Note
|
Release notes are best read in our documentation at elastic.co |
-
Prevent NPE in OpenTelemetry metrics bridge in case of asynchronous agent start - {pull}3880[#3880]
-
Fix random Weblogic ClassNotFoundException related to thread context classloader - {pull}3870[#3870]
-
Skips using NOFOLLOW_LINKS file open option when running on OS/400 as it’s unsupported there - {pull}3905[#3905]
-
Fix JMX metric warning message about unsupported composite value types - {pull}3849[#3849]
-
Fix JAX-WS transaction naming for @WebMethod annotated methods - {pull}3850[#3850]
-
Fix log4j2 log correlation with shaded application jar - {pull}3764[#3764]
-
Improve automatic span class name detection for Scala and nested/anonymous classes - {pull}3746[#3746]
-
Added experimental option to capture HTTP client request bodies for Apache Http Client v4 and v5, HttpUrlConnection and Spring WebClient - {pull}3776[#3776], {pull}3962[#3962], {pull}3724[#3724], {pull}3754[#3754], {pull}3767[#3767]
-
Agent health metrics now GA - {pull}3802[#3802]
-
Restore compatibility with Java 7 - {pull}3657[#3657]
-
Avoid
ClassCastException
and issue warning when trying to use otel span links - {pull}3672[#3672] -
Avoid
NullPointerException
with runtime attach API and invalid map entries - {pull}3712[#3712] -
Enhance invalid state JMX metrics handling - {pull}3713[#3713]
-
Skips using NOFOLLOW_LINKS file open option when running on z/OS as it’s unsupported there - {pull}3722[#3722]
-
Added option to make routing-key part of RabbitMQ transaction/span names - {pull}3636[#3636]
-
Added internal option for capturing request bodies for apache httpclient v4 - {pull}3692[#3692]
-
Added automatic module name to apm-agent-attach - {pull}3743[#3743]
-
Added support for correlating APM data with elastic universal profiling data - {pull}3615[#3615], {pull}3602[#3602], {pull}3607[#3607], {pull}3598[#3598]
-
Excluded latest AppDynamics packages from instrumentation (
com.cisco.mtagent.*
) - {pull}3632[#3632]
-
Fixed edge case where inferred spans could cause cycles in the trace parent-child relationships, subsequently resulting in the UI crashing - {pull}3588[#3588]
-
Fix NPE in dropped spans statistics - {pull}3590[#3590]
-
Fix too small activation stack size for small
transaction_max_spans
values - {pull}3643[#3643]
-
Fixed problems with public API annotation inheritance - {pull}3551[#3551]
-
Differentiate Lambda URLs from API Gateway in AWS Lambda integration - {pull}3417[#3417]
-
Added lambda support for ELB triggers {pull}#3411[#3411]
-
Add exclusion list option for calling DatabaseMetaData.getUserName - {pull}#3568[#3568]
-
Avoid another case where we might touch application exceptions for
safe_exceptions
- {pull}3553[#3553] -
More robust hostname detection on Windows - {pull}3556[#3556]
-
Bumped base alpine docker image version - {pull}3524[#3524]
-
Replace statement parser cache with an LRU cache to improve efficiency in certain cases {pull}3492[#3492]
-
Added missing support for TracerBuilder in OpenTelemetry bridge - {pull}3535[#3535]
-
Fixed some locations to not touch exceptions when
safe_exception
is configured - {pull}3543[#3543]
-
The minimum supported OpenTelemetry version has been increased to 1.4.0 - {pull}3535[#3535]
-
Added internal
safe_exceptions
config option to workaround JVM bugs related to touching exceptions - {pull}3528[#3528]
-
Cleanup extra servlet request attribute used for Spring exception handler - {pull}3527[#3527]
-
Added a configuration option to use queues in names of spring-rabbit transactions - {pull}3424[#3424]
-
Add option to retry JMX metrics capture in case of exception - {pull}3511[#3511]
-
Add support to CLI attach download for new agent signature for 1.46.0+ - {pull}3513[#3513]
-
Added support for OpenTelementry Attributes db.statement and db.user - {pull}3475[#3475]
-
Fixed NPE in ApacheHttpClientApiAdapter#getHostName - {pull}3479[#3479]
-
Fix span stack trace when combined with span compression - {pull}3474[#3474]
-
Fix
UnsupportedClassVersionError
java 7 compatibility for jctools - {pull}3483[#3483]
-
Added support for OpenTelemetry annotations -
WithSpan
andSpanAttribute
- {pull}3406[#3406] -
