You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+build()concurrency(int concurrency) connectionFactory(org.apache.activemq.ActiveMQConnectionFactory connectionFactory) metrics(CollectorMetrics metrics) sampler(CollectorSampler sampler) storage(StorageComponent storage) storage in class CollectorComponent.Buildersampler in class CollectorComponent.Buildermetrics in class CollectorComponent.Builderbuild in class CollectorComponent.BuilderCloseable, AutoCloseablestatic final class start in class CollectorComponentclose in interface AutoCloseableclose in interface Closeableclose in class ComponentIOExceptionstatic ActiveMQCollector.Builderbuilder()concurrency(int concurrency) connectionFactory(org.apache.activemq.ActiveMQConnectionFactory connectionFactory) metrics(CollectorMetrics metrics) sampler(CollectorSampler sampler) storage(StorageComponent storage) You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+bootstrapServers(String bootstrapServers) build()metrics(CollectorMetrics metrics) final KafkaCollector.Buildersampler(CollectorSampler sampler) storage(StorageComponent storage) streams(int streams) storage in class CollectorComponent.Buildersampler in class CollectorComponent.Buildermetrics in class CollectorComponent.BuilderFor example: Only consume spans since you connected by setting the below. + +
+ Map<String, String> overrides = new LinkedHashMap<>();
+ overrides.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "latest");
+ builder.overrides(overrides);
+ ConsumerConfigbuild in class CollectorComponent.BuilderCloseable, AutoCloseablespan consumer.
+
+ This collector uses a Kafka 0.10+ consumer.
static final class start in class CollectorComponentbootstrapServers(String bootstrapServers) static KafkaCollector.Builderbuilder()metrics(CollectorMetrics metrics) final KafkaCollector.Buildersampler(CollectorSampler sampler) storage(StorageComponent storage) streams(int streams) You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+build()concurrency(int concurrency) connectionFactory(com.rabbitmq.client.ConnectionFactory connectionFactory) metrics(CollectorMetrics metrics) sampler(CollectorSampler sampler) storage(StorageComponent storage) storage in class CollectorComponent.Buildersampler in class CollectorComponent.Buildermetrics in class CollectorComponent.Builderbuild in class CollectorComponent.BuilderCloseable, AutoCloseablestatic final class start in class CollectorComponentclose in interface AutoCloseableclose in interface Closeableclose in class ComponentIOExceptionstatic RabbitMQCollector.Builderbuilder()concurrency(int concurrency) connectionFactory(com.rabbitmq.client.ConnectionFactory connectionFactory) metrics(CollectorMetrics metrics) sampler(CollectorSampler sampler) storage(StorageComponent storage) You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+The Overview page is the front page of this API document and provides a list of all packages with a summary for each. This page can also contain an overall description of the set of packages.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
Each serializable or externalizable class has a description of its serialization fields and methods. This information is of interest to those who implement rather than use the API. While there is no link in the navigation bar, you can get to this information by going to any serialized class and clicking "Serialized Form" in the "See Also" section of the class description.
+The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+ScribeCollector.start() was called.port is already in use.IOExceptionClassNotFoundExceptionIOExceptionIOExceptionClassNotFoundExceptionIOExceptionIOExceptionClassNotFoundExceptionIOExceptionResultCode success+
ResultCodebuild()metrics(CollectorMetrics metrics) port(int port) sampler(CollectorSampler sampler) storage(StorageComponent storage) storage in class CollectorComponent.Buildermetrics in class CollectorComponent.Buildersampler in class CollectorComponent.Builderbuild in class CollectorComponent.BuilderCloseable, AutoCloseableasynchronous span consumer.static final class port is already in use.start in class CollectorComponentstart() was called.metrics(CollectorMetrics metrics) static ScribeCollector.BuildernewBuilder()port(int port) sampler(CollectorSampler sampler) storage(StorageComponent storage) Serializable, Comparable<LogEntry._Fields>, org.apache.thrift.TFieldIdEnumstatic LogEntry._FieldsfindByName(String name) static LogEntry._FieldsfindByThriftId(int fieldId) static LogEntry._FieldsfindByThriftIdOrThrow(int fieldId) shortstatic LogEntry._Fieldsstatic LogEntry._Fields[]values()name - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullgetThriftFieldId in interface org.apache.thrift.TFieldIdEnumgetFieldName in interface org.apache.thrift.TFieldIdEnumSerializable, Cloneable, Comparable<LogEntry>, org.apache.thrift.TBase<LogEntry,LogEntry._Fields> , org.apache.thrift.TSerializablestatic enum static final Map<LogEntry._Fields,org.apache.thrift.meta_data.FieldMetaData> voidclear()intdeepCopy()booleanbooleanfieldForId(int fieldId) getFieldValue(LogEntry._Fields field) inthashCode()booleanisSet(LogEntry._Fields field) booleanbooleanvoidread(org.apache.thrift.protocol.TProtocol iprot) setCategory(String category) voidsetCategoryIsSet(boolean value) voidsetFieldValue(LogEntry._Fields field,
+ Object value) setMessage(String message) voidsetMessageIsSet(boolean value) toString()voidvoidvoidvalidate()voidwrite(org.apache.thrift.protocol.TProtocol oprot) deepCopy in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> clear in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> setFieldValue in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> getFieldValue in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> isSet in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> compareTo in interface Comparable<LogEntry>fieldForId in interface org.apache.thrift.TBase<LogEntry,LogEntry._Fields> read in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionwrite in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionorg.apache.thrift.TExceptionSerializable, Comparable<ResultCode>, org.apache.thrift.TEnumstatic ResultCodefindByValue(int value) intgetValue()static ResultCodestatic ResultCode[]values()name - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullgetValue in interface org.apache.thrift.TEnumorg.apache.thrift.async.TAsyncClientFactory<Scribe.AsyncClient>Factory(org.apache.thrift.async.TAsyncClientManager clientManager,
+ org.apache.thrift.protocol.TProtocolFactory protocolFactory) getAsyncClient(org.apache.thrift.transport.TNonblockingTransport transport) getAsyncClient in interface org.apache.thrift.async.TAsyncClientFactory<Scribe.AsyncClient>org.apache.thrift.async.TAsyncMethodCall.Stateclient, transportLog_call(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler,
+ org.apache.thrift.async.TAsyncClient client,
+ org.apache.thrift.protocol.TProtocolFactory protocolFactory,
+ org.apache.thrift.transport.TNonblockingTransport transport) voidwrite_args(org.apache.thrift.protocol.TProtocol prot) getClient, getFrameBuffer, getSequenceId, getStartTime, getState, getTimeoutTimestamp, hasTimeout, isFinished, onError, prepareMethodCallorg.apache.thrift.TExceptionwrite_args in class org.apache.thrift.async.TAsyncMethodCall<ResultCode>org.apache.thrift.TExceptiongetResult in class org.apache.thrift.async.TAsyncMethodCall<ResultCode>org.apache.thrift.TExceptionScribe.AsyncIfacestatic class static class ___currentMethod, ___manager, ___protocolFactory, ___transportAsyncClient(org.apache.thrift.protocol.TProtocolFactory protocolFactory,
+ org.apache.thrift.async.TAsyncClientManager clientManager,
+ org.apache.thrift.transport.TNonblockingTransport transport) voidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) checkReady, getError, getProtocolFactory, getTimeout, hasError, hasTimeout, onComplete, onError, setTimeoutLog in interface Scribe.AsyncIfaceorg.apache.thrift.TExceptionScribe.AsyncClientvoidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) org.apache.thrift.TExceptionorg.apache.thrift.async.AsyncMethodCallback<ResultCode>getResultHandler(org.apache.thrift.server.AbstractNonblockingServer.AsyncFrameBuffer fb,
+ int seqid) protected booleanisOneway()voidstart(I iface,
+ Scribe.Log_args args,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) getMethodName, sendResponsegetEmptyArgsInstance in class org.apache.thrift.AsyncProcessFunction<I extends Scribe.AsyncIface,Scribe.Log_args,ResultCode> getResultHandler in class org.apache.thrift.AsyncProcessFunction<I extends Scribe.AsyncIface,Scribe.Log_args,ResultCode> isOneway in class org.apache.thrift.AsyncProcessFunction<I extends Scribe.AsyncIface,Scribe.Log_args,ResultCode> start in class org.apache.thrift.AsyncProcessFunction<I extends Scribe.AsyncIface,Scribe.Log_args,ResultCode> org.apache.thrift.TExceptionorg.apache.thrift.TAsyncProcessor, org.apache.thrift.TProcessorstatic class Scribe.AsyncProcessor.Log<I extends Scribe.AsyncIface>LOGGER AsyncProcessor(I iface) protected AsyncProcessor(I iface,
+ Map<String,org.apache.thrift.AsyncProcessFunction<I,? extends org.apache.thrift.TBase,?>> processMap) getProcessMapView, process, processorg.apache.thrift.TServiceClientFactory<Scribe.Client>getClient in interface org.apache.thrift.TServiceClientFactory<Scribe.Client>getClient in interface org.apache.thrift.TServiceClientFactory<Scribe.Client>Scribe.Ifaceiprot_, oprot_, seqid_getInputProtocol, getOutputProtocol, receiveBase, sendBase, sendBaseOnewayLog in interface Scribe.Ifaceorg.apache.thrift.TExceptionorg.apache.thrift.TExceptionorg.apache.thrift.TExceptionScribe.Clientorg.apache.thrift.TExceptionSerializable, Comparable<Scribe.Log_args._Fields>, org.apache.thrift.TFieldIdEnumstatic Scribe.Log_args._FieldsfindByName(String name) static Scribe.Log_args._FieldsfindByThriftId(int fieldId) static Scribe.Log_args._FieldsfindByThriftIdOrThrow(int fieldId) shortstatic Scribe.Log_args._Fieldsstatic Scribe.Log_args._Fields[]values()name - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullgetThriftFieldId in interface org.apache.thrift.TFieldIdEnumgetFieldName in interface org.apache.thrift.TFieldIdEnumSerializable, Cloneable, Comparable<Scribe.Log_args>, org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> , org.apache.thrift.TSerializablestatic enum static final Map<Scribe.Log_args._Fields,org.apache.thrift.meta_data.FieldMetaData> voidaddToMessages(LogEntry elem) voidclear()intcompareTo(Scribe.Log_args other) deepCopy()booleanbooleanequals(Scribe.Log_args that) fieldForId(int fieldId) intinthashCode()booleanisSet(Scribe.Log_args._Fields field) booleanvoidread(org.apache.thrift.protocol.TProtocol iprot) voidsetFieldValue(Scribe.Log_args._Fields field,
+ Object value) setMessages(List<LogEntry> messages) voidsetMessagesIsSet(boolean value) toString()voidvoidvalidate()voidwrite(org.apache.thrift.protocol.TProtocol oprot) deepCopy in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> clear in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> setFieldValue in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> getFieldValue in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> isSet in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> compareTo in interface Comparable<Scribe.Log_args>fieldForId in interface org.apache.thrift.TBase<Scribe.Log_args,Scribe.Log_args._Fields> read in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionwrite in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionorg.apache.thrift.TExceptionSerializable, Comparable<Scribe.Log_result._Fields>, org.apache.thrift.TFieldIdEnumstatic Scribe.Log_result._FieldsfindByName(String name) static Scribe.Log_result._FieldsfindByThriftId(int fieldId) static Scribe.Log_result._FieldsfindByThriftIdOrThrow(int fieldId) shortstatic Scribe.Log_result._Fieldsstatic Scribe.Log_result._Fields[]values()ResultCodename - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullgetThriftFieldId in interface org.apache.thrift.TFieldIdEnumgetFieldName in interface org.apache.thrift.TFieldIdEnumSerializable, Cloneable, Comparable<Scribe.Log_result>, org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> , org.apache.thrift.TSerializablestatic enum static final Map<Scribe.Log_result._Fields,org.apache.thrift.meta_data.FieldMetaData> Log_result(ResultCode success) Log_result(Scribe.Log_result other) voidclear()intcompareTo(Scribe.Log_result other) deepCopy()booleanbooleanequals(Scribe.Log_result that) fieldForId(int fieldId) inthashCode()booleanisSet(Scribe.Log_result._Fields field) booleanvoidread(org.apache.thrift.protocol.TProtocol iprot) voidsetFieldValue(Scribe.Log_result._Fields field,
+ Object value) setSuccess(ResultCode success) voidsetSuccessIsSet(boolean value) toString()voidvoidvalidate()voidwrite(org.apache.thrift.protocol.TProtocol oprot) ResultCodedeepCopy in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> clear in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> ResultCodeResultCodesetFieldValue in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> getFieldValue in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> isSet in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> compareTo in interface Comparable<Scribe.Log_result>fieldForId in interface org.apache.thrift.TBase<Scribe.Log_result,Scribe.Log_result._Fields> read in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionwrite in interface org.apache.thrift.TSerializableorg.apache.thrift.TExceptionorg.apache.thrift.TExceptiongetResult(I iface,
