CRL 检查,支持值的完整列表由 CertGetCertificateChain 的 dwFlags 记录: https://msdn.microsoft.com/library/windows/desktop/aa376078(v=vs.85).aspxDefault certificate chain validation flag; may be overridden by component-specific flag; e.g. Federation/X509CertChainFlags 0x10000000 CERT_CHAIN_REVOCATION_CHECK_END_CERT 0x20000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN 0x40000000 CERT_CHAIN_REVOCATION_CHECK_CHAIN_EXCLUDE_ROOT 0x80000000 CERT_CHAIN_REVOCATION_CHECK_CACHE_ONLY Setting to 0 disables CRL checking Full list of supported values is documented by dwFlags of CertGetCertificateChain: https://msdn.microsoft.com/library/windows/desktop/aa376078(v=vs.85).aspx
CrlDisablePeriodCrlDisablePeriod TimeSpan,默认值为 Common::TimeSpan::FromMinutes(15)TimeSpan, default is Common::TimeSpan::FromMinutes(15) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 遇到脱机错误后给定证书的 CRL 检查的禁用时间;是否可以忽略 CRL 脱机错误。How long CRL checking is disabled for a given certificate after encountering offline error; if CRL offline error can be ignored.
CrlOfflineHealthReportTtlCrlOfflineHealthReportTtl TimeSpan,默认值为 Common::TimeSpan::FromMinutes(1440)TimeSpan, default is Common::TimeSpan::FromMinutes(1440) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds.
DisableFirewallRuleForDomainProfileDisableFirewallRuleForDomainProfile bool,默认值为 TRUEbool, default is TRUE 静态Static 指示是否不应对域配置文件启用防火墙规则Indicates if firewall rule should not be enabled for domain profile
DisableFirewallRuleForPrivateProfileDisableFirewallRuleForPrivateProfile bool,默认值为 TRUEbool, default is TRUE 静态Static 指示是否不应对专用配置文件启用防火墙规则Indicates if firewall rule should not be enabled for private profile
DisableFirewallRuleForPublicProfileDisableFirewallRuleForPublicProfile bool,默认值为 TRUEbool, default is TRUE 静态Static 指示是否不应对公共配置文件启用防火墙规则Indicates if firewall rule should not be enabled for public profile
EnforceLinuxMinTlsVersionEnforceLinuxMinTlsVersion bool,默认值为 FALSEbool, default is FALSE 静态Static 如果设置为 true,则仅支持 TLS 版本 1.2+。If set to true; only TLS version 1.2+ is supported. 如果为 false,则支持更早期的 TLS 版本。If false; support earlier TLS versions. 仅适用于 LinuxApplies to Linux only
EnforcePrevalidationOnSecurityChangesEnforcePrevalidationOnSecurityChanges bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 用于控制检测到安全设置更改时群集升级行为的标志。Flag controlling the behavior of cluster upgrade upon detecting changes of its security settings. 如果设置为“true”,群集升级将尝试确保与任何显示规则匹配的证书中至少有一个可传递相应的验证规则。If set to 'true', the cluster upgrade will attempt to ensure that at least one of the certificates matching any of the presentation rules can pass a corresponding validation rule. 在向任何节点应用新设置之前都会执行预验证,但仅在启动升级时,在托管群集管理器服务的主要副本的节点上运行。The pre-validation is executed before the new settings are applied to any node, but runs only on the node hosting the primary replica of the Cluster Manager service at the time of initiating the upgrade. 当前默认设置为“false”;自版本 7.1 开始,新的 Azure Service Fabric 群集的此设置将改为“true”。The default is currently set to 'false'; starting with release 7.1, the setting will be set to 'true' for new Azure Service Fabric clusters.
FabricHostSpnFabricHostSpn string,默认值为“”string, default is "" 静态Static FabricHost 的服务主体名称;结构作为单个域用户(gMSA/域用户帐户)运行并且 FabricHost 在计算机帐户下运行时。Service principal name of FabricHost; when fabric runs as a single domain user (gMSA/domain user account) and FabricHost runs under machine account. 该参数是 FabricHost 的 IPC 侦听器的 SPN;由于 FabricHost 在计算机帐户下运行,所以该参数默认留空It is the SPN of IPC listener for FabricHost; which by default should be left empty since FabricHost runs under machine account
IgnoreCrlOfflineErrorIgnoreCrlOfflineError bool,默认值为 FALSEbool, default is FALSE 动态Dynamic 服务器端验证传入客户端证书时,是否忽略 CRL 脱机错误Whether to ignore CRL offline error when server-side verifies incoming client certificates
IgnoreSvrCrlOfflineErrorIgnoreSvrCrlOfflineError bool,默认值为 TRUEbool, default is TRUE 动态Dynamic 客户端验证传入服务器证书时,是否忽略 CRL 脱机错误;默认值为 true。Whether to ignore CRL offline error when client side verifies incoming server certificates; default to true. 具有吊销的服务器证书的攻击需要破坏 DNS;比具有吊销的客户端证书的攻击更难。Attacks with revoked server certificates require compromising DNS; harder than with revoked client certificates.
ServerAuthCredentialTypeServerAuthCredentialType string,默认值为“None”string, default is "None" 静态Static 指示为保护 FabricClient 和群集之间的通信而使用的安全凭据的类型。Indicates the type of security credentials to use in order to secure the communication between FabricClient and the Cluster. 有效值为 "None/X509/Windows"Valid values are "None/X509/Windows"
ServerCertThumbprintsServerCertThumbprints string,默认值为“”string, default is "" 动态Dynamic 群集用于与客户端通信的服务器证书的指纹;客户端使用该指纹对群集进行身份验证。Thumbprints of server certificates used by cluster to talk to clients; clients use this to authenticate the cluster. 该参数是以逗号分隔的名称列表。It is a comma-separated name list.
SettingsX509StoreNameSettingsX509StoreName string,默认值为“MY”string, default is "MY" 动态Dynamic 结构用于保护配置的 X509 证书存储X509 certificate store used by fabric for configuration protection
UseClusterCertForIpcServerTlsSecurityUseClusterCertForIpcServerTlsSecurity bool,默认值为 FALSEbool, default is FALSE 静态Static 是否使用群集证书保护 IPC 服务器 TLS 传输单元Whether to use cluster certificate to secure IPC Server TLS transport unit
X509FolderX509Folder string,默认值为 /var/lib/waagentstring, default is /var/lib/waagent 静态Static X509 证书和私钥所在的文件夹Folder where X509 certificates and private keys are located
TLS1_2_CipherListTLS1_2_CipherList stringstring 静态Static 对于 TLS1.2 及更低版本,如果设置为非空字符串,则替代受支持的密码列表。If set to a non-empty string; overrides the supported cipher list for TLS1.2 and below. 请参阅“openssl-ciphers”文档来检索受支持的密码列表和列表格式。TLS1.2 的强密码列表的示例为:“ECDHE-ECDSA-AES256-GCM-SHA384:ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-RSA-AES256-GCM-SHA384:ECDHE-RSA-AES-128-GCM-SHA256:ECDHE-ECDSA-AES256-CBC-SHA384:ECDHE-ECDSA-AES128-CBC-SHA256:ECDHE-RSA-AES256-CBC-SHA384:ECDHE-RSA-AES128-CBC-SHA256”。仅适用于 Linux。See the 'openssl-ciphers' documentation for retrieving the supported cipher list and the list format Example of strong cipher list for TLS1.2: "ECDHE-ECDSA-AES256-GCM-SHA384:ECDHE-ECDSA-AES128-GCM-SHA256:ECDHE-RSA-AES256-GCM-SHA384:ECDHE-RSA-AES-128-GCM-SHA256:ECDHE-ECDSA-AES256-CBC-SHA384:ECDHE-ECDSA-AES128-CBC-SHA256:ECDHE-RSA-AES256-CBC-SHA384:ECDHE-RSA-AES128-CBC-SHA256" Applies to Linux only.