Only automatically apply redacted exceptions for Corretto JVM 17-20. Outside that, user should use capture_exception_details=false to workaround the JVM race-condition bug if it gets triggered: {pull}3438[#3438]
-
Added support for Spring 6.1 / Spring-Boot 3.2 - {pull}3440[#3440]
-
Add support for Apache HTTP client 5.x - {pull}3419[#3419]
-
Added support for OpenTelemetry
1.32.0
. As a result,custom_metrics_histogram_boundaries
will not work when you bring your ownMeterProvider
from an SDK with version1.32.0
or newer. As a workaround, you should manually register a corresponding View in yourMeterProvider
. Note that this change will not affect you, if you are using the OpenTelemetry API only and not the SDK. - {pull}3447[#3447]
-
Added protection against invalid timestamps provided by manual instrumentation - {pull}3363[#3363]
-
Added support for AWS SDK 2.21 - {pull}3373[#3373]
-
Capture bucket and object key to Lambda transaction as OTel attributes -
aws.s3.bucket
,aws.s3.key
- {pull}3364[#3364] -
Added
context_propagation_only
configuration option - {pull}3358[#3358] -
Added attribute[] for JMX pattern metrics (all metrics can now be generated with
object_name[
:type=,name=] attribute[*]
) - {pull}3376[#3376]
-
Fixed too many spans being created for
HTTPUrlConnection
requests with methodHEAD
- {pull}3353[#3353] -
Enhance k8s container/pod and host name detection heuristics - {pull}3418[#3418]
-
Add support for Elasticsearch client 8.9 - {pull}3283[#3283]
-
Added
baggage_to_attach
config option to allow automatic lifting of baggage into transaction, span and error attributes - {pull}3288[#3288], {pull}3289[#3289] -
Exclude elasticsearch 8.10 and newer clients from instrumentation because they natively support OpenTelemetry - {pull}3303[#3303]
-
Switched to OpenTelemetry compatible context propagation for Kafka - {pull}3300[#3300]
-
Changed
cloud.project.id
collected in Google Cloud (GCP) to be theproject-id
- {issues}3311[#3311] -
Allow running the IntelliJ debug agent in parallel - {pull}3315[#3315]
-
Capture
span.sync
=false
for ES restclient async spans plugins
-
Prevent bad serialization in edge cases for span compression - {pull}3293[#3293]
-
Allow overriding of transaction type for Servlet-API transactions - {pull}3226[#3226]
-
Fix micrometer histogram serialization - {pull}3290[#3290], {pull}3304[#3304]
-
Fix transactions not being correctly handled in certain edge cases - {pull}3294[#3294]
-
Fixed JDBC instrumentation for DB2 - {pull}3313[#3313]
-
Fixed OpenTelemetry metrics export breaking when
instrument=false
is configured - {pull}3326[#3326]
-
Virtual thread support - {pull}3244[#3244], {pull}3286[#3286]
-
Include
application_packages
in JMS listener naming heuristic - {pull}3299[#3299]
-
Fix JVM memory usage capture - {pull}3279[#3279]
-
Fixed Micrometer histograms to be correctly exported with non-cumulative bucket counts - {pull}3264[#3264]
-
Fixed SQS NoClassDefFoundError in AWS SDK instrumentation for async clients - {pull}3266[#3266]
-
Replaced thread-local IO buffers with pooled ones for virtual thread friendliness - {pull}#3239[#3239]
-
Added W3C baggage propagation - {pull}3236[#3236], {pull}3248[#3248]
-
Added support for baggage in OpenTelemetry bridge - {pull}3249[#3249]
-
Improved span naming and attribute collection for 7.16+ elasticsearch clients - {pull}3157[#3157]
-
Fixed SQS NoClassDefFoundError in AWS SDK instrumentation - {pull}3254[#3254]
-
Fixed reference counting issues in elasticsearch instrumentation - {pull}3256[#3256]
-
Capture
container.id
for cgroups v2 - {pull}3199[#3199]
-
fix jakarta.jms support (wasn’t fully implemented) - {pull}3198[#3198]
-
Fixed agent programmatic attach with immutable config - {pull}3170[#3170]
-
Prevent overriding
ELASTIC_APM_AWS_LAMBDA_HANDLER
in AWS lambda execution when explicitly set - {pull}3205[#3205] -
Ignore gc allocation metrics when unsupported - {pull}3225[#3225]
-
Avoid warning log message when grpc transactions are cancelled before end - {pull}3223[#3223]
-
Align agent hostname capture to FQDN - {pull}3188[#3188]
-
Capture S3 operation details as OTel attributes - {pull}3136[#3136]
-
Added support for recording AWS lambda transactions even if the JVM crashes or runs into a timeout - {pull}3134[#3134]
-
Add extra built-in metrics:
jvm.fd.
