You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cxf.apache.org by bu...@apache.org on 2017/01/30 22:47:06 UTC
svn commit: r1005944 - in /websites/production/cxf/content:
cache/docs.pageCache docs/using-apache-htrace.html
Author: buildbot
Date: Mon Jan 30 22:47:06 2017
New Revision: 1005944
Log:
Production update by buildbot for cxf
Modified:
websites/production/cxf/content/cache/docs.pageCache
websites/production/cxf/content/docs/using-apache-htrace.html
Modified: websites/production/cxf/content/cache/docs.pageCache
==============================================================================
Binary files - no diff available.
Modified: websites/production/cxf/content/docs/using-apache-htrace.html
==============================================================================
--- websites/production/cxf/content/docs/using-apache-htrace.html (original)
+++ websites/production/cxf/content/docs/using-apache-htrace.html Mon Jan 30 22:47:06 2017
@@ -117,19 +117,19 @@ Apache CXF -- Using Apache HTrace
<!-- Content -->
<div class="wiki-content">
<div id="ConfluenceContent"><p><style type="text/css">/*<![CDATA[*/
-div.rbtoc1445395593960 {padding: 0px;}
-div.rbtoc1445395593960 ul {list-style: disc;margin-left: 0px;}
-div.rbtoc1445395593960 li {margin-left: 0px;padding-left: 0px;}
+div.rbtoc1485816393306 {padding: 0px;}
+div.rbtoc1485816393306 ul {list-style: disc;margin-left: 0px;}
+div.rbtoc1485816393306 li {margin-left: 0px;padding-left: 0px;}
-/*]]>*/</style></p><div class="toc-macro rbtoc1445395593960">
-<ul class="toc-indentation"><li><a shape="rect" href="#UsingApacheHTrace-Overview">Overview</a></li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracinginNutshell">Distributed Tracing in Nutshell</a></li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracinginApacheCXF">Distributed Tracing in Apache CXF</a></li><li><a shape="rect" href="#UsingApacheHTrace-ConfiguringClientconfigure.client">Configuring Client</a>
+/*]]>*/</style></p><div class="toc-macro rbtoc1485816393306">
+<ul class="toc-indentation"><li><a shape="rect" href="#UsingApacheHTrace-Overview">Overview</a></li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracinginNutshell">Distributed Tracing in Nutshell</a></li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracinginApacheCXFusingApacheHTrace">Distributed Tracing in Apache CXF using Apache HTrace</a></li><li><a shape="rect" href="#UsingApacheHTrace-ConfiguringClientconfigure.client">Configuring Client</a>
<ul class="toc-indentation"><li><a shape="rect" href="#UsingApacheHTrace-Configuringtracingheadernames">Configuring tracing header names</a></li></ul>
</li><li><a shape="rect" href="#UsingApacheHTrace-ConfiguringServerconfigure.server">Configuring Server</a>
<ul class="toc-indentation"><li><a shape="rect" href="#UsingApacheHTrace-Configuringtracingheadernames.1">Configuring tracing header names</a></li></ul>
</li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracingInAction:UsageScenarios">Distributed Tracing In Action: Usage Scenarios</a>
<ul class="toc-indentation"><li><a shape="rect" href="#UsingApacheHTrace-Example#1:ClientandServerwithdefaultdistributedtracingconfigured">Example #1: Client and Server with default distributed tracing configured</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#2:ClientandServerwithnestedtrace">Example #2: Client and Server with nested trace</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#3:ClientandServertracewithtimeline">Example #3: Client and Server trace with timeline</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#4:ClientandServerwithannotatedtrace(key/value)">Example #4: Client and Server with annotated trace (key/value)</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#5:ClientandServerwithparalleltrace(involvingthreadpools)">Example #5: Client and Server with parallel trace (involving thread pools)</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#6:ClientandServerwithasynchronousJAX-RSservice(server-side)">Exampl
e #6: Client and Server with asynchronous JAX-RS service (server-side)</a></li><li><a shape="rect" href="#UsingApacheHTrace-Example#7:ClientandServerwithasynchronousinvocation(client-side)">Example #7: Client and Server with asynchronous invocation (client-side)</a></li></ul>
</li><li><a shape="rect" href="#UsingApacheHTrace-DistributedTracingandJAX-WSsupport">Distributed Tracing and JAX-WS support</a></li><li><a shape="rect" href="#UsingApacheHTrace-FutureWork">Future Work</a></li></ul>
-</div><h1 id="UsingApacheHTrace-Overview">Overview</h1><p><a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> is a tracing framework intended for use with distributed systems written in java. Since version <strong>3.1.3</strong>, Apache CXF fully supports integration with <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a>, both on client side and server side. This section gives a complete overview on how distributed tracing support is supported in JAX-RS applications built on top of Apache CXF.</p><h1 id="UsingApacheHTrace-DistributedTracinginNutshell">Distributed Tracing in Nutshell</h1><p>Distributed tracing, first described by Google in <a shape="rect" class="external-link" href="http://research.google.com/pubs/pub36356.html" rel="nofollow">Dapper, a Large-Scale Distributed Systems Tracing Infrastructure</a> paper became increasingly important topic these days. With