Security/AdminClientX509NamesSecurity/AdminClientX509Names

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup X509NameMap,默认值为 NoneX509NameMap, default is None 动态Dynamic 这是“名称”和“值”对的列表。This is a list of "Name" and "Value" pair. 每个“名称”是为管理员客户端操作授权的 X509 证书的使用者公用名或 DnsName。Each "Name" is of subject common name or DnsName of X509 certificates authorized for admin client operations. 对于给定的“名称”,“值”是用于颁发者固定的证书指纹的逗号分隔列表,如果不为空,则管理员客户端证书的直接颁发者必须在列表中。For a given "Name", "Value" is a comma separate list of certificate thumbprints for issuer pinning, if not empty, the direct issuer of admin client certificates must be in the list.

Security/ClientAccessSecurity/ClientAccess

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ActivateNodeActivateNode string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于激活节点的安全性配置。Security configuration for activation a node.
CancelTestCommandCancelTestCommand string,默认值为“Admin”string, default is "Admin" 动态Dynamic 取消特定的 TestCommand(如果处于运行中)。Cancels a specific TestCommand - if it is in flight.
CodePackageControlCodePackageControl string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于重启代码包的安全性配置。Security configuration for restarting code packages.
CreateApplicationCreateApplication string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于应用程序创建的安全性配置。Security configuration for application creation.
CreateComposeDeploymentCreateComposeDeployment string,默认值为“Admin”string, default is "Admin" 动态Dynamic 创建由组合文件说明的组合部署Creates a compose deployment described by compose files
CreateGatewayResourceCreateGatewayResource string,默认值为“Admin”string, default is "Admin" 动态Dynamic 创建网关资源Create a gateway resource
CreateNameCreateName string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于命名 URI 创建的安全性配置。Security configuration for Naming URI creation.
CreateNetworkCreateNetwork string,默认值为“Admin”string, default is "Admin" 动态Dynamic 创建容器网络Creates a container network
CreateServiceCreateService string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于服务创建的安全性配置。Security configuration for service creation.
CreateServiceFromTemplateCreateServiceFromTemplate string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于通过模板创建服务的安全性配置。Security configuration for service creation from template.
CreateVolumeCreateVolume string,默认值为“Admin”string, default is "Admin" 动态Dynamic 创建卷Creates a volume
DeactivateNodeDeactivateNode string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于停用节点的安全性配置。Security configuration for deactivating a node.
DeactivateNodesBatchDeactivateNodesBatch string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于停用多个节点的安全性配置。Security configuration for deactivating multiple nodes.
删除Delete string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于映像存储区客户端删除操作的安全性配置。Security configurations for image store client delete operation.
DeleteApplicationDeleteApplication string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于应用程序删除的安全性配置。Security configuration for application deletion.
DeleteComposeDeploymentDeleteComposeDeployment string,默认值为“Admin”string, default is "Admin" 动态Dynamic 删除组合部署Deletes the compose deployment
DeleteGatewayResourceDeleteGatewayResource string,默认值为“Admin”string, default is "Admin" 动态Dynamic 删除网关资源Deletes a gateway resource
DeleteNameDeleteName string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于命名 URI 删除的安全性配置。Security configuration for Naming URI deletion.
DeleteNetworkDeleteNetwork string,默认值为“Admin”string, default is "Admin" 动态Dynamic 删除容器网络Deletes a container network
DeleteServiceDeleteService string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于服务删除的安全性配置。Security configuration for service deletion.
DeleteVolumeDeleteVolume string,默认值为“Admin”string, default is "Admin" 动态Dynamic 删除卷。Deletes a volume.
EnumeratePropertiesEnumerateProperties string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于枚举命名属性的安全性配置。Security configuration for Naming property enumeration.
EnumerateSubnamesEnumerateSubnames string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于枚举命名 URI 的安全性配置。Security configuration for Naming URI enumeration.
FileContentFileContent string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于传输映像存储客户端文件(群集外部)的安全性配置。Security configuration for image store client file transfer (external to cluster).
FileDownloadFileDownload string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动映像存储客户端文件下载(群集外部)的安全性配置。Security configuration for image store client file download initiation (external to cluster).
FinishInfrastructureTaskFinishInfrastructureTask string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于完成基础结构任务的安全性配置。Security configuration for finishing infrastructure tasks.
GetChaosReportGetChaosReport string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 获取给定时间范围内混沌的状态。Fetches the status of Chaos within a given time range.
GetClusterConfigurationGetClusterConfiguration string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 在分区上引入 GetClusterConfiguration。Induces GetClusterConfiguration on a partition.
GetClusterConfigurationUpgradeStatusGetClusterConfigurationUpgradeStatus string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 在分区上引入 GetClusterConfigurationUpgradeStatus。Induces GetClusterConfigurationUpgradeStatus on a partition.
GetFabricUpgradeStatusGetFabricUpgradeStatus string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于轮询群集升级状态的安全性配置。Security configuration for polling cluster upgrade status.
GetFolderSizeGetFolderSize string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于 FileStoreService 的获取文件夹大小的安全配置Security configuration for FileStoreService's getting folder size
GetNodeDeactivationStatusGetNodeDeactivationStatus string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于检查停用状态的安全性配置。Security configuration for checking deactivation status.