andjvm.memory.pool.non_heap.
- {pull}[#3147] -
Capture
span.sync
=false
for some async spans plugins - {pull}[#3164]
-
Fixed classloading for OpenTelemetry dependencies in external plugins - {pull}3154[#3154]
-
Handled an edge case where exceptions thrown by instrumentation code could escape into the application - {pull}3159[#3159]
-
Added guard to gracefully handle the presence of pre 3.0 Servlet API versions in the spring service name discovery mechanism - {pull}3172[#3172]
-
Added tests for Quarkus / RestEasy, adjusted vert.x router transaction name priority - {pull}1765[#1765]
-
Added support for Spring WebMVC 6.x and Spring Boot 3.x - {pull}3094[#3094]
-
Added
service.environment
to logs for service correlation - {pull}3115[#3115] -
Optimize agent overhead when an excessive number of spans is created with
trace_methods
,@Traced
or@CaptureSpan
- {pull}3151[#3151]
-
Do not use proxy to retrieve cloud metadata - {pull}3108[#3108]
-
The list of instrumentation groups for log-related features have been simplified to
logging
,log-reformatting
,log-correlation
andlog-error
, as a consequence any agent configuration withenable_instrumentations
ordisable_instrumentations
might require adjustments as the following instrumentation groups no longer exist:jul-ecs
,jul-error
,log4j1-correlation
,log4j1-ecs
,log4j1-error
,log4j2-correlation
,log4j2-ecs
,log4j2-error
,logback-correlation
,logback-ecs
andslf4j-error
.
-
Add the [config-disable-outgoing-tracecontext-headers] config option to disable injection of
tracecontext
on outgoing communication - {pull}2996[#2996] -
Add the [config-profiling-inferred-spans-logging-enabled] config option to suppress async profiler warning messages - {pull}3002[#3002]
-
Added support for OpenTelemetry metrics - {pull}2968[#2968], {pull}3014[#3014]
-
Added agent.activation_method telemetry - {pull}2926[#2926]
-
Allow creation of exit spans with
@CaptureSpan
and@Traced
annotations - {pull}3046[#3046] -
Add the [config-long-field-max-length] config to enable capturing larger values for specific fields - {pull}3027[#3027]
-
Provide fallback correlation when
ecs-logging-java
is used - {pull}3064[#3064] -
Added separate Java 8 build with updated log4j2 - {pull}3076[#3076]
-
Add elasticsearch_capture_body_urls option to customize which Elasticsearch request bodies are captured - {pull}3091[#3091]
-
Fixed used instrumentations printed on shutdown {pull}3001[#3001]
-
Prevent potential connection leak on network failure - {pull}2869[#2869]
-
Fix for inferred spans where the parent id was also a child id - {pull}2686[#2686]
-
Fix context propagation for async 7.x and 8.x Elasticsearch clients - {pull}3015[#3015]
-
Fix exceptions filtering based on [config-ignore-exceptions] when those are nested - {pull}3025[#3025]
-
Fix usage of
HttpUrlConnection.getResponseCode()
causing an error event due to exception capturing, even when it is internally handled - {pull}3024[#3024] -
Fix source code jar to contain apm-agent sources - {pull}3063[#3063]
-
Fix security exception when security manager is used with
log_level=debug
- {pull}3077[#3077] -
Fix slim attacher when downloading agent version - {pull}3096[#3096]
-
Previously, agent instrumented all implementations of
javax.jms.MessageListener
, which could cause performance issues. From now on, the instrumentation is limited to implementations that fit at least one of the following criteria:-
have
Message
orListener
in their class name -
is an anonymous inner class or a lambda
-
is within a package provided by the
jms_listener_packages
configuration.