+ Scribe.Log_args args) protected booleanisOneway()protected booleangetMethodName, processgetEmptyArgsInstance in class org.apache.thrift.ProcessFunction<I extends Scribe.Iface,Scribe.Log_args> isOneway in class org.apache.thrift.ProcessFunction<I extends Scribe.Iface,Scribe.Log_args> rethrowUnhandledExceptions in class org.apache.thrift.ProcessFunction<I extends Scribe.Iface,Scribe.Log_args> getResult in class org.apache.thrift.ProcessFunction<I extends Scribe.Iface,Scribe.Log_args> org.apache.thrift.TExceptionorg.apache.thrift.TProcessorstatic class Scribe.Processor.Log<I extends Scribe.Iface>getProcessMapView, processclone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, waitprocessstatic class static interface static class Scribe.AsyncProcessor<I extends Scribe.AsyncIface>static class static interface static class static class static class Scribe.Processor<I extends Scribe.Iface>static final Map<LogEntry._Fields,org.apache.thrift.meta_data.FieldMetaData> metaDataMapfieldForId(int fieldId) static LogEntry._FieldsfindByName(String name) static LogEntry._FieldsfindByThriftId(int fieldId) static LogEntry._FieldsfindByThriftIdOrThrow(int fieldId) static LogEntry._Fieldsstatic LogEntry._Fields[]values()getFieldValue(LogEntry._Fields field) booleanisSet(LogEntry._Fields field) voidsetFieldValue(LogEntry._Fields field,
+ Object value) deepCopy()setCategory(String category) setMessage(String message) getMessages()getMessagesIterator()voidaddToMessages(LogEntry elem) intbooleanvoidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) voidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) voidsetMessages(List<LogEntry> messages) Log_call(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler,
+ org.apache.thrift.async.TAsyncClient client,
+ org.apache.thrift.protocol.TProtocolFactory protocolFactory,
+ org.apache.thrift.transport.TNonblockingTransport transport) static ResultCodefindByValue(int value) getResult()getSuccess()recv_Log()static ResultCodestatic ResultCode[]values()org.apache.thrift.async.AsyncMethodCallback<ResultCode>getResultHandler(org.apache.thrift.server.AbstractNonblockingServer.AsyncFrameBuffer fb,
+ int seqid) voidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) voidLog(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) voidstart(I iface,
+ Scribe.Log_args args,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) Log_call(List<LogEntry> messages,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler,
+ org.apache.thrift.async.TAsyncClient client,
+ org.apache.thrift.protocol.TProtocolFactory protocolFactory,
+ org.apache.thrift.transport.TNonblockingTransport transport) getAsyncClient(org.apache.thrift.transport.TNonblockingTransport transport) static class Scribe.AsyncProcessor<I extends Scribe.AsyncIface>static class Scribe.AsyncProcessor.Log<I extends Scribe.AsyncIface>static class Scribe.Processor<I extends Scribe.Iface>static class Scribe.Processor.Log<I extends Scribe.Iface>static final Map<Scribe.Log_args._Fields,org.apache.thrift.meta_data.FieldMetaData> metaDataMapfieldForId(int fieldId) static Scribe.Log_args._FieldsfindByName(String name) static Scribe.Log_args._FieldsfindByThriftId(int fieldId) static Scribe.Log_args._FieldsfindByThriftIdOrThrow(int fieldId) static Scribe.Log_args._Fieldsstatic Scribe.Log_args._Fields[]values()getFieldValue(Scribe.Log_args._Fields field) booleanisSet(Scribe.Log_args._Fields field) voidsetFieldValue(Scribe.Log_args._Fields field,
+ Object value) deepCopy()getEmptyArgsInstance()getEmptyArgsInstance()setMessages(List<LogEntry> messages) intcompareTo(Scribe.Log_args other) booleanequals(Scribe.Log_args that) getResult(I iface,
+ Scribe.Log_args args) voidstart(I iface,
+ Scribe.Log_args args,
+ org.apache.thrift.async.AsyncMethodCallback<ResultCode> resultHandler) Log_args(Scribe.Log_args other) static final Map<Scribe.Log_result._Fields,org.apache.thrift.meta_data.FieldMetaData> metaDataMapfieldForId(int fieldId) static Scribe.Log_result._FieldsfindByName(String name) static Scribe.Log_result._FieldsfindByThriftId(int fieldId) static Scribe.Log_result._FieldsfindByThriftIdOrThrow(int fieldId) static Scribe.Log_result._Fieldsstatic Scribe.Log_result._Fields[]values()getFieldValue(Scribe.Log_result._Fields field) booleanisSet(Scribe.Log_result._Fields field) voidsetFieldValue(Scribe.Log_result._Fields field,
+ Object value) deepCopy()getResult(I iface,
+ Scribe.Log_args args) setSuccess(ResultCode success) intcompareTo(Scribe.Log_result other) booleanequals(Scribe.Log_result that) Log_result(Scribe.Log_result other) You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+CollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int).CollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int).Collector.acceptSpans(byte[], BytesDecoder, Callback), except using a byte buffer.samples spans to reduce load on the
+ storage system.SpanConsumer.accept(List) queued for storage} using
+ this component.build()metrics(CollectorMetrics metrics) sampler(CollectorSampler sampler) storage(StorageComponent storage) Callbacks passed do not propagate to the storage layer. They only return success or failures + before storage is attempted. This ensures that calling threads are disconnected from storage + threads.
voidvoidvoidacceptSpans(byte[] serialized,
+ Callback<Void> callback) CollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int).voidacceptSpans(byte[] serializedSpans,
+ BytesDecoder<Span> decoder,
+ Callback<Void> callback) CollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int).voidacceptSpans(ByteBuffer encoded,
+ SpanBytesDecoder decoder,
+ Callback<Void> callback,
+ Executor executor) acceptSpans(byte[], BytesDecoder, Callback), except using a byte buffer.static Collector.BuildernewBuilder(Class<?> loggingClass) executor - the executor used to enqueue the storage request.acceptSpans(byte[], BytesDecoder, Callback), except using a byte buffer.CollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int). Do not call any other metrics callbacks as those are
+ handled internal to this method.serialized - not empty messageCollectorMetrics.incrementMessages(), and CollectorMetrics.incrementBytes(int). Do not call any other metrics callbacks as those are
+ handled internal to this method.serializedSpans - not empty messageabstract CollectorComponentbuild()abstract CollectorComponent.Buildermetrics(CollectorMetrics metrics) abstract CollectorComponent.Buildersampler(CollectorSampler sampler) samples spans to reduce load on the
+ storage system.abstract CollectorComponent.Builderstorage(StorageComponent storage) SpanConsumer.accept(List) queued for storage} using
+ this component.SpanConsumer.accept(List) queued for storage} using
+ this component.scoped to this transport. Defaults to no-op.samples spans to reduce load on the
+ storage system. Defaults to always sample.Closeable, AutoCloseableCall start() to start collecting spans.
abstract CollectorComponentstart()Many implementations block the calling thread until services are available.
InMemoryCollectorMetricsdecoding them into spans,
+ apply sampling, and queues them for storage.
+
+ Callbacks on this type are invoked by zipkin collectors to improve the visibility of the + system. A typical implementation will report metrics to a telemetry system for analysis and + reporting. + +
A span queried may be comprised of multiple spans collected. While instrumentation should + report complete spans, Instrumentation often patch the same span twice, ex adding annotations. + Also, RPC spans include at least 2 messages due to the client and the server reporting + separately. Finally, some storage components merge patches at ingest. For these reasons, you + should be cautious to alert on queryable spans vs stored spans, unless you control the + instrumentation in such a way that queryable spans/message is reliable. + +
The following relationships can be used to consider health of the tracing system. + +
+
Accepted messages -
+ Dropped messages. Alert when this is less than amount of
+ messages sent from instrumentation.Accepted spans - Dropped spans. Alert when this drops below the
+ collection-tier sample rate.
+ forTransport(String transportType) voidincrementBytes(int quantity) voidvoidvoidincrementSpans(int quantity) voidincrementSpansDropped(int quantity) For example, an implementation may by default report incremented
+ spans to the key "zipkin.collector.span.accepted". When metrics.forTransport("kafka"
+ is called, the counter would report to "zipkin.collector.kafka.span.accepted"
transportType - ex "http", "rabbitmq", "kafka"Note: this count should relate to the raw data structures, like json or thrift, and discount + compression, enveloping, etc.
Debug spans are always stored.
+
+ Accepts a percentage of trace ids by comparing their absolute value against a potentially
+ dynamic boundary. eg isSampled == abs(traceId) <= boundary
+
+
While idempotent, this implementation's sample rate won't exactly match the input rate because
+ trace ids are not perfectly distributed across 64bits. For example, tests have shown an error
+ rate of 3% when 100K trace ids are random.
protected abstract longboundary()static CollectorSamplercreate(float rate) booleantoString()rate - minimum sample rate is 0.0001, or 0.01% of tracesZipkin v1 allows storage-layer sampling, which can help prevent spikes in traffic from + overloading the system. Debug spans are always stored. + +
This uses only the lower 64 bits of the trace ID as instrumentation still send mixed trace + ID width.
hexTraceId - the lower 64 bits of the span's trace ID are checked against the boundarydebug - when true, always passes samplingCollectorMetricsNOOP_METRICSintbytes()voidclear()forTransport(String transportType) voidincrementBytes(int quantity) voidvoidvoidincrementSpans(int quantity) voidincrementSpansDropped(int quantity) intmessages()intintspans()intCollectorMetricsFor example, an implementation may by default report incremented
+ spans to the key "zipkin.collector.span.accepted". When metrics.forTransport("kafka"
+ is called, the counter would report to "zipkin.collector.kafka.span.accepted"
forTransport in interface CollectorMetricstransportType - ex "http", "rabbitmq", "kafka"CollectorMetricsincrementMessages in interface CollectorMetricsCollectorMetricsincrementMessagesDropped in interface CollectorMetricsCollectorMetricsNote: this count should relate to the raw data structures, like json or thrift, and discount + compression, enveloping, etc.
incrementBytes in interface CollectorMetricsCollectorMetricsincrementSpans in interface CollectorMetricsCollectorMetricsincrementSpansDropped in interface CollectorMetricsmetrics(CollectorMetrics metrics) static Collector.BuildernewBuilder(Class<?> loggingClass) sampler(CollectorSampler sampler) storage(StorageComponent storage) abstract CollectorComponent.Buildermetrics(CollectorMetrics metrics) abstract CollectorComponent.Buildersampler(CollectorSampler sampler) samples spans to reduce load on the
+ storage system.abstract CollectorComponent.Builderstorage(StorageComponent storage) SpanConsumer.accept(List) queued for storage} using
+ this component.abstract CollectorComponentbuild()abstract CollectorComponentstart()static final CollectorMetricsNOOP_METRICSforTransport(String transportType) metrics(CollectorMetrics metrics) abstract CollectorComponent.Buildermetrics(CollectorMetrics metrics) static final CollectorSamplerALWAYS_SAMPLEstatic CollectorSamplercreate(float rate) sampler(CollectorSampler sampler) abstract CollectorComponent.Buildersampler(CollectorSampler sampler) samples spans to reduce load on the
+ storage system.forTransport(String transportType) POST endpoints need to survive failures.ZipkinExtension.httpUrl().You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+POST endpoints need to survive failures.ZipkinExtension.httpUrl().POST endpoints need to survive failures. Besides simply not
+ starting the zipkin server, you can enqueue failures like this to test edge cases. For example,
+ that you log a failure when a 400 code is returned.static HttpFailurestatic HttpFailuresendErrorResponse(int code,
+ String body) org.junit.jupiter.api.extension.AfterEachCallback, org.junit.jupiter.api.extension.BeforeEachCallback, org.junit.jupiter.api.extension.ExtensionhttpUrl().
+
+ This can be used to test instrumentation. For example, you can POST spans directly to this + server. + +
See http://openzipkin.github.io/zipkin-api/#/
voidafterEach(org.junit.jupiter.api.extension.ExtensionContext extensionContext) voidbeforeEach(org.junit.jupiter.api.extension.ExtensionContext extensionContext) enqueueFailure(HttpFailure failure) inthttpUrl()voidshutdown()voidstart(int httpPort) storeSpans(List<Span> spans) For example, if you are testing what happens when instrumentation adds a child to a trace, + you'd add the parent here.