GetNodeTransitionProgressGetNodeTransitionProgress string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于获取节点转换命令进度的安全配置。Security configuration for getting progress on a node transition command.
GetPartitionDataLossProgressGetPartitionDataLossProgress string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 获取调用数据丢失 API 调用的进度。Fetches the progress for an invoke data loss api call.
GetPartitionQuorumLossProgressGetPartitionQuorumLossProgress string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 获取调用仲裁丢失 API 调用的进度。Fetches the progress for an invoke quorum loss api call.
GetPartitionRestartProgressGetPartitionRestartProgress string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 获取重启分区 API 调用的进度。Fetches the progress for a restart partition api call.
GetSecretsGetSecrets string,默认值为“Admin”string, default is "Admin" 动态Dynamic 获取机密值Get secret values
GetServiceDescriptionGetServiceDescription string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于长时间轮询服务通知和读取服务描述的安全性配置。Security configuration for long-poll service notifications and reading service descriptions.
GetStagingLocationGetStagingLocation string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于检索映像存储客户端暂存位置的安全性配置。Security configuration for image store client staging location retrieval.
GetStoreLocationGetStoreLocation string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于检索映像存储客户端存储位置的安全性配置。Security configuration for image store client store location retrieval.
GetUpgradeOrchestrationServiceStateGetUpgradeOrchestrationServiceState string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在每个分区上引发 GetUpgradeOrchestrationServiceStateInduces GetUpgradeOrchestrationServiceState on a partition
GetUpgradesPendingApprovalGetUpgradesPendingApproval string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引入 GetUpgradesPendingApproval。Induces GetUpgradesPendingApproval on a partition.
GetUpgradeStatusGetUpgradeStatus string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于轮询应用程序升级状态的安全性配置。Security configuration for polling application upgrade status.
InternalListInternalList string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于映像存储客户端文件列表操作(内部)的安全性配置。Security configuration for image store client file list operation (internal).
InvokeContainerApiInvokeContainerApi string,默认值为“Admin”string,default is "Admin" 动态Dynamic 调用容器 APIInvoke container API
InvokeInfrastructureCommandInvokeInfrastructureCommand string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于基础结构任务管理命令的安全性配置。Security configuration for infrastructure task management commands.
InvokeInfrastructureQueryInvokeInfrastructureQuery string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于查询基础结构任务的安全性配置。Security configuration for querying infrastructure tasks.
列出List string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于映像存储客户端文件列表操作的安全性配置。Security configuration for image store client file list operation.
MoveNextFabricUpgradeDomainMoveNextFabricUpgradeDomain string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于使用显式升级域恢复群集升级的安全性配置。Security configuration for resuming cluster upgrades with an explicit Upgrade Domain.
MoveNextUpgradeDomainMoveNextUpgradeDomain string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于使用显式升级域恢复应用程序升级的安全性配置。Security configuration for resuming application upgrades with an explicit Upgrade Domain.
MoveReplicaControlMoveReplicaControl string,默认值为“Admin”string, default is "Admin" 动态Dynamic 移动副本。Move replica.
NameExistsNameExists string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于检查命名 URI 存在的安全性配置。Security configuration for Naming URI existence checks.
NodeControlNodeControl string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动、停止和重启节点的安全性配置。Security configuration for starting; stopping; and restarting nodes.
NodeStateRemovedNodeStateRemoved string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于报告删的节点状态的安全性配置。Security configuration for reporting node state removed.
PingPing string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于客户端 ping 的安全性配置。Security configuration for client pings.
PredeployPackageToNodePredeployPackageToNode string,默认值为“Admin”string, default is "Admin" 动态Dynamic 预部署 API。Predeployment api.
PrefixResolveServicePrefixResolveService string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于基于投诉的服务前缀解析的安全配置。Security configuration for complaint-based service prefix resolution.
PropertyReadBatchPropertyReadBatch string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于命名属性读取操作的安全性配置。Security configuration for Naming property read operations.
PropertyWriteBatchPropertyWriteBatch string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于命名属性写入操作的安全性配置。Security configurations for Naming property write operations.
ProvisionApplicationTypeProvisionApplicationType string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于应用程序类型预配的安全性配置。Security configuration for application type provisioning.
ProvisionFabricProvisionFabric string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于预配 MSI 和/或群集清单的安全性配置。Security configuration for MSI and/or Cluster Manifest provisioning.
查询Query string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于查询的安全性配置。Security configuration for queries.
RecoverPartitionRecoverPartition string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复分区的安全性配置。Security configuration for recovering a partition.
RecoverPartitionsRecoverPartitions string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复多个分区的安全性配置。Security configuration for recovering partitions.
RecoverServicePartitionsRecoverServicePartitions string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复服务分区的安全性配置。Security configuration for recovering service partitions.