-
-
The
context.destination.service.resource
span field has been changed for AWS S3, SQS and DynamoDB. As a result, the history of metrics relying on this field might break. - {pull}2947[#2947]
-
Add experimental log sending from the agent with
log_sending
- {pull}2694[#2694] -
Add bootstrap checks that enable [jvm-filtering] on startup - {pull}2951[#2951]
-
Added support for LDAP - {pull}2977[#2977]
-
Use
127.0.0.1
as default forserver_url
to prevent ipv6 ambiguity - {pull}2927[#2927] -
Fix some span-compression concurrency issues - {pull}2865[#2865]
-
Add warning when agent is accidentally started on a JVM/JDK command-line tool - {pull}2924[#2924]
-
Fix
NullPointerException
caused by the Elasticsearch REST client instrumentation when collecting dropped span metrics - {pull}2959[#2959] -
Fix SQS Instrumentation for Non-MessageReceive actions to avoid NoSuchElementException - {pull}2979[#2979]
-
Fix
java.lang.NoSuchMethodError
when using the agent with WebFlux and Spring 6.x/Spring Boot 3.x - {pull}2935[#2935] -
Optimize JMS listener matcher - {pull}2930[#2930]
-
Handle Corretto causing a sigsegv when accessing Throwables inside some instrumentation with AWS - {pull}2958[#2958]
-
Change Micrometer logs to DEBUG from INFO - {pull}2914[#2914]
-
Add support for log correlation for
java.util.logging
(JUL) - {pull}2724[#2724] -
Add support for spring-kafka batch listeners - {pull}2815[#2815]
-
Improved instrumentation for legacy Apache HttpClient (when not using an
HttpUriRequest
, such asBasicHttpRequest
) -
Prevented exclusion of agent-packages via
classes_excluded_from_instrumentation
to avoid unintended side effects -
Add Tomcat support for log reformatting - {pull}2839[#2839]
-
Capture Elastic cluster name on Elastic Cloud - {pull}2796[#2796]
-
Attacher CLI: added a
--no-fork
config to opt out from executing a forked process as different user.If--no-fork
is used alongside discovery rules that contain only--include-pid
rules, the attacher will not execute JVM discovery - {pull}2863[#2863] -
Add the option to instrument very old bytecode through the
instrument_ancient_bytecode
config option - {pull}2866[#2866] -
Capture MongoDB statements - {pull}2806[#2806]
-
Added agent health and background overhead metrics (experimental) - {pull}2864[#2864], {pull}2888[#2888]
-
Added support for Finagle Http Client - {pull}2795[#2795]
-
Add support for Apache HTTP client 3.x - {pull}2853[#2853]
-
Made
api_key
andsecret_token
configuration options dynamic - {pull}2889[#2889] -
Misaligned micrometer interval vs Elastic agent metrics reporting interval now handled - {pull}2801[#2801]
-
Histograms now reported from micrometer metrics - {pull}2895[#2895]
-
Remove
context.db
fields from S3 instrumentation - {pull}2821[#2821] -
Allowed OpenTelemetry
Span.updateName
to update names provided by elastic - {pull}2838[#2838] -
Prevent random
NullPointerException
when span compression is not possible - {pull}2859[#2859] -
Fix security manager issues on OpenJDK 17, with errors like:
java.lang.UnsupportedOperationException: Could not access Unsafe class
- {pull}2874[#2874] -
Fix security manager compatibility with Tomcat - {pull}2871[#2871] and {pull}2883[#2883]
-
Fix NPE with Spring MVC - {pull}2896[#2896]
-
Redact
principal
headers by default - {pull}2798[#2798]
-
Activation stack was extracted from
ElasticApmTracer
into a separate class, where it also enforces a stack depth to eliminate activation leaks - {pull}2783[#2783]