microservices (aka SOA) gaining more and more adoption, the typical applications are built using dozens or even hundreds of small, distributed pieces. The end-to-end traceability of the requests (or any kind of work performed on user's behalf) is hard task to accomplish, particularly taking into account asyncronous or/and concurrent invocations. <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> is inspired by <a shape="rect" class="external-link" href="http://research.google.com/pubs/pub36356.html" rel="nofollow">Dapper, a Large-Scale Distributed Systems Tracing Infrastructure</a> paper and essentially is a full-fledged distributed tracing framework.</p><p>Distributed tracing is additional instrumentation layer on top of new or existing applications. In terms of distributed tracing, <strong>span</strong> represents a basic unit of work. For example, executing database query is a <strong>span</strong>. <strong>Spans</strong>
are identified by a unique 128-bit ID. <strong>Spans</strong> also have other data, such as <strong>descriptions</strong>, <strong>timelines</strong>,<strong> key-value annotations</strong>, the <strong>ID</strong> of the <strong>span</strong> that caused them (parent), and <strong>process/tracer</strong> ID’s (normally IP address and process name). Spans are started and stopped, and they keep track of their timing information. Once <strong>span</strong> is created, it should be stopped at some point in the future. In turn, <strong>trace</strong> is a set of spans forming a tree-like structure. For example, if you are running a JAX-RS service, a trace might be formed by a <strong>PUT</strong> request and downstream work.</p><p>From implementation prospective, and in context of Java applications, <strong>spans</strong> are attached to their threads (in general, thread which created the <strong>span</strong> should close it). However it is possible to transfer <strong>spans</str
ong> from thread to thread in order to model a complex execution flows. It is also possible to have many <strong>spans</strong> in the same thread, as long as they are properly created and closed. In the next sections we are going to see the examples of that.</p><p>Another two important concepts in context of distributed tracing are <strong>span receivers</strong> and <strong>samplers</strong>. Essentially, all spans (including start/stop time, key/value annotations, timelines, ..) should be persisted (or collected) somewhere. <strong>Span receiver</strong> is a collector within a process that is the destination of <strong>spans</strong> when a trace is running (it could be a console, local file, data store, ...). <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> provides span receivers for <a shape="rect" class="external-link" href="http://hbase.apache.org">Apache HBase</a>, <a shape="rect" class="external-link" href="https
://flume.apache.org/">Apache Flume</a> and <a shape="rect" class="external-link" href="http://zipkin.io/" rel="nofollow">Twitter Zipkin</a>. From other side, <strong>samplers</strong> allow to control the frequency of the tracing (all the time, never, probability driven, ...). Using the <strong>sampler</strong> is the way to minimize tracing overhead (or just amount of traces) by limiting them to particular conditions.</p><h1 id="UsingApacheHTrace-DistributedTracinginApacheCXF">Distributed Tracing in Apache CXF</h1><p><a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> is a very popular framework for building services and web APIs. No doubts, it is going to play even more important role in context of microservices architecture letting developers to quickly build and deploy individual JAX-RS/JAX-WS services. As it was just mentioned before, distributed tracing is an essential technique to monitor the application as whole, breaking the request to individual service traces as
it goes through and crosses the boundaries of threads, processes and machines.</p><p>The current integration of distributed tracing in <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> supports <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> (<strong>4.x+</strong> release branch) only in JAX-RS 2.x applications. From high-level prospective, it consists of three main parts:</p><ul style="list-style-type: square;"><li><strong>TracerContext</strong> (injectable through <strong>@Context</strong> annotation)</li><li><strong>HTraceProvider</strong> (server-side JAX-RS provider) and <strong>HTraceClientProvider</strong> (client-side JAX-RS provider)</li><li><strong>HTraceFeature</strong> (server-side <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> feature to simplify <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> configuration and integration)</