Ex. If you want to test that you don't repeatedly send bad data, you could send a 400 back. + +
+ zipkin.enqueueFailure(sendErrorResponse(400, "bad format"));
+ failure - type of failure the next call to the http endpoint responds withhttpPort - choose 0 to select an available portIOExceptionIOExceptionbeforeEach in interface org.junit.jupiter.api.extension.BeforeEachCallbackafterEach in interface org.junit.jupiter.api.extension.AfterEachCallbackstatic HttpFailuredisconnectDuringBody()static HttpFailuresendErrorResponse(int code,
+ String body) enqueueFailure(HttpFailure failure) enqueueFailure(HttpFailure failure) storeSpans(List<Span> spans) POST endpoints need to survive failures.httpUrl().POST endpoints need to survive failures.ZipkinExtension.httpUrl().You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+autocompleteCardinality, autocompleteKeys, autocompleteTtl, contactPoints, ensureSchema, indexFetchMultiplier, keyspace, localDc, maxTraceCols, password, searchEnabled, sslHostnameValidation, strictTraceId, username, useSslbuild()ensureSchema(boolean ensureSchema) sessionFactory(CassandraStorage.SessionFactory sessionFactory) autocompleteCardinality, autocompleteKeys, autocompleteTtl, contactPoints, indexFetchMultiplier, localDc, maxConnections, maxTraceCols, password, poolingOptions, searchEnabled, sslHostnameValidation, strictTraceId, username, useSslkeyspace in class zipkin2.storage.cassandra.internal.CassandraStorageBuilder<CassandraStorage.Builder>ensureSchema in class zipkin2.storage.cassandra.internal.CassandraStorageBuilder<CassandraStorage.Builder>build in class StorageComponent.Buildercom.datastax.oss.driver.api.core.CqlSessioncreate(CassandraStorage storage) Closeable, AutoCloseableQueries are logged to the category "com.datastax.oss.driver.api.core.cql.QueryLogger" when + debug or trace is enabled via SLF4J. Trace level includes bound values. + +
Schema is installed by default from "/zipkin2-schema.cql" + +
When StorageComponent.Builder.strictTraceId(boolean) is disabled, span and index data
+ are uniformly written with 64-bit trace ID length. When retrieving data, an extra "trace_id_high"
+ field clarifies if a 128-bit trace ID was sent.
static final class static interface check()voidclose()booleanstatic CassandraStorage.BuildertoString()traces()spanStore in class StorageComponenttraces in class StorageComponentserviceAndSpanNames in class StorageComponentautocompleteTags in class StorageComponentspanConsumer in class StorageComponentisOverCapacity in class StorageComponentensureSchema(boolean ensureSchema) static CassandraStorage.BuildernewBuilder()sessionFactory(CassandraStorage.SessionFactory sessionFactory) static final CassandraStorage.SessionFactoryDEFAULTsessionFactory(CassandraStorage.SessionFactory sessionFactory) com.datastax.oss.driver.api.core.CqlSessioncreate(CassandraStorage storage) WebClient.You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+WebClient.RejectedExecutionException when work is
+ rejected.Span.timestamp() are at or after
+ (now - lookback) in milliseconds.ElasticsearchStorage.close()abstract ElasticsearchStorage.BuilderautocompleteCardinality(int autocompleteCardinality) abstract ElasticsearchStorage.BuilderautocompleteKeys(List<String> autocompleteKeys) abstract ElasticsearchStorage.BuilderautocompleteTtl(int autocompleteTtl) abstract ElasticsearchStoragebuild()dateSeparator(char dateSeparator) abstract ElasticsearchStorage.BuilderensureTemplates(boolean ensureTemplates) abstract ElasticsearchStorage.BuilderflushOnWrites(boolean flushOnWrites) abstract ElasticsearchStorage.BuilderindexReplicas(int indexReplicas) abstract ElasticsearchStorage.BuilderindexShards(int indexShards) abstract ElasticsearchStorage.BuildernamesLookback(int namesLookback) Span.timestamp() are at or after
+ (now - lookback) in milliseconds.abstract ElasticsearchStorage.Builderabstract ElasticsearchStorage.BuildersearchEnabled(boolean searchEnabled) abstract ElasticsearchStorage.BuilderstrictTraceId(boolean strictTraceId) abstract ElasticsearchStorage.BuildertemplatePriority(Integer templatePriority) See https://www.elastic.co/guide/en/elasticsearch/reference/master/pipeline.html
Span.timestamp() are at or after
+ (now - lookback) in milliseconds. Defaults to 1 day (86400000).See https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-refresh.html
By default, spans with a timestamp falling on 2016/03/19 end up in the index + 'zipkin-span-2016-03-19'. When the date separator is '.', the index would be + 'zipkin-span-2016.03.19'. If the date separator is 0, there is no delimiter. Ex the index + would be 'zipkin-span-20160319'
Corresponds to index.number_of_shards
Corresponds to index.number_of_replicas
See https://www.elastic.co/guide/en/elasticsearch/reference/7.8/_index_template_and_settings_priority.html
strictTraceId in class StorageComponent.BuildersearchEnabled in class StorageComponent.BuilderautocompleteKeys in class StorageComponent.BuilderautocompleteTtl in class StorageComponent.BuilderautocompleteCardinality in class StorageComponent.Builderbuild in class StorageComponent.BuilderAutoCloseable, Closeable, Supplier<com.linecorp.armeria.client.WebClient>WebClient. This is needed because routinely, I/O occurs in
+ constructors and this can delay or cause startup to crash. For example, an underlying EndpointGroup could be delayed due to DNS, implicit api calls or health checks.default voidclose()com.linecorp.armeria.client.WebClientget()toString()close in interface AutoCloseableclose in interface CloseableCloseable, AutoCloseablestatic class static interface WebClient.check()voidclear()voidclose()abstract booleanabstract zipkin2.elasticsearch.internal.IndexNameFormatterbooleanRejectedExecutionException when work is
+ rejected.abstract intstatic ElasticsearchStorage.BuildernewBuilder(ElasticsearchStorage.LazyHttpClient lazyHttpClient) close()abstract Stringpipeline()abstract booleanfinal StringtoString()traces()version()close()spanStore in class StorageComponenttraces in class StorageComponentserviceAndSpanNames in class StorageComponentautocompleteTags in class StorageComponentspanConsumer in class StorageComponentIOExceptionRejectedExecutionException when work is
+ rejected. We also classify ResponseTimeoutException as a capacity related exception
+ even though capacity is not the only reason (timeout could also result from a misconfiguration
+ or a network problem).isOverCapacity in class StorageComponentComparable<ElasticsearchVersion>static final ElasticsearchVersionstatic final ElasticsearchVersionstatic final ElasticsearchVersionstatic final ElasticsearchVersionstatic final ElasticsearchVersionstatic final ElasticsearchVersioncompareTo in interface Comparable<ElasticsearchVersion>abstract ElasticsearchStorage.BuilderautocompleteCardinality(int autocompleteCardinality) abstract ElasticsearchStorage.BuilderautocompleteKeys(List<String> autocompleteKeys) abstract ElasticsearchStorage.BuilderautocompleteTtl(int autocompleteTtl) dateSeparator(char dateSeparator) abstract ElasticsearchStorage.BuilderensureTemplates(boolean ensureTemplates) abstract ElasticsearchStorage.BuilderflushOnWrites(boolean flushOnWrites) abstract ElasticsearchStorage.BuilderindexReplicas(int indexReplicas) abstract ElasticsearchStorage.BuilderindexShards(int indexShards) abstract ElasticsearchStorage.BuildernamesLookback(int namesLookback) Span.timestamp() are at or after
+ (now - lookback) in milliseconds.static ElasticsearchStorage.BuildernewBuilder(ElasticsearchStorage.LazyHttpClient lazyHttpClient) ElasticsearchStorage.close()abstract ElasticsearchStorage.Builderabstract ElasticsearchStorage.BuildersearchEnabled(boolean searchEnabled) abstract ElasticsearchStorage.BuilderstrictTraceId(boolean strictTraceId) abstract ElasticsearchStorage.BuildertemplatePriority(Integer templatePriority) static ElasticsearchStorage.BuildernewBuilder(ElasticsearchStorage.LazyHttpClient lazyHttpClient) ElasticsearchStorage.close()abstract ElasticsearchStoragebuild()static final ElasticsearchVersionV5_0static final ElasticsearchVersionV6_0static final ElasticsearchVersionV6_7static final ElasticsearchVersionV7_0static final ElasticsearchVersionV7_8static final ElasticsearchVersionV9_0version()intcompareTo(ElasticsearchVersion other) WebClient.WebClient.You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+autocompleteKeys(List<String> keys) build()datasource(DataSource datasource) listenerProvider(org.jooq.ExecuteListenerProvider listenerProvider) searchEnabled(boolean searchEnabled) settings(org.jooq.conf.Settings settings) strictTraceId(boolean strictTraceId) autocompleteCardinality, autocompleteTtlstrictTraceId in class StorageComponent.BuildersearchEnabled in class StorageComponent.BuilderautocompleteKeys in class StorageComponent.Builderbuild in class StorageComponent.BuilderCloseable, AutoCloseableisOverCapacityspanStore in class StorageComponenttraces in class StorageComponentserviceAndSpanNames in class StorageComponentautocompleteTags in class StorageComponentspanConsumer in class StorageComponentautocompleteKeys(List<String> keys) datasource(DataSource datasource) listenerProvider(org.jooq.ExecuteListenerProvider listenerProvider) static MySQLStorage.BuildernewBuilder()searchEnabled(boolean searchEnabled) settings(org.jooq.conf.Settings settings) strictTraceId(boolean strictTraceId) StorageComponent.Builder.autocompleteKeys(List) has values.SpanStore implementations that support dependency aggregation.SpanStore implementations that support dependency aggregation.searchEnabled ==
+ false.ServiceAndSpanNames.SpanStore.SpanStore implementations.StorageComponent integration tests.strictTraceId ==
+ false.Traces.You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+The Overview page is the front page of this API document and provides a list of all packages with a summary for each. This page can also contain an overall description of the set of packages.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+StorageComponent.Builder with parameters for the test being executed.Span.Kind.SERVER and the Span.localEndpoint() indicates the service.StorageComponent before each test rather than the test
+ class.StorageComponent.Builder.autocompleteKeys(List) has values.SpanStore implementations that support dependency aggregation.SpanStore implementations that support dependency aggregation.searchEnabled ==
+ false.ServiceAndSpanNames.SpanStore.SpanStore implementations.StorageComponent integration tests.strictTraceId ==
+ false.Traces.StorageComponent.Builder for connecting to the backend for the test.static final Endpointstatic final Spanstatic final longstatic final Endpointstatic final Endpointstatic final Endpointstatic final Span[]static final longstatic final Charsetstatic StringappendSuffix(String serviceName,
+ String serviceNameSuffix) static longstatic longmidnightUTC(long epochMillis) static SpannewClientSpan(String serviceNameSuffix) static Stringstatic Spanspan(long traceId) static Span.BuilderspanBuilder(String serviceNameSuffix) static longstatic EndpointsuffixServiceName(Endpoint endpoint,
+ String serviceNameSuffix) StorageComponent.Builder.autocompleteKeys(List) has values.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetTagsAndValues(org.junit.jupiter.api.TestInfo testInfo) protected voidignores_when_key_not_in_autocompleteTags(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionSpanStore implementations that support dependency aggregation. Subtypes
+ should create a connection to a real backend, even if that backend is in-process.