-
Fix imports (leading to
NoClassDefFoundError
) in the AWS SDK instrumentation - {pull}2800[#2800]
-
Changed the main agent class loader to work in a child-first delegation model, thus making it more isolated by preferring self packaged version of classes that are available also in the parent (bootstrap) class loader - {pull}2728[#2728]
-
Capture Oracle SID in connection string - {pull}2709[#2709]
-
Implemented span links in the OTel bridge - {pull}2685[#2685]
-
Added support for MongoDB 4.x Sync Driver - {pull}2241{#2241}
-
Capture keyspace in
db.instance
for Cassandra database - {pull}2684[#2684] -
Added support for AWS SQS - {pull}2637[#2637]
-
Add
trace_continuation_strategy
configuration option - {pull}2760[#2760] -
Capture user from Azure SSO with Servlet-based app containers - {pull}2767[#2767]
-
Promote WebFlux & Reactor to GA and enable it by default - {pull}2782[#2782]
-
Fix unexpected side effects of
toString
calls within reactor instrumentation - {pull}2708[#2708] -
Fix Vert.x instrumentation for 4.3.2 - {pull}2700[#2700]
-
Fix
NullPointerException
inAmazonHttpClientInstrumentation
- {pull}2740[#2740] -
Fix stack frame exclusion patterns - {pull}2758[#2758]
-
Fix
NullPointerException
with compressed spans - {pull}2755[#2755] -
Fix empty Servlet path with proper fallback - {pull}2748[#2748]
-
Fix
ClosedByInterruptException
during indy bootstrap method resolution - {pull}2752[#2752] -
Enhance exclusion of other APM agents - {pull}2766[#2766]
-
Avoid warning when log correlation option is provided in remote config - {pull}2765[#2765]
-
Update
async-profiler
to 1.8.8 to avoid missing symbol log spam - {pull}2775[#2775] -
Fix container ID discovery for containers managed through AWS Fargate - {pull}2772[#2772]
-
Make
traceparent
header computation thread-safe - {pull}2747[#2747] -
Fix OTel bridge with multiple OTel APIs or external plugins - {pull}2735[#2735]
As of version 1.33.0, Java 7 support is deprecated and will be removed in a future release (not expected to be removed before January 2024) - {pull}2677[#2677]
-
Add support for Spring WebClient - {pull}2229[#2229]
-
Added undocumented and unsupported configuration
metric_set_limit
to increase the metric set limit - {pull}2148[#2148] -
Added
transaction_name_groups
configuration option - {pull}2676[#2676]-
Deprecated
url_groups
in favor oftransaction_name_groups
.
-
-
Fix for JAX-WS (SOAP) transaction names. The agent now properly names transaction for web service methods that are not annotated with
@WebMethod
. - {pull}2667[#2667] -
Fix public API backward compatibility that was broken in 1.32.0. With this version you can use any version of the public API once again - {pull}2682[#2682]
-
Fix flaky transaction name with Webflux+Servlet - {pull}2695[#2695]
-
If using the public API of version < 1.32.0 and using the
@CaptureSpan
or@Traced
annotations, upgrading the agent to 1.32.0 without upgrading the API version may causeVerifyError
. This was fixed in version 1.33.0, which can once again be used with any version of the public API. -
For relational databases, the agent now captures the database name and makes it part of service dependencies and service map. For example, with a
MySQL
database, previously a singlemysql
item was shown in the map and in service dependencies, the agent will now include the database name in the dependency, thusmysql/my-db1
,mysql/my-db2
will now be captured.