li></ul><p><a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> uses HTTP headers to hand off tracing context from the client to the service and from the service to service. Those headers are used internally by <strong>HTraceProvider</strong> and <strong>HTraceClientProvider</strong>, but are configurable. The default header names are declared in the TracerHeaders class:</p><ul style="list-style-type: square;"><li><strong>X-Span-Id</strong>: contains a current span ID</li></ul><p>By default, <strong>HTraceProvider</strong> will try to pass the currently active <strong>span</strong> through HTTP headers on each service invocation. If there is no active spans, the new span will be created and passed through HTTP headers on per-invocation basis. Essentially, just registering the <strong>HTraceProvider</strong> on the client and <strong>HTraceClientProvider</strong> on the server is enough to have tracing context to be properly passed everywhere. The only configuration part whic
h is necessary are <strong>span receiver(s)</strong> and <strong>sampler</strong>(s).</p><p>It is also worth to mention the way <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> attaches the description to <strong>spans</strong>. With regards to the client integration, the description becomes a full URL being invoked prefixed by HTTP method, for example: <strong>GET </strong><a shape="rect" class="external-link" href="http://localhost:8282/books" rel="nofollow"><strong>http://localhost:8282</strong>/books</a>. On the server side integration, the description becomes a relation JAX-RS resource path prefixed by HTTP method, f.e.: <strong>GET books, POST book/123</strong></p><h1 id="UsingApacheHTrace-ConfiguringClientconfigure.client">Configuring Client <span class="confluence-anchor-link" id="UsingApacheHTrace-configure.client"></span></h1><p>There are a couple of ways the JAX-RS client could be configured, depending on the client implementation. <a shape="rect" href="http:/
/cxf.apache.org/">Apache CXF</a> provides its own <strong>WebClient</strong> which could be configured just like that (in future versions, there would be a simpler ways to do that using client specific features):</p><div class="code panel pdl" style="border-width: 1px;"><div class="codeContent panelContent pdl">
+</div><h1 id="UsingApacheHTrace-Overview">Overview</h1><p><a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> is a tracing framework intended for use with distributed systems written in java. Since version <strong>3.1.3</strong>, Apache CXF fully supports integration with <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a>, both on client side and server side. This section gives a complete overview on how distributed tracing support is supported in JAX-RS applications built on top of Apache CXF.</p><h1 id="UsingApacheHTrace-DistributedTracinginNutshell">Distributed Tracing in Nutshell</h1><p>Distributed tracing, first described by Google in <a shape="rect" class="external-link" href="http://research.google.com/pubs/pub36356.html" rel="nofollow">Dapper, a Large-Scale Distributed Systems Tracing Infrastructure</a> paper became increasingly important topic these days. With
microservices (aka SOA) gaining more and more adoption, the typical applications are built using dozens or even hundreds of small, distributed pieces. The end-to-end traceability of the requests (or any kind of work performed on user's behalf) is hard task to accomplish, particularly taking into account asyncronous or/and concurrent invocations. <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> is inspired by <a shape="rect" class="external-link" href="http://research.google.com/pubs/pub36356.html" rel="nofollow">Dapper, a Large-Scale Distributed Systems Tracing Infrastructure</a> paper and essentially is a full-fledged distributed tracing framework.</p><p>Distributed tracing is additional instrumentation layer on top of new or existing applications. In terms of distributed tracing, <strong>span</strong> represents a basic unit of work. For example, executing database query is a <strong>span</strong>. <strong>Spans</strong>
are identified by a unique 128-bit ID. <strong>Spans</strong> also have other data, such as <strong>descriptions</strong>, <strong>timelines</strong>,<strong> key-value annotations</strong>, the <strong>ID</strong> of the <strong>span</strong> that caused them (parent), and <strong>process/tracer</strong> ID’s (normally IP address and process name). Spans are started and stopped, and they keep track of their timing information. Once <strong>span</strong> is created, it should be stopped at some point in the future. In turn, <strong>trace</strong> is a set of spans forming a tree-like structure. For example, if you are running a JAX-RS service, a trace might be formed by a <strong>PUT</strong> request and downstream work.</p><p>From implementation prospective, and in context of Java applications, <strong>spans</strong> are attached to their threads (in general, thread which created the <strong>span</strong> should close it). However it is possible to transfer <strong>spans</str