+
+ This is a replacement for zipkin.storage.DependenciesTest. There is some redundancy
+ as zipkin2.internal.DependencyLinkerTest also defines many of these tests. The redundancy
+ helps ensure integrated storage doesn't fail due to mismapping of data, for example.
static Map<Long,List<DependencyLink>> aggregateLinks(List<Span> spans) protected voidannotationNamedErrorIsntError(org.junit.jupiter.api.TestInfo testInfo) protected voidcanSearchForIntervalsBesidesToday(org.junit.jupiter.api.TestInfo testInfo) protected final voidStorageComponent.Builder with parameters for the test being executed.protected voiddependencies_headlessTrace(org.junit.jupiter.api.TestInfo testInfo) protected voiddependencies_loopback(org.junit.jupiter.api.TestInfo testInfo) protected voidduplicateAddress(org.junit.jupiter.api.TestInfo testInfo) protected voidempty()protected voidendTimeBeforeData(org.junit.jupiter.api.TestInfo testInfo) protected voidprotected voidendTsInsideTheTrace(org.junit.jupiter.api.TestInfo testInfo) protected voidgetDependencies(org.junit.jupiter.api.TestInfo testInfo) protected voidgetDependencies_linksMixedTraceId(org.junit.jupiter.api.TestInfo testInfo) protected voidgetDependenciesAllInstrumented(org.junit.jupiter.api.TestInfo testInfo) Span.Kind.SERVER and the Span.localEndpoint() indicates the service.protected voidinstrumentedClientAndServer(org.junit.jupiter.api.TestInfo testInfo) protected voidinstrumentedProducerAndConsumer(org.junit.jupiter.api.TestInfo testInfo) protected voidintermediateSpans(org.junit.jupiter.api.TestInfo testInfo) protected voidlookbackAfterData(org.junit.jupiter.api.TestInfo testInfo) protected voidlooksBackIndefinitely(org.junit.jupiter.api.TestInfo testInfo) protected voidmissingIntermediateSpan(org.junit.jupiter.api.TestInfo testInfo) protected voidnotInstrumentedClientAndServer(org.junit.jupiter.api.TestInfo testInfo) protected voidoneway(org.junit.jupiter.api.TestInfo testInfo) protected voidoneway_noClient(org.junit.jupiter.api.TestInfo testInfo) protected voidprocessDependencies(List<Span> spans) protected voidreplayOverwrites(org.junit.jupiter.api.TestInfo testInfo) protected voidspanKindIsNotRequiredWhenEndpointsArePresent(org.junit.jupiter.api.TestInfo testInfo) protected voidunnamedEndpointsAreSkipped(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionExceptionSpan.Kind.SERVER and the Span.localEndpoint() indicates the service.ExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionException+ span1: SR SS span2: intermediate call span3: CS SR SS CR: Dependency 1
Exception+ span1: SR SS span2: intermediate call span3: CS SR SS CR: Dependency 1
ExceptionExceptionExceptionExceptionSpanStore implementations that support dependency aggregation.
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
+
+ As these tests create a lot of data, implementations may wish to isolate them from other
+ integration tests such as ITDependencies
protected voidStorageComponent.Builder with parameters for the test being executed.protected booleanStorageComponent before each test rather than the test
+ class.protected voidprotected voidprocessDependencies(List<Span> spans) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent before each test rather than the test
+ class. Generally, tests will run faster if the storage is initialized as infrequently as
+ possibly while clearing data between runs, but for certain backends like Cassandra, it's
+ difficult to reliably clear data between runs and tends to be very slow anyways.initializeStoragePerTest in class ITStorage<T extends StorageComponent>ITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionsearchEnabled ==
+ false.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetRemoteServiceNames_isEmpty(org.junit.jupiter.api.TestInfo testInfo) protected voidgetServiceNames_isEmpty(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_isEmpty(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_indexDataReturnsNothing(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionServiceAndSpanNames.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetLocalServiceNames_includesLocalServiceName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetLocalServiceNames_noServiceName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetRemoteServiceNames(org.junit.jupiter.api.TestInfo testInfo) protected voidgetRemoteServiceNames_allReturned(org.junit.jupiter.api.TestInfo testInfo) protected voidgetRemoteServiceNames_dedupes(org.junit.jupiter.api.TestInfo testInfo) protected voidgetRemoteServiceNames_noRemoteServiceName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetRemoteServiceNames_serviceNameGoesLowercase(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_allReturned(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_dedupes(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_doesNotMapNameToRemoteServiceName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_noSpanName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetSpanNames_serviceNameGoesLowercase(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionSpanStore.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
protected voidprotected voidprotected final voidStorageComponent.Builder with parameters for the test being executed.protected voidconsumer_implementsCall_execute(org.junit.jupiter.api.TestInfo testInfo) protected voidconsumer_implementsCall_submit(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_absentWhenNoTimestamp(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_annotation(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_considersBitsAbove64bit(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_differentiateOnServiceName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_differentiatesDebugFromShared(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_duration(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_endTsAndLookback(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_endTsInsideTheTrace(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_filteringMatchesMostRecentTraces(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_groupsTracesTogether(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_lateDuration(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_limit(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_maxDuration(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_minDuration(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_multipleAnnotationsBecomeAndFilter(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_serviceNames(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_serviceNames_mixedTraceIdLength(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_spanName(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_spanName_mixedTraceIdLength(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_tags(org.junit.jupiter.api.TestInfo testInfo) protected voidnames_goLowercase(org.junit.jupiter.api.TestInfo testInfo) protected voidreadback_minimalErrorSpan(org.junit.jupiter.api.TestInfo testInfo) protected voidreadsBackLargeValues(org.junit.jupiter.api.TestInfo testInfo) protected voidspanWithProblematicData(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionNotably this guards empty tag values work
ExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionExceptionSpanStore implementations. Subtypes should create a connection to a
+ real backend, even if that backend is in-process.
+
+ As these tests create a lot of data, implementations may wish to isolate them from other
+ integration tests such as ITSpanStore
protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetTraces_manyTraces(org.junit.jupiter.api.TestInfo testInfo) protected booleanStorageComponent before each test rather than the test
+ class.protected voidtraceWithManySpans(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent before each test rather than the test
+ class. Generally, tests will run faster if the storage is initialized as infrequently as
+ possibly while clearing data between runs, but for certain backends like Cassandra, it's
+ difficult to reliably clear data between runs and tends to be very slow anyways.initializeStoragePerTest in class ITStorage<T extends StorageComponent>ITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionITAutocompleteTags, ITDependencies, ITDependenciesHeavy, ITSearchEnabledFalse, ITServiceAndSpanNames, ITSpanStore, ITSpanStoreHeavy, ITStrictTraceIdFalse, ITTracesStorageComponent integration tests.protected final voidprotected final voidprotected voidassertGetTraceReturns(String traceId,
+ List<Span> trace) protected voidassertGetTraceReturns(Span onlySpan) protected voidassertGetTraceReturnsEmpty(String traceId) protected voidassertGetTracesReturns(List<String> traceIds,
+ List<Span>... traces) protected voidassertGetTracesReturns(QueryRequest request,
+ List<Span>... traces) protected voidassertGetTracesReturnsCount(QueryRequest request,
+ int traceCount) protected voidassertGetTracesReturnsEmpty(List<String> traceIds) protected voidassertGetTracesReturnsEmpty(QueryRequest request) protected voidprotected voidprotected abstract voidclear()protected abstract voidStorageComponent.Builder with parameters for the test being executed.protected booleanStorageComponent before each test rather than the test
+ class.protected ServiceAndSpanNamesnames()protected abstract StorageComponent.BuildernewStorageBuilder(org.junit.jupiter.api.TestInfo testInfo) StorageComponent.Builder for connecting to the backend for the test.protected static QueryRequest.Builderprotected booleanprotected SpanStorestore()protected static StringtestSuffix(org.junit.jupiter.api.TestInfo testInfo) protected Tracestraces()StorageComponent before each test rather than the test
+ class. Generally, tests will run faster if the storage is initialized as infrequently as
+ possibly while clearing data between runs, but for certain backends like Cassandra, it's
+ difficult to reliably clear data between runs and tends to be very slow anyways.StorageComponent.Builder for connecting to the backend for the test.StorageComponent.Builder with parameters for the test being executed.IOExceptionIOExceptionExceptionIOExceptionIOExceptionIOExceptionIOExceptionIOExceptionIOExceptionIOExceptionIOExceptionstrictTraceId ==
+ false.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process. + +
This is a replacement for zipkin.storage.StrictTraceIdFalseTest.
accept128BitTrace(StorageComponent storage,
+ org.junit.jupiter.api.TestInfo testInfo) protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetTrace_retrievesBy64Or128BitTraceId(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTrace_retrievesBy64Or128BitTraceId_mixed(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_128BitTraceId(List<Span> trace,
+ org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_128BitTraceId(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_128BitTraceId_mixed(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_retrievesBy64Or128BitTraceId(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionExceptionExceptionExceptionTraces.
+
+ Subtypes should create a connection to a real backend, even if that backend is in-process.
protected final voidStorageComponent.Builder with parameters for the test being executed.protected voidgetTrace_deduplicates(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTrace_differentiatesDebugFromShared(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTrace_returnsEmptyOnNotFound(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_differentiatesDebugFromShared(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_onlyReturnsTracesThatMatch(org.junit.jupiter.api.TestInfo testInfo) protected voidgetTraces_returnsEmptyOnNotFound(org.junit.jupiter.api.TestInfo testInfo) accept, accept, assertGetTraceReturns, assertGetTraceReturns, assertGetTraceReturnsEmpty, assertGetTracesReturns, assertGetTracesReturns, assertGetTracesReturnsCount, assertGetTracesReturnsEmpty, assertGetTracesReturnsEmpty, blockWhileInFlight, checkStorage, clear, initializeStoragePerTest, names, newStorageBuilder, requestBuilder, returnsRawSpans, sortTrace, store, testSuffix, tracesITStorageStorageComponent.Builder with parameters for the test being executed.configureStorageForTest in class ITStorage<T extends StorageComponent>ExceptionExceptionExceptionExceptionExceptionExceptionclass ITAutocompleteTags<T extends StorageComponent>StorageComponent.Builder.autocompleteKeys(List) has values.class ITDependencies<T extends StorageComponent>SpanStore implementations that support dependency aggregation.class ITDependenciesHeavy<T extends StorageComponent>SpanStore implementations that support dependency aggregation.class ITSearchEnabledFalse<T extends StorageComponent>searchEnabled ==
+ false.class ITServiceAndSpanNames<T extends StorageComponent>ServiceAndSpanNames.class ITSpanStore<T extends StorageComponent>SpanStore.class ITSpanStoreHeavy<T extends StorageComponent>SpanStore implementations.class ITStrictTraceIdFalse<T extends StorageComponent>strictTraceId ==
+ false.class ITTraces<T extends StorageComponent>Traces.StorageComponent.Builder.autocompleteKeys(List) has values.SpanStore implementations that support dependency aggregation.SpanStore implementations that support dependency aggregation.searchEnabled ==
+ false.ServiceAndSpanNames.SpanStore.SpanStore implementations.StorageComponent integration tests.strictTraceId ==
+ false.Traces.SpanConsumer.synchronously or asynchronously.SpanConsumer.SpanBytesEncoder, as it isn't needed for instrumentationSpanConsumer.strict trace ID is disabled.SpanConsumer.Span.tags().Span.public static final int0public static final int6Annotation.Span.tags().Span.You can search for definitions of modules, packages, types, fields, methods, system properties and other terms defined in the API, using some or all of the name, optionally using "camelCase" abbreviations. For example:
+j.l.obj will match "java.lang.Object"InpStr will match "java.io.InputStream"HM.cK will match "java.util.HashMap.containsKey(Object)"Refer to the Javadoc Search Specification for a full description of search features.
+The Overview page is the front page of this API document and provides a list of all packages with a summary for each. This page can also contain an overall description of the set of packages.
+Each package has a page that contains a list of its classes and interfaces, with a summary for each. These pages may contain the following categories:
+Each class, interface, nested class and nested interface has its own separate page. Each of these pages has three sections consisting of a declaration and description, member summary tables, and detailed member descriptions. Entries in each of these sections are omitted if they are empty or not applicable.
+Note: Annotation interfaces have required and optional elements, but not methods. Only enum classes have enum constants. The components of a record class are displayed as part of the declaration of the record class. Properties are a feature of JavaFX.
+The summary entries are alphabetical, while the detailed descriptions are in the order they appear in the source code. This preserves the logical groupings established by the programmer.
+Packages and modules may contain pages with additional information related to the declarations nearby.
+Each documented package, class and interface has its own Use page. This page describes what packages, classes, methods, constructors and fields use any part of the given class or package. Given a class or interface A, its Use page includes subclasses of A, fields declared as A, methods that return A, and methods and constructors with parameters of type A. You can access this page by first going to the package, class or interface, then clicking on the USE link in the navigation bar.
+There is a Class Hierarchy page for all packages, plus a hierarchy for each package. Each hierarchy page contains a list of classes and a list of interfaces. Classes are organized by inheritance structure starting with java.lang.Object. Interfaces do not inherit from java.lang.Object.
The Deprecated API page lists all of the API that have been deprecated. A deprecated API is not recommended for use, generally due to shortcomings, and a replacement API is usually given. Deprecated APIs may be removed in future implementations.
+The Constant Field Values page lists the static final fields and their values.
+Each serializable or externalizable class has a description of its serialization fields and methods. This information is of interest to those who implement rather than use the API. While there is no link in the navigation bar, you can get to this information by going to any serialized class and clicking "Serialized Form" in the "See Also" section of the class description.
+The All Packages page contains an alphabetic index of all packages contained in the documentation.