-
Promote mature agent features as Generaly Available (GA) - {pull}2632[#2632]
-
OpenTelemetry bridge is now enabled by default
-
Circuit breaker marked as GA (still disabled by default)
-
API Attach and CLI Attach marked as GA
-
use_path_as_transaction_name
configuration option marked as GA -
Dubbo instrumentation is now enabled by default
-
com.sun.net.httpserver.HttpServer
instrumentation marked as GA
-
-
Struts action invocations via an action chain result start a new span - {pull}2513[#2513]
-
Added official support for Elasticsearch Java API client - {pull}2211[#2211]
-
Added the ability to make spans non-discardable through the public API and the OpenTelemetry bridge - {pull}2632[#2632]
-
Added support for the new service target fields - {pull}2578[#2578]
-
Capture the database name from JDBC connection string - {pull}2642[#2642]
-
Added an additional span around Javalin template renderers - {pull}2381[#2381]
-
Added support for downloading the latest agent version through the attach CLI by setting
--download-agent-version latest
. In addition, when using theapm-agent-attach-cli-slim.jar
, which does not contain a bundled agent, the latest version will be downloaded from maven at runtime unless configured otherwise through--download-agent-version
- {pull}2659[#2659] -
Added span-links to messaging systems instrumentation (supported by APM Server 8.3+ only) - {pull}2610[#2610]
-
Fix missing attributes in bridged OTel transactions - {pull}2657[#2657]
-
Fix
transaction.result
with bridged OTel transactions - {pull}2660[#2660]
-
Starting this version, when using [config-log-ecs-reformatting], the agent will automatically set the
service.version
field. If you are using ECS-logging and set theservice.version
through a custom field, the behaviour is not strictly defined. Remove the customservice.name
field setting and either allow the agent to automatically discover and set it, or use the [config-service-version] config option to set it manually.
-
Vert.x 3.x instrumentation was refactored to remove constructor instrumentation as well as wrapping of response handler. In addition, in HTTP 2 request handling, transactions are ended when the request end event occurs and are kept alive until response end, when they are allowed to recycle. This allows for spans representing asynchronous handling of requests for which the corresponding transaction has ended - {pull}2564[#2564]
-
Jedis clients instrumentation was changed
-
Set the service version when using the ECS reformatting of the application logs: {pull}2603[#2603]
-
Add ECS-reformatting support for
java.util.logging
- {pull}2591[#2591] -
Added support for setting the service version on Log4j2’s EcsLayout - {pull}2604[#2604]
-
Added support for AWS S3 and DynamoDB - {pull}2606[#2606]
-
Added support for Jedis 4.x clients - {pull}2626[#2626]
-
Fixed multiple dropped stats types in a transaction producing invalid JSON: {pull}2589[#2589]
-
Fixed NoClassDefFoundError when using OTel bridge and span.*current() : {pull}2596[#2596]
-
Fallback to standard output when Security Manager prevents writing to log file - {pull}2581[#2581]
-
Fix missing transactions when using Vert.x 3.x with HTTP 1 - {pull}2564[#2564]
-
Fix Vert.x
GET null
transactions to be namedGET unknown route
, according to spec - {pull}2564[#2564] -
Fix OpenTelemetry bridge span end with explicit timestamp - {pull}2615[#2615]
-
Fix improper naming for
scheduled
transactions created byjava.util.TimerTask
instrumentation - {pull}2620[#2620] -
Properly handle
java.lang.IllegalStateException
related to premature invocation ofServletConfig#getServletContext()
inServlet#init()
instrumentations - {pull}2627[#2627]
-
Fixed AWS Lambda instrumentation for AWS handler classes with input object types that are not AWS Events classes - {pull}2551[#2551]
-
Fixed service name discovery based on MANIFEST.MF file through
ServletContainerInitializer#onStartup
on Jakarta Servlet containers - {pull}2546[#2546] -
Fix shaded classloader package definition - {pull}2566[#2566]
-
Fix logging initialization with Security Manager - {pull}2568[#2568]
-
normalize empty
transaction.type
andspan.type
- {pull}2525[#2525] -
Allowing square brackets within the [config-capture-jmx-metrics] config value - {pull}2547[#2547]
-
Fixed duplicated ending of
HttpUrlConnection
spans - {pull}2530[#2530] -
Compressed span fixes - {pull}2576[#2576], {pull}2552[#2552], {pull}2558[#2558]
-
Create the JDBC spans as exit spans- {pull}2484[#2484]
-
WebSocket requests are now captured with transaction type request instead of custom - {pull}2501[#2501]
-
Logging frameworks instrumentations - {pull}2428[#2428]. This refactoring includes:
-
Log correlation now works based on bytecode instrumentation rather than
ActivationListener
that directly updates the MDC -