RecoverSystemPartitionsRecoverSystemPartitions string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复系统服务分区的安全性配置。Security configuration for recovering system service partitions.
RemoveNodeDeactivationsRemoveNodeDeactivations string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于在多个节点上还原停用操作的安全性配置。Security configuration for reverting deactivation on multiple nodes.
ReportFabricUpgradeHealthReportFabricUpgradeHealth string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复群集升级并提供当前升级进度的安全性配置。Security configuration for resuming cluster upgrades with the current upgrade progress.
ReportFaultReportFault string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于报告故障的安全性配置。Security configuration for reporting fault.
ReportHealthReportHealth string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于报告运行状况的安全性配置。Security configuration for reporting health.
ReportUpgradeHealthReportUpgradeHealth string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于恢复应用程序升级并提供当前升级进度的安全性配置。Security configuration for resuming application upgrades with the current upgrade progress.
ResetPartitionLoadResetPartitionLoad string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于 failoverUnit 的重置负载的安全性配置。Security configuration for reset load for a failoverUnit.
ResolveNameOwnerResolveNameOwner string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于解析命名 URI 所有者的安全性配置。Security configuration for resolving Naming URI owner.
ResolvePartitionResolvePartition string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于解析系统服务的安全性配置。Security configuration for resolving system services.
ResolveServiceResolveService string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于基于投诉的服务解析的安全配置。Security configuration for complaint-based service resolution.
ResolveSystemServiceResolveSystemService string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于解析系统服务的安全性配置Security configuration for resolving system services
RollbackApplicationUpgradeRollbackApplicationUpgrade string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于回滚应用程序升级的安全性配置。Security configuration for rolling back application upgrades.
RollbackFabricUpgradeRollbackFabricUpgrade string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于回滚群集升级的安全性配置。Security configuration for rolling back cluster upgrades.
ServiceNotificationsServiceNotifications string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于基于时间的服务通知的安全配置。Security configuration for event-based service notifications.
SetUpgradeOrchestrationServiceStateSetUpgradeOrchestrationServiceState string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引发 SetUpgradeOrchestrationServiceStateInduces SetUpgradeOrchestrationServiceState on a partition
StartApprovedUpgradesStartApprovedUpgrades string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引入 StartApprovedUpgrades。Induces StartApprovedUpgrades on a partition.
StartChaosStartChaos string,默认值为“Admin”string, default is "Admin" 动态Dynamic 启动混沌 - 如果尚未启动。Starts Chaos - if it is not already started.
StartClusterConfigurationUpgradeStartClusterConfigurationUpgrade string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引入 StartClusterConfigurationUpgrade。Induces StartClusterConfigurationUpgrade on a partition.
StartInfrastructureTaskStartInfrastructureTask string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动基础结构任务的安全性配置。Security configuration for starting infrastructure tasks.
StartNodeTransitionStartNodeTransition string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动节点转换的安全性配置。Security configuration for starting a node transition.
StartPartitionDataLossStartPartitionDataLoss string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引入数据丢失。Induces data loss on a partition.
StartPartitionQuorumLossStartPartitionQuorumLoss string,默认值为“Admin”string, default is "Admin" 动态Dynamic 在分区上引入仲裁丢失。Induces quorum loss on a partition.
StartPartitionRestartStartPartitionRestart string,默认值为“Admin”string, default is "Admin" 动态Dynamic 同时重启分区的部分或所有副本。Simultaneously restarts some or all the replicas of a partition.
StopChaosStopChaos string,默认值为“Admin”string, default is "Admin" 动态Dynamic 停止混沌 - 如果已启动。Stops Chaos - if it has been started.
ToggleVerboseServicePlacementHealthReportingToggleVerboseServicePlacementHealthReporting string,默认值为“Admin||User”string, default is "Admin||User" 动态Dynamic 用于切换详细服务放置运行状况报告的安全配置。Security configuration for Toggling Verbose ServicePlacement HealthReporting.
UnprovisionApplicationTypeUnprovisionApplicationType string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于取消设置应用程序类型的安全性配置。Security configuration for application type unprovisioning.
UnprovisionFabricUnprovisionFabric string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于取消设置 MSI 和/或群集清单的安全性配置。Security configuration for MSI and/or Cluster Manifest unprovisioning.
UnreliableTransportControlUnreliableTransportControl string,默认值为“Admin”string, default is "Admin" 动态Dynamic 添加和删除行为的不可靠传输。Unreliable Transport for adding and removing behaviors.
UpdateServiceUpdateService string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于服务更新的安全性配置。Security configuration for service updates.
UpgradeApplicationUpgradeApplication string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动或中断应用程序升级的安全性配置。Security configuration for starting or interrupting application upgrades.
UpgradeComposeDeploymentUpgradeComposeDeployment string,默认值为“Admin”string, default is "Admin" 动态Dynamic 升级组合部署Upgrades the compose deployment
UpgradeFabricUpgradeFabric string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于启动群集升级的安全性配置。Security configuration for starting cluster upgrades.
上载Upload string,默认值为“Admin”string, default is "Admin" 动态Dynamic 用于映像存储客户端上传操作的安全性配置。Security configuration for image store client upload operation.