ong> from thread to thread in order to model a complex execution flows. It is also possible to have many <strong>spans</strong> in the same thread, as long as they are properly created and closed. In the next sections we are going to see the examples of that.</p><p>Another two important concepts in context of distributed tracing are <strong>span receivers</strong> and <strong>samplers</strong>. Essentially, all spans (including start/stop time, key/value annotations, timelines, ..) should be persisted (or collected) somewhere. <strong>Span receiver</strong> is a collector within a process that is the destination of <strong>spans</strong> when a trace is running (it could be a console, local file, data store, ...). <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> provides span receivers for <a shape="rect" class="external-link" href="http://hbase.apache.org">Apache HBase</a>, <a shape="rect" class="external-link" href="https
://flume.apache.org/">Apache Flume</a> and <a shape="rect" class="external-link" href="http://zipkin.io/" rel="nofollow">Twitter Zipkin</a>. From other side, <strong>samplers</strong> allow to control the frequency of the tracing (all the time, never, probability driven, ...). Using the <strong>sampler</strong> is the way to minimize tracing overhead (or just amount of traces) by limiting them to particular conditions.</p><h1 id="UsingApacheHTrace-DistributedTracinginApacheCXFusingApacheHTrace">Distributed Tracing in Apache CXF using Apache HTrace</h1><p><a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> is a very popular framework for building services and web APIs. No doubts, it is going to play even more important role in context of microservices architecture letting developers to quickly build and deploy individual JAX-RS/JAX-WS services. As it was just mentioned before, distributed tracing is an essential technique to monitor the application as whole, breaking the req
uest to individual service traces as it goes through and crosses the boundaries of threads, processes and machines.</p><p>The current integration of distributed tracing in <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> supports <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace</a> (<strong>4.x+</strong> release branch) only in JAX-RS 2.x applications. From high-level prospective, it consists of three main parts:</p><ul style="list-style-type: square;"><li><strong>TracerContext</strong> (injectable through <strong>@Context</strong> annotation)</li><li><strong>HTraceProvider</strong> (server-side JAX-RS provider) and <strong>HTraceClientProvider</strong> (client-side JAX-RS provider)</li><li><strong>HTraceFeature</strong> (server-side <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> feature to simplify <a shape="rect" class="external-link" href="http://htrace.incubator.apache.org/index.html">Apache HTrace
</a> configuration and integration)</li></ul><p><a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> uses HTTP headers to hand off tracing context from the client to the service and from the service to service. Those headers are used internally by <strong>HTraceProvider</strong> and <strong>HTraceClientProvider</strong>, but are configurable. The default header names are declared in the TracerHeaders class:</p><ul style="list-style-type: square;"><li><strong>X-Span-Id</strong>: contains a current span ID</li></ul><p>By default, <strong>HTraceProvider</strong> will try to pass the currently active <strong>span</strong> through HTTP headers on each service invocation. If there is no active spans, the new span will be created and passed through HTTP headers on per-invocation basis. Essentially, just registering the <strong>HTraceProvider</strong> on the client and <strong>HTraceClientProvider</strong> on the server is enough to have tracing context to be properly passed everywh
ere. The only configuration part which is necessary are <strong>span receiver(s)</strong> and <strong>sampler</strong>(s).</p><p>It is also worth to mention the way <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> attaches the description to <strong>spans</strong>. With regards to the client integration, the description becomes a full URL being invoked prefixed by HTTP method, for example: <strong>GET </strong><a shape="rect" class="external-link" href="http://localhost:8282/books" rel="nofollow"><strong>http://localhost:8282</strong>/books</a>. On the server side integration, the description becomes a relation JAX-RS resource path prefixed by HTTP method, f.e.: <strong>GET books, POST book/123</strong></p><h1 id="UsingApacheHTrace-ConfiguringClientconfigure.client">Configuring Client <span class="confluence-anchor-link" id="UsingApacheHTrace-configure.client"></span></h1><p>There are a couple of ways the JAX-RS client could be configured, depending on the client impleme
ntation. <a shape="rect" href="http://cxf.apache.org/">Apache CXF</a> provides its own <strong>WebClient</strong> which could be configured just like that (in future versions, there would be a simpler ways to do that using client specific features):</p><div class="code panel pdl" style="border-width: 1px;"><div class="codeContent panelContent pdl">
<pre class="brush: java; gutter: false; theme: Default" style="font-size:12px;">final Map<String, String> properties = new HashMap<String, String>();
properties.put(Tracer.SPAN_RECEIVER_CLASSES_KEY, ...);
properties.put(Tracer.SAMPLER_CLASSES_KEY, ...);