Merging the different instrumentations (log-correlation, error-capturing and ECS-reformatting) into a single plugin
-
Module structure and package naming changes
-
-
Added support for setting service name and version for a transaction via the public api - {pull}2451[#2451]
-
Added support for en-/disabling each public annotation on each own - {pull}2472[#2472]
-
Added support for compressing spans - {pull}2477[#2477]
-
Added microsecond durations with
us
as unit - {pull}2496[#2496] -
Added support for dropping fast exit spans - {pull}2491[#2491]
-
Added support for collecting statistics about dropped exit spans - {pull}2505[#2505]
-
Making AWS Lambda instrumentation GA - includes some changes in Lambda transaction metadata fields and a dedicated flush HTTP request to the AWS Lambda extension - {pull}2424[#2424]
-
Changed logging correlation to be on by default. This change includes the removal of the now redundant
enable_log_correlation
config option. If there’s a need to disable the log correlation mechanism, this can be done now through thedisable_instrumentations
config - {pull}2428[#2428] -
Added automatic error event capturing for log4j1 and JBoss LogManager - {pull}2428[#2428]
-
Issue a warning when security manager is mis-configured - {pull}2510[#2510]
-
Add experimental OpenTelemetry API bridge - {pull}1631[#1631]
-
Proxy classes are excluded from instrumentation in more cases - {pull}2474[#2474]
-
Only time type/method matching if the debug logging is enabled as the results are only used when debug logging is enabled - {pull}2471[#2471]
-
Fix cross-plugin dependencies triggering NoClassDefFound - {pull}2509[#2509]
-
Fix status code setting in AWS Lambda transactions triggered by API Gateway V1 - {pull}2346[#2346]
-
Fix classloading OSGi bundles with partial dependency on Servlet API + avoid SecurityException with Apache Sling - {pull}2418[2418]
-
Respect
transaction_ignore_urls
andtransaction_ignore_user_agents
when creating transactions in the spring webflux instrumentation - {pull}2515[#2515]
-
Changes in service name auto-discovery of jar files (see Features section)
-
Exceptions that are logged using the fatal log level are now captured (log4j2 only) - {pull}2377[#2377]
-
Replaced
authorization
in the default value ofsanitize_field_names
withauth
- {pull}2326[#2326] -
Unsampled transactions are dropped and not sent to the APM-Server if the APM-Server version is 8.0+ - {pull}2329[#2329]
-
Adding agent logging capabilities to our SDK, making it available for external plugins - {pull}2390[#2390]
-
Service name auto-discovery improvements
-
For applications deployed to application servers (
war
files) and standalone jars that are started withjava -jar
, the agent now discovers theMETA-INF/MANIFEST.MF
file. -
If the manifest contains the
Implementation-Title
attribute, it is used as the default service name - {pull}1921[#1921], {pull}2434[#2434]
Note: this may change your service names if you relied on the auto-discovery that uses the name of the jar file. If that jar file also contains anImplementation-Title
attribute in theMANIFEST.MF
file, the latter will take precedence. -
When the manifest contains the
Implementation-Version
attribute, it is used as the default service version - {pull}1726[#1726], {pull}1922[#1922], {pull}2434[#2434]
-
-
Added support for instrumenting Struts 2 static resource requests - {pull}1949[#1949]
-
Added support for Java/Jakarta WebSocket ServerEndpoint - {pull}2281[#2281]
-
Added support for setting the service name on Log4j2’s EcsLayout - {pull}2296[#2296]
-
Print the used instrumentation groups when the application stops - {pull}2448[#2448]
-
Add
elastic.apm.start_async
property that makes the agent start on a non-premain/main thread - {pull}2454[#2454]
-
Fix runtime attach with some docker images - {pull}2385[#2385]
-
Restore dynamic capability to
log_level
config for plugin loggers - {pull}2384[#2384] -
Fix slf4j-related
LinkageError
- {pull}2390[#2390] and {pull}2376[#2376] -
Fix possible deadlock occurring when Byte Buddy reads System properties by warming up bytecode instrumentation code paths. The BCI warmup is on by default and may be disabled through the internal
warmup_byte_buddy
config option - {pull}2368[#2368] -
Fixed few dubbo plugin issues - {pull}2149[#2149]
-
Dubbo transaction will should be created at the provider side
-
APM headers conversion issue within dubbo transaction
-
-
Fix External plugins automatic setting of span outcome - {pull}2376[#2376]
-
Avoid early initialization of JMX on Weblogic - {pull}2420[#2420]
-
Automatically disable class sharing on AWS lambda layer - {pull}2438[#2438]
-
Avoid standalone spring applications to have two different service names, one based on the jar name, the other based on
spring.application.name
.