Security/ClientCertificateIssuerStoresSecurity/ClientCertificateIssuerStores

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup IssuerStoreKeyValueMap,默认值为 NoneIssuerStoreKeyValueMap, default is None 动态Dynamic 客户端证书的 X509 颁发者证书存储;名称 = clientIssuerCN;值 = 逗号分隔的存储列表X509 issuer certificate stores for client certificates; Name = clientIssuerCN; Value = comma separated list of stores

Security/ClientX509NamesSecurity/ClientX509Names

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup X509NameMap,默认值为 NoneX509NameMap, default is None 动态Dynamic 这是“名称”和“值”对的列表。This is a list of "Name" and "Value" pair. 每个“名称”是为客户端操作授权的 X509 证书的使用者公用名或 DnsName。Each "Name" is of subject common name or DnsName of X509 certificates authorized for client operations. 对于给定的“名称”,“值”是用于颁发者固定的证书指纹的逗号分隔列表,如果不为空,则客户端证书的直接颁发者必须在列表中。For a given "Name", "Value" is a comma separate list of certificate thumbprints for issuer pinning, if not empty, the direct issuer of client certificates must be in the list.

Security/ClusterCertificateIssuerStoresSecurity/ClusterCertificateIssuerStores

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup IssuerStoreKeyValueMap,默认值为 NoneIssuerStoreKeyValueMap, default is None 动态Dynamic 群集证书的 X509 颁发者证书存储;名称 = clusterIssuerCN;值 = 逗号分隔的存储列表X509 issuer certificate stores for cluster certificates; Name = clusterIssuerCN; Value = comma separated list of stores

Security/ClusterX509NamesSecurity/ClusterX509Names

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup X509NameMap,默认值为 NoneX509NameMap, default is None 动态Dynamic 这是“名称”和“值”对的列表。This is a list of "Name" and "Value" pair. 每个“名称”是为群集操作授权的 X509 证书的使用者公用名或 DnsName。Each "Name" is of subject common name or DnsName of X509 certificates authorized for cluster operations. 对于给定的“名称”,“值”是用于颁发者固定的证书指纹的逗号分隔列表,如果不为空,则群集证书的直接颁发者必须在列表中。For a given "Name","Value" is a comma separate list of certificate thumbprints for issuer pinning, if not empty, the direct issuer of cluster certificates must be in the list.