+The All Classes and Interfaces page contains an alphabetic index of all classes and interfaces contained in the documentation, including annotation interfaces, enum classes, and record classes.
+The Index contains an alphabetic index of all classes, interfaces, constructors, methods, and fields in the documentation, as well as summary pages such as All Packages, All Classes and Interfaces.
+Span.annotationsV1Span.annotations()Span.tags() except duplicating the
+ endpoint.Span.remoteEndpoint()Span.annotations()
+ include a value in this set, or where Span.tags() include a key is in this set.Span.annotations(), except each may be associated with Span.localEndpoint()SpanConsumer.Span.tags() are allocated to binary annotations with a V1BinaryAnnotation.stringValue().synchronously or asynchronously.DependencyLink.parent to DependencyLink.childSpan.localServiceName() if instrumented.Span.annotationsSpan.tagsAnnotation.timestamp, then Annotation.value.V1Annotation.timestamp(), then V1Annotation.value().V1BinaryAnnotation.keySpan.timestamp() is the moment a consumer received a message from an
+ origin.Span.remoteEndpoint()Span.tags() except duplicating the
+ endpoint.Span.debug()Span.debug()Span.debug()V1Span.debug()list.BytesDecoder.decodeList(byte[], Collection)true if an element was decoded or throws IllegalArgumentException.BytesDecoder.decodeList(byte[]).BytesDecoder.decodeList(byte[]).BytesDecoder.decodeOne(byte[])IllegalArgumentException if the input isn't a json, proto3 or thrift list
+ message.IllegalArgumentException if the input isn't a v1 json or thrift single-span
+ messageSpan.durationAsLong()V1Span.duration()Span.duration()Span.duration() except returns a primitive where zero implies absent.V1BinaryAnnotation.stringValue() is present, this is the same as the Span.localEndpoint()
+ Otherwise, it is the same as the Span.remoteEndpoint().Span.timestamp() are at or before this time in epoch
+ milliseconds.QueryRequest.endTs()callback when complete.flatMapper function.AutocompleteTags.getValues(String).local service names,
+ sorted lexicographically.span names recorded by a service, sorted lexicographically.V1Span.id()Span.id() except packed into a long.Span.id() or throws IllegalArgumentException if the
+ input is zero.V1Span.id()Span.id() or throws IllegalArgumentException if not lower-hex format.Span.Builder.id(String)Endpoint.Builder.parseIp(String)Endpoint.Builder.parseIp(InetAddress)Span.tags() v2 span tag}Span.remoteEndpointSpan.kindQueryRequest.limit()Encoding.listSizeInBytes(List), except for a single element.Span.localEndpointSpan.timestamp() are at or after (endTs - lookback) in
+ milliseconds.mapper function.Span.duration() is less than or equal to maxDuration
+ microseconds.Span.duration() is greater than or equal to minDuration
+ microseconds.Span.name()Span.nameV1Span.name()Span.localServiceName() if instrumented.Span.id or null if this the root span in a trace.V1Span.id() or zero if this the root span in a trace.Span.parentId() or unsets if the input is zero.V1Span.parentId()Span.parentId() or throws IllegalArgumentException if not lower-hex
+ format.Span.Builder.parentId(String)Endpoint.Builder.parseIp(String) except this accepts a byte array.Endpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.Endpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.Endpoint.portAsInt()Endpoint.port() except returns a primitive where zero implies absent.Span.timestamp() is the moment a producer sent a message to a destination.Span.tagsSpan.remoteEndpointremote service
+ name.StorageComponent.traces()
+ supported.SpanConsumer.local service name and
+ constrains all other parameters.Endpoint.serviceNameSpan.shared()Span.shared()SpanBytesEncoder, as it isn't needed for instrumentationSpan.name()SpanConsumer.strict trace ID is disabled.Span.tags() v2 span tag} or null if this is an addressSpan.timestampAsLong()V1Span.timestamp()Span.timestamp()Span.timestamp() except returns a primitive where zero implies absent.V1Span.traceId()Span.traceId()V1Span.traceId()Span.id() or throws IllegalArgumentException if not lower-hex format.Span.Builder.traceId(String)V1Span.traceIdHigh()V1Span.traceIdHigh()SpanConsumer.Annotation.Span.tags().Span.cache.miss or errorAnnotation.value()ObjectStreamExceptionlong timestamp+
String value+
ObjectStreamExceptionObjectStreamExceptionObjectStreamExceptionList<Annotation> annotations+
long duration+
int flags+
String id+
Span.Kind kind+
Endpoint localEndpoint+
String name+
String parentId+
Endpoint remoteEndpoint+
Map<String,+String> tags
long timestamp+
String traceId+
Serializable, Comparable<Annotation>Unlike log statements, annotations are often codes: Ex. cache.miss.
intcompareTo(Annotation that) timestamp, then value.static AnnotationbooleaninthashCode()longtoString()value()cache.miss or errorThis value should be set directly by instrumentation, using the most precise value possible.
+ For example, gettimeofday or multiplying System.currentTimeMillis() by 1000.
cache.miss or errortimestamp, then value.compareTo in interface Comparable<Annotation>CloneableCall.Base<V>, Call.ErrorHandler<V>, Call.FlatMapper<V1,V2>, Call.Mapper<V1,V2> final voidcancel()protected voiddoCancel()protected abstract voidprotected abstract Vprotected booleanfinal voidcallback when complete.final Vexecute()final booleanclone, create, emptyList, flatMap, handleError, map, propagateIfFatalCallEventhough this is a blocking call, implementations may honor calls to Call.cancel() from a different thread.
execute in class Call<V>V is Void.IOExceptionIOExceptionCallcallback when complete. Invoking this
+ more than once will result in an error. To repeat a call, make a copy with Call.clone().CallCallCalls can fail before being canceled, so true does always mean cancelation caused a call to + fail. That said, successful cancellation does result in a failure.
isCanceled in class Call<V>voidonErrorReturn(Throwable error,
+ Callback<V> callback) GroupByTraceIdV - the success type, typically not null except when V is Void.CloneableCall.Basesynchronously or asynchronously. At any time, from any thread, you
+ can call cancel(), which might stop an in-flight request or prevent one from
+ occurring.
+
+ Implementations should prepare a call such that there's little or no likelihood of late
+ runtime exceptions. For example, if the call is to get a trace, the call to listSpans
+ should propagate input errors vs delay them until a call to execute() or
+ enqueue(Callback).
+
+
Ex. +
+ // Any translation of an input request to remote parameters should happen here, and any related
+ // errors should propagate here.
+ Call<List<List<Span>>> listTraces = spanStore.listTraces(request);
+ // When this executes, it should simply run the remote request.
+ List<Span> trace = getTraceCall.execute();
+ An instance of call cannot be invoked more than once, but you can clone() an + instance if you need to replay the call. There is no relationship between a call and a number of + remote requests. For example, an implementation that stores spans may make hundreds of remote + requests, possibly retrying on your behalf. + +
This type owes its design to retrofit2.Call, which is nearly the same, except limited
+ to HTTP transports.
static class static interface static interface static interface abstract voidcancel()clone()static <V> Call<V>create(V v) abstract voidcallback when complete.abstract Vexecute()final <R> Call<R>flatMap(Call.FlatMapper<V,R> flatMapper) flatMapper function.handleError(Call.ErrorHandler<V> errorHandler) abstract booleanfinal <R> Call<R>map(Call.Mapper<V,R> mapper) mapper function.static voidmapper function.
+ This is used to convert values from one type to another. For example, you could use this to
+ convert between zipkin v1 and v2 span format.
+
+
+ getTracesV1Call = getTracesV2Call.map(traces -> v2TracesConverter);
+ This method intends to be used for chaining. That means "this" instance should be discarded + in favor of the result of this method.
flatMapper function. This
+ is used to chain two remote calls together. For example, you could use this to chain a list IDs
+ call to a get by IDs call.
+
+
+ getTracesCall = getIdsCall.flatMap(ids -> getTraces(ids));
+
+ // this would now invoke the chain
+ traces = getTracesCall.enqueue(tracesCallback);
+ This method intends to be used for chaining. That means "this" instance should be discarded + in favor of the result of this method.
Here's an example of coercing 404 to empty: +
+ call.handleError((error, callback) -> {
+ if (error instanceof HttpException && ((HttpException) error).code == 404) {
+ callback.onSuccess(Collections.emptyList());
+ } else {
+ callback.onError(error);
+ }
+ });
+ Eventhough this is a blocking call, implementations may honor calls to cancel() from a different thread.
V is Void.IOExceptioncallback when complete. Invoking this
+ more than once will result in an error. To repeat a call, make a copy with clone().Calls can fail before being canceled, so true does always mean cancelation caused a call to + fail. That said, successful cancellation does result in a failure.
This is a bridge to async libraries such as CompletableFuture complete, completeExceptionally. + +
Implementations will call either onSuccess(V) or onError(java.lang.Throwable), but not both.
When this is called, onError(java.lang.Throwable) won't be.
When this is called, onSuccess(V) won't be.
Implementations should initialize the component if necessary. It should test a remote + connection, or consult a trusted source to derive the result. They should use least resources + possible to establish a meaningful result, and be safe to call many times, even concurrently.
OKCloseable, AutoCloseableStorageComponentComponents are lazy with regards to I/O. They can be injected directly to other components so + as to avoid crashing the application graph if a network service is unavailable.
Implementations should initialize the component if necessary. It should test a remote + connection, or consult a trusted source to derive the result. They should use least resources + possible to establish a meaningful result, and be safe to call many times, even concurrently.
For example, if this created a connection, it would close it. If it was provided one, this + would close any sessions, but leave the connection open.
close in interface AutoCloseableclose in interface CloseableIOExceptionbuild()callCount(long callCount) errorCount(long errorCount) Serializablelongparent to childchild()Span.localServiceName() if instrumented.booleanlonginthashCode()static DependencyLink.Builderparent()Span.localServiceName() if instrumented.toString()Span.localServiceName() if instrumented.Span.localServiceName() if instrumented.parent to childbuild()parseIp(String)ip(InetAddress addr) parseIp(InetAddress)booleanparseIp(byte[] ipBytes) parseIp(String) except this accepts a byte array.booleanEndpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.booleanparseIp(InetAddress addr) Endpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.port(int port) Endpoint.portAsInt()serviceName(String serviceName) Endpoint.serviceNameEndpoint.serviceNameparseIp(InetAddress)Endpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.
+
+ Returns boolean not this for conditional parsing. For example: +
+ if (!builder.parseIp(input.getHeader("X-Forwarded-For"))) {
+ builder.parseIp(input.getRemoteAddr());
+ }
+ parseIp(String) except this accepts a byte array.ipBytes - byte array whose ownership is exclusively transferred to this endpoint.parseIp(String)Endpoint.ipv4() or Endpoint.ipv6() could be parsed from the
+ input.
+
+ Returns boolean not this for conditional parsing. For example: +
+ if (!builder.parseIp(input.getHeader("X-Forwarded-For"))) {
+ builder.parseIp(input.getRemoteAddr());
+ }
+ port - port associated with the endpoint. zero coerces to null (unknown)Endpoint.portAsInt()SerializablebooleaninthashCode()ipv4()byte[]ipv6()byte[]static Endpoint.Builderport()intport() except returns a primitive where zero implies absent.toString()This is a primary label for trace lookup and aggregation, so it should be intuitive and + consistent. Many use a name from service discovery.
port() except returns a primitive where zero implies absent.
+
+ Using this method will avoid allocation, so is encouraged when copying data.
addAnnotation(long timestamp,
+ String value) Span.annotationsbuild()clear()Span.annotationsSpan.tagsclone()debug(boolean debug) Span.debug()Span.debug()duration(long duration) Span.duration()id(long id) Span.id() or throws IllegalArgumentException if the
+ input is zero.Span.id() or throws IllegalArgumentException if not lower-hex format.kind()Span.kindlocalEndpoint(Endpoint localEndpoint) Span.localEndpointSpan.nameparentId(long parentId) Span.parentId() or unsets if the input is zero.Span.parentId() or throws IllegalArgumentException if not lower-hex
+ format.Span.tagsremoteEndpoint(Endpoint remoteEndpoint) Span.remoteEndpointshared(boolean shared) Span.shared()Span.shared()timestamp(long timestamp) Span.timestamp()traceId(long high,
+ long low) Span.id() or throws IllegalArgumentException if not lower-hex format.Span.id() or throws IllegalArgumentException if not lower-hex format.high - Upper 64bits of the trace ID. Zero means the trace ID is 64-bit.low - Lower 64bits of the trace ID.IllegalArgumentException - if both values are zeroSpan.parentId() or unsets if the input is zero.Span.parentId() or throws IllegalArgumentException if not lower-hex
+ format.Span.id() or throws IllegalArgumentException if the
+ input is zero.Span.id() or throws IllegalArgumentException if not lower-hex format.Span.kindSpan.nameSpan.timestamp()Span.duration()Span.localEndpointSpan.remoteEndpointSpan.annotationsSpan.annotationsSpan.tagsSpan.tagsSpan.debug()Span.debug()Span.shared()Span.shared()Serializable, Comparable<Span.Kind>Span.timestamp() is the moment a consumer received a message from an
+ origin.Span.timestamp() is the moment a producer sent a message to a destination.Span.timestamp() is the moment a producer sent a message to a destination.
