APM, (Application Performance Management) tool for large-scale distributed systems.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @kjkmadness
We have moved our documentation to Gitbook. https://pinpoint-apm.gitbook.io/pinpoint
How to enable the experimental function
with data sampling, better memory usage, better performance, and more intuitive visualization
To provide more features via appling perfetto(https://github.com/google/perfetto).
Please refer to pinpoint gitbook
(Thank you @yjqg6666 for your contribution)
# support 2 types, COUNTING(default) and PERCENT.
# If this value set to COUNTING(default), sampling rate is 1/n.
# If this value set to PERCENT, sampling rate is n%.
profiler.sampling.type=PERCENT
# if it's PERCENT, then first x transactions out of y transactions will be sampled.
# Support from 100% to 0.01%
# eg. 100: 100% 50: 50% 0.01: 0.01%
profiler.sampling.percent.sampling-rate=100
Support TLS Connection for agent and collector
pinpoint-hbase2-collector-boot-2.3.1-p1.jar in download file is the version with the a887052e6599657ed93fdd8932a5e16056fa82ff feature.
to whom provided or suggested valuable features whom fixed or reported bugs whom showed interest in Pinpoint and shared it to others.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @aalinyu @Duytq7 @mariusssi @yjqg6666
# Pinpoint OPTS
-javaagent:${PINPOINT_BOOTSTRAP} -Dpinpoint.agentName=pinpoint_has_been_resolved_24_character_limit_agentName -Dpinpoint.applicationName=${APPLICATION_NAME}
In the future, the module is separated so that the batch is operated as a separate process instead of running the batch job in the web. Please refer to the guide document below for how to run batch.
to whom provided or suggested valuable features whom fixed or reported bugs whom showed interest in Pinpoint and shared it to others.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @brito-wang @cwJohnPark @davide-parini @doll6777 @hoverwinter @imbf @kkojaeh @linux0x5c @messi-gao @stanvl @tankilo @theLazyCat775
There is a bug in the Reactor-netty plugin (from 2.0.0 to 2.2.1) that to insert wrong endPoint. In general, there is no problem, but it can be a problem in situations where the collector is a high load environment. To prevent this, it is recommended to upgrade to version 2.2.2 or higher when using the Reactor-netty plugin. issue : #7659
[#7640] Solve compatibility with v2.3 collector module
[#7462] Add C/C++ servertype
[#7694] Skip handling, If there is no cpuload data in thrift stat
[#7679] Prevent errors that may occur due to lambda classes when executing retransform
[#7628] Fix profile.include
configuration not worked
[#7659] Fix inserting wrong endpoint in Reactor-netty plugin
If there is someone who was inadvertently excluded, please let me know. @snaigle
Server Map
Call Stack
issue : #7463
Compatibility
Pinpoint Agent supports OS environment variable.
$ PROFILER_SAMPLING_RATE=10
this.collectorSpanServerIp = profilerConfig.readString("profiler.collector.span.ip", DEFAULT_IP, placeHolderResolver);
this.collectorSpanServerPort = profilerConfig.readInt("profiler.collector.span.port", 9996);
@Value("${profiler.collector.span.ip}")
private String collectorSpanServerIp = DEFAULT_IP;
@Value("${profiler.collector.span.port}")
private int collectorSpanServerPort = 9996;
Fixed problem that gRPC-stream does not work with below messages. (#7375)
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:1
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:2
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:3
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:4
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:5
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:6
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:7
2020-10-23 11:21:57 [INFO ](c.n.p.p.s.g.SpanGrpcDataSender ) Discard PSpanMessage message, stream not ready. discardCount:8
==================================== ========== !! IMPORTANT ========== If you use grpc transport in Pinpoint Agent, it is strongly recommended to version 2.2.1.
issue : #7375
batch was run in the background of pinpoint-web server until v2.2.0. From v2.2.1 it will be dealt with in pinpoint-batch server. Since the batch logic(code) in pinpoint-web will be deprecated in the future, we advice you to transfer the execution of batch to pinpoint-batch server.
to whom provided or suggested valuable features whom fixed or reported bugs whom showed interest in Pinpoint and shared it to others.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @imbf @yjqg6666
## pinpoint.config
## AUTO, NIO, EPOLL
profiler.transport.grpc.agent.sender.channel-type=AUTO
## pinpoint-collector.properties
## AUTO, NIO, EPOLL
collector.receiver.grpc.span.channel-type=AUTO
AgentID is no longer a required confuguration. If AgentID is not present, the agent automatically generates it.
11-05 17:22:13.251 INFO AgentIdResolver : Failed to resolve AgentId(-Dpinpoint.agentId)
11-05 17:22:13.409 INFO AgentIdResolver : 'BYDnVJbPR6ODxGK61qTT1A' AgentId is automatically generated
11-05 17:22:13.409 INFO IdValidator : check SystemProperties(-D) applicationName:emeroad-app
11-05 17:22:13.410 INFO AgentIdResolver : SystemProperties(-D) emeroad-app=emeroad-app
[#7442] Fix NPE when SocketChannel state is connecting in KafkaPlugin [#7455] Fix race condation of PingStreamContext
Call Stack
issue : #7150 (Thank you @acafela for your contribution)
Add limiter on the number of link data and on build time for preventing OOM when drawing Servermap
# Limit number of link data
web.servermap.linkData.limit=500000000
# ApplicationMap build timeout in milliseconds
web.servermap.build.timeout=600000
Add limiter on the number of Span and SpanChunk Data when making Callstack
screenshot
configuration
# Limit number of Span and SpanChunk data
# If -1, there is no limit
web.callstack.selectSpans.limit=10000
Support that reuse string value in SpanDecoder.
# Limit number of string cache size in SpanMapper
# If -1, cache will be disabled
web.hbase.mapper.cache.string.size=-1
If AgentID is not present, the agent automatically generates it.
Determining whether to insert a header according to the broker in Kafka Plugin
to whom provided or suggested valuable features whom fixed or reported bugs whom showed interest in Pinpoint and shared it to others.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @acafela @dengliming @doll6777 @imbf @qingfeng @yjqg6666
Spring boot was applied to the web and collector. So it supports executable jar.
$ java -jar -Dpinpoint.zookeeper.address=localhost pinpoint-collector-boot-2.1.0.jar
$ java -jar -Dpinpoint.zookeeper.address=localhost pinpoint-web-boot-2.1.0.jar
# record HTTP request headers case-sensitive
# e.g. profiler.http.record.request.headers=X-AccessKey,X-Device-UUID
profiler.http.record.request.headers=user-agent,accept
# record HTTP request cookies(case-sensitive) in Cookie header
# e.g. profiler.http.record.request.cookies=userid,device-id,uuid
profiler.http.record.request.cookies=_ga
to whom provided or suggested valuable features whom fixed or reported bugs whom showed interest in Pinpoint and shared it to others.
Thank you all. If there is someone who was inadvertently excluded, please let me know. @acafela @cwJohnPark @dengliming @guillermomolina @heowc @imbf @Jaeyo @jbnimble @LightMingMing @messi-gao @sumniy @tlqaksqhr @yjqg6666