Security/ServerCertificateIssuerStoresSecurity/ServerCertificateIssuerStores

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup IssuerStoreKeyValueMap,默认值为 NoneIssuerStoreKeyValueMap, default is None 动态Dynamic 服务器证书的 X509 颁发者证书存储;名称 = serverIssuerCN;值 = 逗号分隔的存储列表X509 issuer certificate stores for server certificates; Name = serverIssuerCN; Value = comma separated list of stores

Security/ServerX509NamesSecurity/ServerX509Names

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup X509NameMap,默认值为 NoneX509NameMap, default is None 动态Dynamic 这是“名称”和“值”对的列表。This is a list of "Name" and "Value" pair. 每个“名称”是为服务器操作授权的 X509 证书的使用者公用名或 DnsName。Each "Name" is of subject common name or DnsName of X509 certificates authorized for server operations. 对于给定的“名称”,“值”是用于颁发者固定的证书指纹的逗号分隔列表,如果不为空,则服务器证书的直接颁发者必须在列表中。For a given "Name", "Value" is a comma separate list of certificate thumbprints for issuer pinning, if not empty, the direct issuer of server certificates must be in the list.

设置Setup

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
ContainerNetworkNameContainerNetworkName string,默认值为“”string, default is "" 静态Static 设置容器网络时要使用的网络名称。The network name to use when setting up a container network.
ContainerNetworkSetupContainerNetworkSetup bool,Linux 的默认值为 FALSE,Windows 的默认值为 TRUEbool, default is FALSE (Linux) and default is TRUE (Windows) 静态Static 是否设置容器网络。Whether to set up a container network.
FabricDataRootFabricDataRoot StringString 不允许Not Allowed Service Fabric 数据根目录。Service Fabric data root directory. Azure 默认位置为 d:\svcfabDefault for Azure is d:\svcfab
FabricLogRootFabricLogRoot StringString 不允许Not Allowed Service Fabric 日志根目录。Service fabric log root directory. 这是放置 SF 日志和跟踪信息的位置。This is where SF logs and traces are placed.
NodesToBeRemovedNodesToBeRemoved string,默认值为“”string, default is "" 动态Dynamic 应在配置升级过程中删除的节点。The nodes which should be removed as part of configuration upgrade. (仅适用于独立部署)(Only for Standalone Deployments)
ServiceRunAsAccountNameServiceRunAsAccountName StringString 不允许Not Allowed 运行 Fabric 主机服务的帐户名称。The account name under which to run fabric host service.
SkipContainerNetworkResetOnRebootSkipContainerNetworkResetOnReboot bool,默认值为 FALSEbool, default is FALSE NotAllowedNotAllowed 是否在重启时跳过容器网络重置。Whether to skip resetting container network on reboot.
SkipFirewallConfigurationSkipFirewallConfiguration Bool,默认值为 falseBool, default is false 不允许Not Allowed 指定是否需要由系统设置防火墙设置。Specifies if firewall settings need to be set by the system or not. 仅当使用 Windows 防火墙时适用。This applies only if you are using windows firewall. 如果使用第三方防火墙,则必须打开端口以供系统和应用程序使用If you are using third party firewalls, then you must open the ports for the system and applications to use

TokenValidationServiceTokenValidationService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
提供程序Providers string,默认值为“DSTS”string, default is "DSTS" 静态Static 要启用的令牌验证提供程序的逗号分隔列表(有效的提供程序是:DSTS、AAD)。Comma separated list of token validation providers to enable (valid providers are: DSTS; AAD). 目前只能在任何时候启用单个提供程序。Currently only a single provider can be enabled at any time.

Trace/EtwTrace/Etw

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
LevelLevel Int,默认值为 4Int, default is 4 动态Dynamic 跟踪 etw 级别可以采用值 1、2、3、4。Trace etw level can take values 1, 2, 3, 4. 必须使跟踪级别保持在 4 才可受到支持To be supported you must keep the trace level at 4