+ Span.duration() represents delay sending the message, such as batching, while Span.remoteEndpoint() indicates the destination, such as a broker.
+
+ Unlike CLIENT, messaging spans never share a span ID. For example, the CONSUMER of the same message has Span.parentId() set to this span's Span.id().
Span.timestamp() is the moment a consumer received a message from an
+ origin. Span.duration() represents delay consuming the message, such as from backlog,
+ while Span.remoteEndpoint() indicates the origin, such as a broker.
+
+ Unlike SERVER, messaging spans never share a span ID. For example, the PRODUCER of this message is the Span.parentId() of this span.
name - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullSerializableparentId field establishes the position of one span in the tree.
+
+ The root span is where parentId is null and usually has the longest duration in the trace. However, nested asynchronous work can materialize as child spans whose
+ duration exceed the root span.
+
+
Spans usually represent remote activity such as RPC calls, or messaging producers and + consumers. However, they can also represent in-process activity in any position of the trace. For + example, a root span could represent a server receiving an initial client request. A root span + could also represent a scheduled job that has no remote context. + +
While span identifiers are packed into longs, they should be treated opaquely. ID encoding is + 16 or 32 character lower-hex, to avoid signed interpretation. + +
zipkin.SpanThis type is intended to replace use of zipkin.Span. Particularly, tracers represent
+ a single-host view of an operation. By making one endpoint implicit for all data, this type does
+ not need to repeat endpoints on each data like zipkin.Span does. This results in simpler
+ and smaller data.
static final class static enum debug()duration()longduration() except returns a primitive where zero implies absent.booleaninthashCode()id()kind()remoteEndpointname()static Span.Builderstatic StringnormalizeTraceId(String traceId) parentId()id or null if this the root span in a trace.shared()tags()longtimestamp() except returns a primitive where zero implies absent.toString()traceId()Encoded as 16 or 32 lowercase hex characters corresponding to 64 or 128 bits. For example,
+ a 128bit trace ID looks like 4e441824ec2b6a44ffdc9bb9a6453df3.
+
+
Some systems downgrade trace identifiers to 64bit by dropping the left-most 16 characters.
+ For example, 4e441824ec2b6a44ffdc9bb9a6453df3 becomes ffdc9bb9a6453df3.
id or null if this the root span in a trace.
+
+ This is the same encoding as id. For example ffdc9bb9a6453df3
Encoded as 16 lowercase hex characters. For example ffdc9bb9a6453df3
+
+
A span is uniquely identified in storage by (traceId, id()).
remoteEndpointConventionally, when the span name isn't known, name = "unknown".
This value should be set directly by instrumentation, using the most precise value
+ possible. For example, gettimeofday or multiplying System.currentTimeMillis() by
+ 1000.
+
+
There are three known edge-cases where this could be reported absent: + +
Note: timestamps at or before epoch (0L == 1970) are invalid
timestamp() except returns a primitive where zero implies absent.
+
+ Using this method will avoid allocation, so is encouraged when copying data.
This value should be set directly, as opposed to implicitly via annotation timestamps. + Doing so encourages precision decoupled from problems of clocks, such as skew or NTP updates + causing time to move backwards. + +
If this field is persisted as unset, zipkin will continue to work, except duration query + support will be implementation-specific. Similarly, setting this field non-atomically is + implementation-specific. + +
This field is i64 vs i32 to support spans longer than 35 minutes.
duration() except returns a primitive where zero implies absent.
+
+ Using this method will avoid allocation, so is encouraged when copying data.
Instrumentation should always record this and be consistent as possible with the service + name as it is used in search. This is nullable for legacy reasons.
By recording the remote endpoint, your trace will contain network context even if the peer
+ is not tracing. For example, you can record the IP from the X-Forwarded-For header or
+ the service name and socket of a remote peer.
For example, a tag key could be "http.path".
kind() to be Span.Kind.SERVER.
+
+ When an RPC trace is client-originated, it will be sampled and the same span ID is used for + the server side. However, the server shouldn't set span.timestamp or duration since it didn't + start the span.
IllegalArgumentException - if oversized or not lower-hexFor example, we can identify the encoding and also the format in http via the request path + and content-type. However, in Kafka it could be that folks send mixed Zipkin data without + identifying its format. For example, Kafka historically has no content-type and users don't + always segregate different queues by instrumentation format.
static BytesDecoder<Span>decoderForListMessage(byte[] spans) IllegalArgumentException if the input isn't a json, proto3 or thrift list
+ message.static BytesDecoder<Span>decoderForListMessage(ByteBuffer spans) static BytesDecoder<Span>decoderForMessage(byte[] span) IllegalArgumentException if the input isn't a v1 json or thrift single-span
+ messageIllegalArgumentException if the input isn't a v1 json or thrift single-span
+ messageIllegalArgumentException if the input isn't a json, proto3 or thrift list
+ message.annotations()intcompareTo(Annotation that) timestamp, then value.handleError(Call.ErrorHandler<V> errorHandler) final <R> Call<R>flatMap(Call.FlatMapper<V,R> flatMapper) flatMapper function.final <R> Call<R>map(Call.Mapper<V,R> mapper) mapper function.final class create(boolean strictTraceId) static Call.Mapper<List<Span>,List<Span>> filterSpans(String traceId) filterTraces(Iterable<String> traceIds) filterTraces(QueryRequest request) clone()static <V> Call<V>create(V v) emptyList()final <R> Call<R>flatMap(Call.FlatMapper<V,R> flatMapper) flatMapper function.handleError(Call.ErrorHandler<V> errorHandler) final <R> Call<R>map(Call.Mapper<V,R> mapper) mapper function.getDependencies(long endTs,
+ long lookback) getDependencies(long endTs,
+ long lookback) getKeys()AutocompleteTags.getValues(String).getKeys()getRemoteServiceNames(String service) getRemoteServiceNames(String serviceName) getServiceNames()getServiceNames()local service names,
+ sorted lexicographically.getServiceNames()getSpanNames(String service) getSpanNames(String serviceName) span names recorded by a service, sorted lexicographically.getSpanNames(String serviceName) getTraces(QueryRequest request) getTraces(QueryRequest request) protected abstract voidfinal voidabstract voidcallback when complete.voidonErrorReturn(Throwable error,
+ Callback<V> callback) check()static CheckResultclass final class class callCount(long callCount) errorCount(long errorCount) static DependencyLink.BuildernewBuilder()toBuilder()getDependencies()getDependencies(long endTs,
+ long lookback) getDependencies(long endTs,
+ long lookback) parseIp(String)ip(InetAddress addr) parseIp(InetAddress)static Endpoint.BuildernewBuilder()port(int port) Endpoint.portAsInt()serviceName(String serviceName) Endpoint.serviceNametoBuilder()build()localEndpoint()localEndpoint()remoteEndpoint()localEndpoint(Endpoint localEndpoint) Span.localEndpointremoteEndpoint(Endpoint remoteEndpoint) Span.remoteEndpointendpoint()endpoint()V1BinaryAnnotation.stringValue() is present, this is the same as the Span.localEndpoint()
+ Otherwise, it is the same as the Span.remoteEndpoint().addAnnotation(long timestamp,
+ String value,
+ Endpoint endpoint) V1Span.annotations()addBinaryAnnotation(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.addBinaryAnnotation(String address,
+ Endpoint endpoint) Span.remoteEndpoint()static V1Annotationstatic V1BinaryAnnotationcreateAddress(String address,
+ Endpoint endpoint) Span.remoteEndpoint()static V1BinaryAnnotationcreateString(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.addAnnotation(long timestamp,
+ String value) Span.annotationsclear()clearAnnotations()Span.annotationsclearTags()Span.tagsclone()debug(boolean debug) Span.debug()Span.debug()duration(long duration) Span.duration()id(long id) Span.id() or throws IllegalArgumentException if the
+ input is zero.Span.id() or throws IllegalArgumentException if not lower-hex format.Span.kindlocalEndpoint(Endpoint localEndpoint) Span.localEndpointSpan.namestatic Span.BuildernewBuilder()parentId(long parentId) Span.parentId() or unsets if the input is zero.Span.parentId() or throws IllegalArgumentException if not lower-hex
+ format.Span.tagsremoteEndpoint(Endpoint remoteEndpoint) Span.remoteEndpointshared(boolean shared) Span.shared()Span.shared()timestamp(long timestamp) Span.timestamp()toBuilder()traceId(long high,
+ long low) Span.id() or throws IllegalArgumentException if not lower-hex format.kind()kind()Span.remoteEndpointstatic Span.Kindstatic Span.Kind[]values()Span.kindstatic BytesDecoder<Span>decoderForListMessage(byte[] spans) IllegalArgumentException if the input isn't a json, proto3 or thrift list
+ message.static BytesDecoder<Span>decoderForListMessage(ByteBuffer spans) static BytesDecoder<Span>decoderForMessage(byte[] span) IllegalArgumentException if the input isn't a v1 json or thrift single-span
+ messageabstract SpandecodeOne(ByteBuffer span) BytesDecoder.decodeOne(byte[])decodeList(ByteBuffer spans) BytesDecoder.decodeList(byte[]).abstract booleandecodeList(ByteBuffer spans,
+ Collection<Span> out) BytesDecoder.decodeList(byte[]).abstract intencodeList(List<Span> spans,
+ byte[] out,
+ int pos) create(boolean strictTraceId) create(boolean strictTraceId) static Call.Mapper<List<Span>,List<Span>> filterSpans(String traceId) static Call.Mapper<List<Span>,List<Span>> filterSpans(String traceId) filterTraces(Iterable<String> traceIds) filterTraces(Iterable<String> traceIds) filterTraces(QueryRequest request) filterTraces(QueryRequest request) getTraces()getTraces(QueryRequest request) getTraces(QueryRequest request) booleanvoidconvert(V1Span source,
+ Collection<Span> sink) T - type of the object to deserializeDependencyLinkBytesDecoder, SpanBytesDecoder
+ ArrayList<Span> out = new ArrayList<>();
+ SpanBytesDecoder.JSON_V2.decodeList(spans, out)
+ booleandecode(byte[] serialized,
+ Collection<T> out) list.decodeList(byte[] serialized) decodeList(byte[], Collection)booleandecodeList(byte[] serialized,
+ Collection<T> out) true if an element was decoded or throws IllegalArgumentException.decodeOne(byte[] serialized) encoding()list.
+
+ Note: multiple elements can be consumed from a single serialized object. For example, if the + input is Zipkin v1, the list might receive two elements if the serialized object was a shared + span.
serialized - a single message, for example a json objecttrue if an element was decoded or throws IllegalArgumentException.decodeList(byte[], Collection)T - type of the object to encodeDependencyLinkBytesEncoder, SpanBytesEncoderbyte[]byte[]encodeList(List<T> input) encoding()intsizeInBytes(T input) Serializable, Comparable<DependencyLinkBytesDecoder>, BytesDecoder<DependencyLink>static DependencyLinkBytesDecoderstatic DependencyLinkBytesDecoder[]values()clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOfdecode, decodeList, decodeList, decodeOne, encodingname - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullSerializable, Comparable<DependencyLinkBytesEncoder>, BytesEncoder<DependencyLink>static DependencyLinkBytesEncoderstatic DependencyLinkBytesEncoder[]values()clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOfencode, encodeList, encoding, sizeInBytesname - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullSerializable, Comparable<Encoding>abstract intlistSizeInBytes(int encodedSizeInBytes) listSizeInBytes(List), except for a single element.abstract intlistSizeInBytes(List<byte[]> values) static Encodingstatic Encoding[]values()The message's binary data includes a list header followed by N spans serialized in + TBinaryProtocol
See https://developers.google.com/protocol-buffers/docs/encoding#optional
name - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nulllistSizeInBytes(List), except for a single element.Serializable, Comparable<SpanBytesDecoder>, BytesDecoder<Span>SpanBytesEncoder, as it isn't needed for instrumentationdecodeList(ByteBuffer spans) BytesDecoder.decodeList(byte[]).abstract booleandecodeList(ByteBuffer spans,
+ Collection<Span> out) BytesDecoder.decodeList(byte[]).abstract SpandecodeOne(ByteBuffer span) BytesDecoder.decodeOne(byte[])static SpanBytesDecoderstatic SpanBytesDecoder[]values()clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOfdecode, decodeList, decodeList, decodeOne, encodingname - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullBytesDecoder.decodeList(byte[]).