TransactionalReplicatorTransactionalReplicator

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
BatchAcknowledgementIntervalBatchAcknowledgementInterval 时间以秒为单位,默认值为 0.015Time in seconds, default is 0.015 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 确定接收到操作后到发送回确认之前,复制器等待的时间。Determines the amount of time that the replicator waits after receiving an operation before sending back an acknowledgement. 在该时间段期间接收的其他操作将通过一条消息发送回其确认 -> 减少网络流量,同时可能降低复制器的吞吐量。Other operations received during this time period will have their acknowledgements sent back in a single message-> reducing network traffic but potentially reducing the throughput of the replicator.
MaxCopyQueueSizeMaxCopyQueueSize Uint,默认值为 16384Uint, default is 16384 静态Static 这是用于定义队列初始大小的最大值,该队列用于维护复制操作。This is the maximum value defines the initial size for the queue which maintains replication operations. 请注意,它必须是 2 的幂。Note that it must be a power of 2. 如果在运行时该队列增长到此大小,将限制主复制器和辅助复制器之间的操作。If during runtime the queue grows to this size operation will be throttled between the primary and secondary replicators.
MaxPrimaryReplicationQueueMemorySizeMaxPrimaryReplicationQueueMemorySize Uint,默认值为 0Uint, default is 0 静态Static 这是主复制队列的最大值(以字节为单位)。This is the maximum value of the primary replication queue in bytes.
MaxPrimaryReplicationQueueSizeMaxPrimaryReplicationQueueSize Uint,默认值为 8192Uint, default is 8192 静态Static 这是主复制队列中可以存在的最大操作数量。This is the maximum number of operations that could exist in the primary replication queue. 请注意,它必须是 2 的幂。Note that it must be a power of 2.
MaxReplicationMessageSizeMaxReplicationMessageSize Uint,默认值为 52428800Uint, default is 52428800 静态Static 复制操作的最大消息大小。Maximum message size of replication operations. 默认值为 50MB。Default is 50MB.
MaxSecondaryReplicationQueueMemorySizeMaxSecondaryReplicationQueueMemorySize Uint,默认值为 0Uint, default is 0 静态Static 这是辅助复制队列的最大值(以字节为单位)。This is the maximum value of the secondary replication queue in bytes.
MaxSecondaryReplicationQueueSizeMaxSecondaryReplicationQueueSize Uint,默认值为 16384Uint, default is 16384 静态Static 这是辅助复制队列中可以存在的最大操作数量。This is the maximum number of operations that could exist in the secondary replication queue. 请注意,它必须是 2 的幂。Note that it must be a power of 2.
ReplicatorAddressReplicatorAddress string,默认值为“localhost:0”string, default is "localhost:0" 静态Static 采用字符串形式 -'IP:Port' 的终结点,Windows Fabric 复制器将其用于与其他副本建立连接以发送/接收操作。The endpoint in form of a string -'IP:Port' which is used by the Windows Fabric Replicator to establish connections with other replicas in order to send/receive operations.

传输Transport

ParameterParameter 允许的值Allowed Values 升级策略Upgrade policy 指导或简短说明Guidance or Short Description
ConnectionOpenTimeoutConnectionOpenTimeout 时间跨度,默认值是 Common::TimeSpan::FromSeconds(60)TimeSpan, default is Common::TimeSpan::FromSeconds(60) 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. 正在传入和正在接收的端(包括安全模式下的安全协商)上连接设置的超时时间Time out for connection setup on both incoming and accepting side (including security negotiation in secure mode)
FrameHeaderErrorCheckingEnabledFrameHeaderErrorCheckingEnabled bool,默认值为 TRUEbool, default is TRUE 静态Static 有关对非安全模式下的帧标头执行错误检查的默认设置;组件设置会重写此值。Default setting for error checking on frame header in non-secure mode; component setting overrides this.
MessageErrorCheckingEnabledMessageErrorCheckingEnabled bool,默认值为 FALSEbool,default is FALSE 静态Static 有关对非安全模式下的消息标头和正文执行错误检查的默认设置;组件设置会重写此值。Default setting for error checking on message header and body in non-secure mode; component setting overrides this.
ResolveOptionResolveOption string,默认值为“unspecified”string, default is "unspecified" 静态Static 确定解析 FQDN 的方法。Determines how FQDN is resolved. 有效值为 "unspecified/ipv4/ipv6"。Valid values are "unspecified/ipv4/ipv6".
SendTimeoutSendTimeout TimeSpan,默认值为 Common::TimeSpan::FromSeconds(300)TimeSpan, default is Common::TimeSpan::FromSeconds(300) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 检测停滞连接时的发送超时。Send timeout for detecting stuck connection. 某个环境中的 TCP 故障报告不可靠。TCP failure reports are not reliable in some environment. 可能需要根据可用网络带宽和出站数据大小 (*MaxMessageSize/*SendQueueSizeLimit) 调整此值。This may need to be adjusted according to available network bandwidth and size of outbound data (*MaxMessageSize/*SendQueueSizeLimit).