+
+ Note: only use this when it is ok to modify the underlying ByteBuffer.array().
BytesDecoder.decodeList(byte[]).
+
+ Note: only use this when it is ok to modify the underlying ByteBuffer.array().
BytesDecoder.decodeOne(byte[])
+
+ Note: only use this when it is ok to modify the underlying ByteBuffer.array().
Serializable, Comparable<SpanBytesEncoder>, BytesEncoder<Span>abstract intencodeList(List<Span> spans,
+ byte[] out,
+ int pos) static SpanBytesEncoderstatic SpanBytesEncoder[]values()clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOfencode, encodeList, encoding, sizeInBytesname - the name of the enum constant to be returned.IllegalArgumentException - if this enum type has no constant with the specified nameNullPointerException - if the argument is nullstatic BytesDecoder<Span>decoderForListMessage(byte[] spans) IllegalArgumentException if the input isn't a json, proto3 or thrift list
+ message.static BytesDecoder<Span>decoderForListMessage(ByteBuffer spans) static BytesDecoder<Span>decoderForMessage(byte[] span) IllegalArgumentException if the input isn't a v1 json or thrift single-span
+ messageenum enum SpanBytesEncoder, as it isn't needed for instrumentationenum enum static DependencyLinkBytesDecoderstatic DependencyLinkBytesDecoder[]values()static DependencyLinkBytesEncoderstatic DependencyLinkBytesEncoder[]values()encoding()encoding()static Encodingstatic Encoding[]values()static SpanBytesDecoderstatic SpanBytesDecoder[]values()static SpanBytesEncoderstatic SpanBytesEncoder[]values()SpanBytesEncoder, as it isn't needed for instrumentationSpanBytesEncoder, as it isn't needed for instrumentationsynchronously or asynchronously.synchronously or asynchronously.synchronously or asynchronously.InMemoryStorageSpanConsumer.getValues(String).IllegalArgumentException - if the input is empty.Closeable, AutoCloseableExtending this is better than extending StorageComponent directly because it reduces
+ risk of accidentally masking new methods. For example, if you extended storage component and
+ later a new feature for cache control was added, that feature would be blocked until the wrapper
+ was re-compiled. Such would be worse in most cases than not having decoration on new methods.
StorageComponent.Builderprotected check()voidclose()protected abstract StorageComponentdelegate()booleantoString()traces()spanConsumer in class StorageComponenttraces in class StorageComponentspanStore in class StorageComponentautocompleteTags in class StorageComponentserviceAndSpanNames in class StorageComponentComponentImplementations should initialize the component if necessary. It should test a remote + connection, or consult a trusted source to derive the result. They should use least resources + possible to establish a meaningful result, and be safe to call many times, even concurrently.
StorageComponentCallers of this method will submit an exception raised by Call.execute() or on the
+ error callback of Call.enqueue(Callback).
+
+
By default, this returns true if the input is a RejectedExecutionException. When
+ originating exceptions, use this type to indicate a load related failure.
+
+
It is generally preferred to specialize this method to handle relevant exceptions for the
+ particular storage rather than wrapping them in RejectedExecutionException at call
+ sites. Extra wrapping can make errors harder to read, for example, by making it harder to
+ "google" a solution for a well known error message for the storage client, instead thinking the
+ error is in Zipkin code itself.
+
+
While implementation is flexible, one known use is Netflix + concurrency limits
isOverCapacity in class StorageComponentComponentFor example, if this created a connection, it would close it. If it was provided one, this + would close any sessions, but leave the connection open.
close in interface AutoCloseableclose in interface Closeableclose in class ComponentIOExceptionSpanStore.getTraces(QueryRequest).autocompleteKeys(List<String> autocompleteKeys) build()maxSpanCount(int maxSpanCount) searchEnabled(boolean searchEnabled) StorageComponent.traces()
+ supported.strictTraceId(boolean strictTraceId) autocompleteCardinality, autocompleteTtlStorageComponent.BuilderZipkin historically had 64-bit trace IDs, but it now supports 128-
+ bit trace IDs via 32-character hex representation. While instrumentation update to propagate
+ 128-bit IDs, it can be ambiguous whether a 64-bit trace ID was sent intentionally, or as an
+ accident of truncation. This setting allows Zipkin to be usable until application
+ instrumentation are upgraded to support 128-bit trace IDs.
+
+
Here are a few trace IDs the help explain this setting. + +
In the above example, Trace ID A and Trace ID B might mean they are in the same trace,
+ since the lower-64 bits of the IDs are the same. This could happen if a server A created the
+ trace and propagated it to server B which ran an older tracing library. Server B could have
+ truncated the trace ID to lower-64 bits. When strictTraceId == false, spans matching
+ either trace ID A or B would be returned in the same trace when searching by ID A or B. Spans
+ with trace ID C or D wouldn't be when searching by ID A or B because trace IDs C and D don't
+ share lower 64-bits (right-most 16 characters) with trace IDs A or B.
+
+
It is also possible that all servers are capable of handling 128-bit trace identifiers,
+ but are configured to only send 64-bit ones. In this case, if strictTraceId == false
+ trace ID A and B would clash and be put into the same trace, causing confusion. Moreover,
+ there is overhead associated with indexing spans both by 64 and 128-bit trace IDs. When a
+ site has finished upgrading to 128-bit trace IDs, they should enable this setting.
+
+
See https://github.com/openzipkin/b3-propagation/issues/6 for the status of + known open source libraries on 128-bit trace identifiers.
strictTraceId in class StorageComponent.BuilderStorageComponent.BuilderStorageComponent.traces()
+ supported. For example, query requests will be disabled.
+ + The use case is typically to support 100% sampled data, or when traces are searched using + alternative means such as a logging index. + +
Refer to implementation docs for the impact of this parameter. Operations that use + indexes should return empty as opposed to throwing an exception.
searchEnabled in class StorageComponent.BuilderStorageComponent.Builderspan tags. Good
+ choices for autocomplete are limited in cardinality for the same reasons as service and span
+ names.
+ + For example, "http.url" would be a bad choice for autocomplete, not just because it isn't + site-specific (such as environment would be), but also as there are unlimited getValues due + to factors such as unique ids in the path.
autocompleteKeys in class StorageComponent.BuilderautocompleteKeys - controls the span values stored for auto-complete.build in class StorageComponent.BuilderCloseable, AutoCloseable, AutocompleteTags, ServiceAndSpanNames, SpanConsumer, SpanStore, TracesInternally, spans are indexed on 64-bit trace ID + +
Here's an example of some traces in memory: + +
+ spansByTraceIdTimeStamp:
+ <aaaa,July 4> --> ( spanA(time:July 4, traceId:aaaa, service:foo, name:GET),
+ spanB(time:July 4, traceId:aaaa, service:bar, name:GET) )
+ <cccc,July 4> --> ( spanC(time:July 4, traceId:aaaa, service:foo, name:GET) )
+ <bbbb,July 5> --> ( spanD(time:July 5, traceId:bbbb, service:biz, name:GET) )
+ <bbbb,July 6> --> ( spanE(time:July 6, traceId:bbbb) service:foo, name:POST )
+
+ traceIdToTraceIdTimeStamps:
+ aaaa --> [ <aaaa,July 4> ]
+ bbbb --> [ <bbbb,July 5>, <bbbb,July 6> ]
+ cccc --> [ <cccc,July 4> ]
+
+ serviceToTraceIds:
+ foo --> [ <aaaa>, <cccc>, <bbbb> ]
+ bar --> [ <aaaa> ]
+ biz --> [ <bbbb> ]
+
+ serviceToSpanNames:
+ bar --> ( GET )
+ biz --> ( GET )
+ foo --> ( GET, POST )
+ intvoidclear()voidclose()getDependencies(long endTs,
+ long lookback) getKeys()AutocompleteTags.getValues(String).getRemoteServiceNames(String service) getSpanNames(String service) span names recorded by a service, sorted lexicographically.getTraces(QueryRequest request) static InMemoryStorage.BuildertoString()traces()isOverCapacityaccept in interface SpanConsumerSpanStoreWhen strict trace ID is disabled, spans are grouped by the right-most 16 characters of the + trace ID.
SpanStoreWhen strict trace ID is disabled, spans with the same right-most 16 characters are returned + even if the characters to the left are not. + +
Implementations should use Span.normalizeTraceId(String) to ensure consistency.
TracesWhen strict trace ID is disabled, spans with the same right-most 16 characters are returned + even if the characters to the left are not. + +
Implementations should use Span.normalizeTraceId(String) on each input trace ID to
+ ensure consistency.
SpanStoregetServiceNames in interface ServiceAndSpanNamesgetServiceNames in interface SpanStoreServiceAndSpanNamesgetRemoteServiceNames in interface ServiceAndSpanNamesSpanStorespan names recorded by a service, sorted lexicographically.getSpanNames in interface ServiceAndSpanNamesgetSpanNames in interface SpanStoreSpanStoreImplementations may bucket aggregated data, for example daily. When this is the case, endTs + may be floored to align with that bucket, for example midnight if daily. lookback applies to + the original endTs, even when bucketed. Using the daily example, if endTs was 11pm and lookback + was 25 hours, the implementation would query against 2 buckets. + +
Some implementations parse spans from storage and call DependencyLinker to aggregate links. The reason is certain graph logic, such as skipping up
+ the tree is difficult to implement as a storage query.
+
+
Spans are grouped by the right-most 16 characters of the trace ID. This ensures call counts + are not incremented twice due to one hop downgrading from 128 to 64-bit trace IDs.
getDependencies in interface SpanStoreendTs - only return links from spans where Span.timestamp() are at or before this
+ time in epoch milliseconds.lookback - only return links from spans where Span.timestamp() are at or after
+ (endTs - lookback) in milliseconds.AutocompleteTagsAutocompleteTags.getValues(String).getKeys in interface AutocompleteTagsAutocompleteTagsgetValues in interface AutocompleteTagstraces in class StorageComponentspanStore in class StorageComponentautocompleteTags in class StorageComponentserviceAndSpanNames in class StorageComponentspanConsumer in class StorageComponentComponentFor example, if this created a connection, it would close it. If it was provided one, this + would close any sessions, but leave the connection open.
annotationQuery(Map<String,String> annotationQuery) build()endTs(long endTs) QueryRequest.endTs()limit(int limit) QueryRequest.limit()lookback(long lookback) maxDuration(Long maxDuration) minDuration(Long minDuration) parseAnnotationQuery(String annotationQuery) remoteServiceName(String remoteServiceName) serviceName(String serviceName) QueryRequest.endTs()QueryRequest.limit()Results should be filtered against endTs, subject to limit and lookback. For example, if endTs is 10:20 today, limit is 10, and lookback is 7 days, traces
+ returned should be those nearest to 10:20 today, not 10:20 a week ago.
+
+
Time units of endTs and lookback are milliseconds as opposed to
+ microseconds, the grain of Span.timestamp(). Milliseconds is a more familiar and
+ supported granularity for query, index and windowing functions.
Span.annotations()
+ include a value in this set, or where Span.tags() include a key is in this set.longendTs()Span.timestamp() are at or before this time in epoch
+ milliseconds.intlimit()longlookback()Span.timestamp() are at or after (endTs - lookback) in
+ milliseconds.Span.duration() is less than or equal to maxDuration
+ microseconds.Span.duration() is greater than or equal to minDuration
+ microseconds.static QueryRequest.Builderremote service
+ name.local service name and
+ constrains all other parameters.spanName()Span.name()booleantoString()local service name and
+ constrains all other parameters.remote service
+ name.Span.name()Span.annotations()
+ include a value in this set, or where Span.tags() include a key is in this set. When
+ not, include traces whose Span.tags() an entry in this map.
+
+ Multiple entries are combined with AND, and AND against other conditions.
Span.duration() is greater than or equal to minDuration
+ microseconds.Span.duration() is less than or equal to maxDuration
+ microseconds. Only valid with minDuration.Span.timestamp() are at or before this time in epoch
+ milliseconds. Defaults to current time.Span.timestamp() are at or after (endTs - lookback) in
+ milliseconds. Defaults to endTs.This is used when the backend cannot fully refine a trace query.
InMemoryStorageSpanConsumer.getRemoteServiceNames(String serviceName) local service names,
+ sorted lexicographically.getSpanNames(String serviceName) span names recorded by a service, sorted lexicographically.local service names,
+ sorted lexicographically.span names recorded by a service, sorted lexicographically.InMemoryStorageInMemoryStorageSpanConsumer.
+
+ Note: This is not considered a user-level Api, rather an Spi that can be used to bind + user-level abstractions such as futures or observables.
getDependencies(long endTs,
+ long lookback) getSpanNames(String serviceName) getTraces(QueryRequest request) When strict trace ID is disabled, spans are grouped by the right-most 16 characters of the + trace ID.
When strict trace ID is disabled, spans with the same right-most 16 characters are returned + even if the characters to the left are not. + +
Implementations should use Span.normalizeTraceId(String) to ensure consistency.
traceId - the trace IDspan names recorded by a service, sorted lexicographically.Implementations may bucket aggregated data, for example daily. When this is the case, endTs + may be floored to align with that bucket, for example midnight if daily. lookback applies to + the original endTs, even when bucketed. Using the daily example, if endTs was 11pm and lookback + was 25 hours, the implementation would query against 2 buckets. + +
Some implementations parse spans from storage and call DependencyLinker to aggregate links. The reason is certain graph logic, such as skipping up
+ the tree is difficult to implement as a storage query.
+
+
Spans are grouped by the right-most 16 characters of the trace ID. This ensures call counts + are not incremented twice due to one hop downgrading from 128 to 64-bit trace IDs.
endTs - only return links from spans where Span.timestamp() are at or before this
+ time in epoch milliseconds.lookback - only return links from spans where Span.timestamp() are at or after
+ (endTs - lookback) in milliseconds.InMemoryStorage.BuilderautocompleteCardinality(int autocompleteCardinality) autocompleteKeys(List<String> keys) autocompleteTtl(int autocompleteTtl) abstract StorageComponentbuild()abstract StorageComponent.BuildersearchEnabled(boolean searchEnabled) StorageComponent.traces()
+ supported.abstract StorageComponent.BuilderstrictTraceId(boolean strictTraceId) Zipkin historically had 64-bit trace IDs, but it now supports 128-
+ bit trace IDs via 32-character hex representation. While instrumentation update to propagate
+ 128-bit IDs, it can be ambiguous whether a 64-bit trace ID was sent intentionally, or as an
+ accident of truncation. This setting allows Zipkin to be usable until application
+ instrumentation are upgraded to support 128-bit trace IDs.
+
+
Here are a few trace IDs the help explain this setting. + +
In the above example, Trace ID A and Trace ID B might mean they are in the same trace,
+ since the lower-64 bits of the IDs are the same. This could happen if a server A created the
+ trace and propagated it to server B which ran an older tracing library. Server B could have
+ truncated the trace ID to lower-64 bits. When strictTraceId == false, spans matching
+ either trace ID A or B would be returned in the same trace when searching by ID A or B. Spans
+ with trace ID C or D wouldn't be when searching by ID A or B because trace IDs C and D don't
+ share lower 64-bits (right-most 16 characters) with trace IDs A or B.
+
+
It is also possible that all servers are capable of handling 128-bit trace identifiers,
+ but are configured to only send 64-bit ones. In this case, if strictTraceId == false
+ trace ID A and B would clash and be put into the same trace, causing confusion. Moreover,
+ there is overhead associated with indexing spans both by 64 and 128-bit trace IDs. When a
+ site has finished upgrading to 128-bit trace IDs, they should enable this setting.
+
+
See https://github.com/openzipkin/b3-propagation/issues/6 for the status of + known open source libraries on 128-bit trace identifiers.
StorageComponent.traces()
+ supported. For example, query requests will be disabled.
+ + The use case is typically to support 100% sampled data, or when traces are searched using + alternative means such as a logging index. + +
Refer to implementation docs for the impact of this parameter. Operations that use + indexes should return empty as opposed to throwing an exception.
span tags. Good
+ choices for autocomplete are limited in cardinality for the same reasons as service and span
+ names.
+ + For example, "http.url" would be a bad choice for autocomplete, not just because it isn't + site-specific (such as environment would be), but also as there are unlimited getValues due + to factors such as unique ids in the path.
keys - controls the span values stored for auto-complete.Closeable, AutoCloseableForwardingStorageComponent, InMemoryStoragebooleanabstract SpanConsumerabstract SpanStoretraces()Callers of this method will submit an exception raised by Call.execute() or on the
+ error callback of Call.enqueue(Callback).
+
+
By default, this returns true if the input is a RejectedExecutionException. When
+ originating exceptions, use this type to indicate a load related failure.
+
+
It is generally preferred to specialize this method to handle relevant exceptions for the
+ particular storage rather than wrapping them in RejectedExecutionException at call
+ sites. Extra wrapping can make errors harder to read, for example, by making it harder to
+ "google" a solution for a well known error message for the storage client, instead thinking the
+ error is in Zipkin code itself.
+
+
While implementation is flexible, one known use is Netflix + concurrency limits
strict trace ID is disabled.static Call.Mapper<List<Span>,List<Span>> filterSpans(String traceId) filterTraces(Iterable<String> traceIds) filterTraces(QueryRequest request) FilterTracesMake sure the input IDs are unique and normalized.
InMemoryStorageSpanConsumer.
+
+ Specifically, this provides apis present when search is disabled.
+
+
Note: This is not considered a user-level Api, rather an Spi that can be used to bind + user-level abstractions such as futures or observables.
When strict trace ID is disabled, spans with the same right-most 16 characters are returned + even if the characters to the left are not. + +
Implementations should use Span.normalizeTraceId(String) to ensure consistency.
traceId - the trace IDWhen strict trace ID is disabled, spans with the same right-most 16 characters are returned + even if the characters to the left are not. + +
Implementations should use Span.normalizeTraceId(String) on each input trace ID to
+ ensure consistency.
traceIds - a list of unique trace IDs.final class autocompleteTags()autocompleteTags()autocompleteKeys(List<String> autocompleteKeys) maxSpanCount(int maxSpanCount) static InMemoryStorage.BuildernewBuilder()searchEnabled(boolean searchEnabled) strictTraceId(boolean strictTraceId) autocompleteTags()build()serviceAndSpanNames()spanStore()traces()annotationQuery(Map<String,String> annotationQuery) endTs(long endTs) QueryRequest.endTs()limit(int limit) QueryRequest.limit()lookback(long lookback) maxDuration(Long maxDuration) minDuration(Long minDuration) static QueryRequest.BuildernewBuilder()parseAnnotationQuery(String annotationQuery) remoteServiceName(String remoteServiceName) serviceName(String serviceName) toBuilder()filterTraces(QueryRequest request) getTraces(QueryRequest request) getTraces(QueryRequest request) final class serviceAndSpanNames()serviceAndSpanNames()final class spanConsumer()spanConsumer()abstract SpanConsumerspanConsumer()final class autocompleteCardinality(int autocompleteCardinality) autocompleteKeys(List<String> keys) autocompleteTtl(int autocompleteTtl) abstract StorageComponent.BuildersearchEnabled(boolean searchEnabled) StorageComponent.traces()
+ supported.abstract StorageComponent.BuilderstrictTraceId(boolean strictTraceId) class final class abstract StorageComponentbuild()protected abstract StorageComponentdelegate()final class SpanConsumer.SpanConsumer.SpanConsumer.strict trace ID is disabled.SpanConsumer.SpanConsumer.SpanConsumer.SpanConsumer.SpanConsumer.Comparable<V1Annotation>Annotation.Annotation, except in v1 format the Span.localEndpoint() was
+ repeated for each annotation.intcompareTo(V1Annotation that) timestamp(), then value().static V1Annotationendpoint()booleaninthashCode()longvalue()Annotation.value()Annotation.value()In v2 format, this is analogous to Span.localEndpoint().
timestamp(), then value().compareTo in interface Comparable<V1Annotation>Comparable<V1BinaryAnnotation>Span.tags().v2 span data. Namely, this
+ supports Span.tags(), Span.localEndpoint() and Span.remoteEndpoint().
+
+ Specifically, this maps String and Boolean binary annotations, ignoring others.
static final intstatic final intintcompareTo(V1BinaryAnnotation that) keystatic V1BinaryAnnotationcreateAddress(String address,
+ Endpoint endpoint) Span.remoteEndpoint()static V1BinaryAnnotationcreateString(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.endpoint()stringValue() is present, this is the same as the Span.localEndpoint()
+ Otherwise, it is the same as the Span.remoteEndpoint().booleaninthashCode()key()Span.tags() v2 span tag}Span.tags() v2 span tag} or null if this is an addressinttype()Span.remoteEndpoint()Span.tags() except duplicating the
+ endpoint.
+
+ A special case is when the key is "lc" and value is empty: This substitutes for the Span.localEndpoint().
Span.tags() v2 span tag}Span.tags() v2 span tag} or null if this is an addressstringValue() is present, this is the same as the Span.localEndpoint()
+ Otherwise, it is the same as the Span.remoteEndpoint().keycompareTo in interface Comparable<V1BinaryAnnotation>addAnnotation(long timestamp,
+ String value,
+ Endpoint endpoint) V1Span.annotations()addBinaryAnnotation(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.addBinaryAnnotation(String address,
+ Endpoint endpoint) Span.remoteEndpoint()build()clear()V1Span.debug()duration(long duration) V1Span.duration()longid()V1Span.id()id(long id) V1Span.id()Span.Builder.id(String)V1Span.name()parentId(long parentId) V1Span.parentId()Span.Builder.parentId(String)timestamp(long timestamp) V1Span.timestamp()longtraceId()V1Span.traceId()traceId(long traceId) V1Span.traceId()Span.Builder.traceId(String)longV1Span.traceIdHigh()traceIdHigh(long traceIdHigh) V1Span.traceIdHigh()V1Span.traceIdHigh()V1Span.traceId()V1Span.id()Span.Builder.traceId(String)V1Span.traceId()V1Span.traceIdHigh()V1Span.id()Span.Builder.id(String)Span.Builder.parentId(String)V1Span.parentId()V1Span.name()V1Span.timestamp()V1Span.duration()V1Span.annotations()Span.remoteEndpoint()Span.tags() except duplicating the
+ endpoint.
+
+ A key of "lc" and empty value substitutes for Span.localEndpoint().
V1Span.debug()Span.Span.annotations(), except each may be associated with Span.localEndpoint()Span.tags() are allocated to binary annotations with a V1BinaryAnnotation.stringValue().debug()Span.debug()longduration()Span.durationAsLong()booleaninthashCode()longid()Span.id() except packed into a long.name()Span.name()static V1Span.BuilderlongparentId()id() or zero if this the root span in a trace.longSpan.timestampAsLong()longtraceId()Span.traceId()longSpan.traceId()Span.id() except packed into a long. Zero means root span.Span.name()id() or zero if this the root span in a trace.Span.timestampAsLong()Span.durationAsLong()Span.annotations(), except each may be associated with Span.localEndpoint()Span.tags() are allocated to binary annotations with a V1BinaryAnnotation.stringValue(). Span.remoteEndpoint() to those without.Span.debug()This type isn't thread-safe: it re-uses state to avoid re-allocations in conversion loops.
voidconvert(V1Span source,
+ Collection<Span> sink) static V1SpanConvertercreate()This type isn't thread-safe: it re-uses state to avoid re-allocations in conversion loops.
annotations()Span.annotations(), except each may be associated with Span.localEndpoint()intcompareTo(V1Annotation that) timestamp(), then value().static V1BinaryAnnotationcreateAddress(String address,
+ Endpoint endpoint) Span.remoteEndpoint()static V1BinaryAnnotationcreateString(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.binaryAnnotations()Span.tags() are allocated to binary annotations with a stringValue().intcompareTo(V1BinaryAnnotation that) keyaddAnnotation(long timestamp,
+ String value,
+ Endpoint endpoint) V1Span.annotations()addBinaryAnnotation(String key,
+ String value,
+ Endpoint endpoint) Span.tags() except duplicating the
+ endpoint.addBinaryAnnotation(String address,
+ Endpoint endpoint) Span.remoteEndpoint()clear()V1Span.debug()duration(long duration) V1Span.duration()id(long id) V1Span.id()Span.Builder.id(String)V1Span.name()static V1Span.BuildernewBuilder()parentId(long parentId) V1Span.parentId()Span.Builder.parentId(String)timestamp(long timestamp) V1Span.timestamp()traceId(long traceId) V1Span.traceId()Span.Builder.traceId(String)traceIdHigh(long traceIdHigh) V1Span.traceIdHigh()Annotation.Span.tags().Span.Annotation.Span.tags().Span.
Annotation.