UpgradeOrchestrationServiceUpgradeOrchestrationService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
AutoupgradeEnabledAutoupgradeEnabled Bool,默认值为 trueBool, default is true 静态Static 基于目标状态文件的自动轮训和升级操作。Automatic polling and upgrade action based on a goal-state file.
AutoupgradeInstallEnabledAutoupgradeInstallEnabled 布尔值,默认为 FALSEBool, default is FALSE 静态Static 基于目标状态文件自动轮询、预配和安装代码升级操作。Automatic polling, provisioning and install of code upgrade action based on a goal-state file.
GoalStateExpirationReminderInDaysGoalStateExpirationReminderInDays 整数,默认为 30int, default is 30 静态Static 设置剩余天数,在该天数过后应显示目标状态提醒。Sets the number of remaining days after which goal state reminder should be shown.
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 0Int, default is 0 静态Static UpgradeOrchestrationService 的 MinReplicaSetSize。The MinReplicaSetSize for UpgradeOrchestrationService.
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 静态Static UpgradeOrchestrationService 的 PlacementConstraints。The PlacementConstraints for UpgradeOrchestrationService.
QuorumLossWaitDurationQuorumLossWaitDuration 以秒为单位的时间,默认值为 MaxValueTime in seconds, default is MaxValue 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. UpgradeOrchestrationService 的 QuorumLossWaitDuration。The QuorumLossWaitDuration for UpgradeOrchestrationService.
ReplicaRestartWaitDurationReplicaRestartWaitDuration 以秒为单位的时间,默认值为 60 分钟Time in seconds, default is 60 minutes 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. UpgradeOrchestrationService 的 ReplicaRestartWaitDuration。The ReplicaRestartWaitDuration for UpgradeOrchestrationService.
StandByReplicaKeepDurationStandByReplicaKeepDuration 以秒为单位的时间,默认值为 60247 分钟Time in seconds, default is 60247 minutes 静态Static 指定以秒为单位的时间范围。Specify timespan in seconds. UpgradeOrchestrationService 的 StandByReplicaKeepDuration。The StandByReplicaKeepDuration for UpgradeOrchestrationService.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 0Int, default is 0 静态Static UpgradeOrchestrationService 的 TargetReplicaSetSize。The TargetReplicaSetSize for UpgradeOrchestrationService.
UpgradeApprovalRequiredUpgradeApprovalRequired Bool,默认值为 falseBool, default is false 静态Static 此设置可让升级代码需要管理员批准才能继续操作。Setting to make code upgrade require administrator approval before proceeding.

UpgradeServiceUpgradeService

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
BaseUrlBaseUrl string,默认值为“”string, default is "" 静态Static UpgradeService 的 BaseUrl。BaseUrl for UpgradeService.
ClusterIdClusterId string,默认值为“”string, default is "" 静态Static UpgradeService 的 ClusterId。ClusterId for UpgradeService.
CoordinatorTypeCoordinatorType string,默认值为“WUTest”string, default is "WUTest" 不允许Not Allowed UpgradeService 的 CoordinatorType。The CoordinatorType for UpgradeService.
MinReplicaSetSizeMinReplicaSetSize Int,默认值为 2Int, default is 2 不允许Not Allowed UpgradeService 的 MinReplicaSetSize。The MinReplicaSetSize for UpgradeService.
OnlyBaseUpgradeOnlyBaseUpgrade Bool,默认值为 falseBool, default is false 动态Dynamic UpgradeService 的 OnlyBaseUpgrade。OnlyBaseUpgrade for UpgradeService.
PlacementConstraintsPlacementConstraints string,默认值为“”string, default is "" 不允许Not Allowed 升级服务的 PlacementConstraints。The PlacementConstraints for Upgrade service.
PollIntervalInSecondsPollIntervalInSeconds 时间跨度,默认为 Common::TimeSpan::FromSeconds(60)Timespan, default is Common::TimeSpan::FromSeconds(60) 动态Dynamic 指定以秒为单位的时间范围。Specify timespan in seconds. 适用于 ARM 的 UpgradeService 轮询管理操作的时间间隔。The interval between UpgradeService poll for ARM management operations.
TargetReplicaSetSizeTargetReplicaSetSize Int,默认值为 3Int, default is 3 不允许Not Allowed UpgradeService 的 TargetReplicaSetSize。The TargetReplicaSetSize for UpgradeService.
TestCabFolderTestCabFolder string,默认值为“”string, default is "" 静态Static UpgradeService 的 TestCabFolder。TestCabFolder for UpgradeService.
X509FindTypeX509FindType string,默认值为“”string, default is "" 动态Dynamic UpgradeService 的 X509FindType。X509FindType for UpgradeService.
X509FindValueX509FindValue string,默认值为“”string, default is "" 动态Dynamic UpgradeService 的 X509FindValue。X509FindValue for UpgradeService.
X509SecondaryFindValueX509SecondaryFindValue string,默认值为“”string, default is "" 动态Dynamic UpgradeService 的 X509SecondaryFindValue。X509SecondaryFindValue for UpgradeService.
X509StoreLocationX509StoreLocation string,默认值为“”string, default is "" 动态Dynamic UpgradeService 的 X509StoreLocation。X509StoreLocation for UpgradeService.
X509StoreNameX509StoreName string,默认值为“My”string, default is "My" 动态Dynamic UpgradeService 的 X509StoreName。X509StoreName for UpgradeService.

UserServiceMetricCapacitiesUserServiceMetricCapacities

ParameterParameter 允许的值Allowed Values 升级策略Upgrade Policy 指导或简短说明Guidance or Short Description
PropertyGroupPropertyGroup UserServiceMetricCapacitiesMap,默认值为“无”UserServiceMetricCapacitiesMap, default is None 静态Static 用户服务资源管理限制的集合。由于它会影响 AutoDetection 逻辑,因此必须是静态的A collection of user services resource governance limits Needs to be static as it affects AutoDetection logic

后续步骤Next steps

有关详细信息,请参阅升级独立群集的配置For more information, see Upgrade the configuration of a standalone cluster.