-
Fix
@Traced
annotation to return proper outcome instead offailed
- {pull}2370[#2370]
-
Update Log4j to 2.12.4 and log4j2-ecs-layout to 1.3.2 - {pull}2378[#2378]
-
Update Log4j to 2.12.3
-
Update ecs-logging-java to 1.3.0
-
If the agent cannot discover a service name, it now uses
unknown-java-service
instead ofmy-service
- {pull}2325[#2325]
-
Gracefully handle JDBC drivers which don’t support
Connection#getCatalog
- {pull}2340[#2340] -
Fix using JVM keystore options for communication with APM Server - {pull}2362[#2362]
-
Update Log4j to 2.12.2
-
Fix module loading errors on J9 JVM - {pull}2341[#2341]
-
Fixing log4j configuration error - {pull}2343[#2343]
-
Fix for "Log4Shell" RCE 0-day exploit in log4j CVE-2021-44228 - {pull}2332[#2332]
-
Added support to selectively enable instrumentations - {pull}2292[#2292]
-
Preferring controller names for Spring MVC transactions,
use_path_as_transaction_name
only as a fallback - {pull}2320[#2320]
-
Adding experimental support for AWS Lambda - {pull}1951[#1951]
-
Now supporting tomcat 10 - {pull}2229[#2229]
-
Fix error with parsing APM Server version for 7.16+ - {pull}2313[#2313]
-
Resolves Local Privilege Escalation issue ESA-2021-30 CVE-2021-37942
-
Add support to Jakarta EE for JSF - {pull}2254[#2254]
-
Fixing missing Micrometer metrics in Spring boot due to premature initialization - {pull}2255[#2255]
-
Fixing hostname trimming of FQDN too aggressive - {pull}2286[#2286]
-
Fixing agent
unknown
version - {pull}2289[#2289] -
Improve runtime attach configuration reliability - {pull}2283[#2283]
-
Resolves Local Privilege Escalation issue ESA-2021-29 CVE-2021-37941
-
transaction_ignore_urls
now relies on full request URL path - {pull}2146[#2146]-
On a typical application server like Tomcat, deploying an
app.war
application to the non-ROOT context makes it accessible withhttp://localhost:8080/app/
-
Ignoring the whole webapp through
/app/*
was not possible until now. -
Existing configuration may need to be updated to include the deployment context, thus for example
/static/.js
used to exclude known static files in all applications might be changed to/app/static/
.js
or/static/.js
. -
It only impacts prefix patterns due to the additional context path in pattern.
-
It does not impact deployment within the
ROOT
context like Spring-boot which do not have such context path prefix.
-
-
The metrics
transaction.duration.sum.us
,transaction.duration.count
andtransaciton.breakdown.count
are no longer recorded - {pull}2194[#2194] -
Automatic hostname discovery mechanism had changed, so the resulted
host.name
andhost.hostname
in events reported by the agent may be different. This was done in order to improve the integration with host metrics in the APM UI.
-
Improved capturing of logged exceptions when using Log4j2 - {pull}2139[#2139]
-
Update to async-profiler 1.8.7 and set configured
safemode
at load time though a new system property - {pull}2165[#2165] -
Added support to capture
context.message.routing-key
in rabbitmq, spring amqp instrumentations - {pull}1767[#1767] -
Breakdown metrics are now tracked per service (when using APM Server 8.0) - {pull}2208[#2208]
-
Add support for Spring AMQP batch API - {pull}1716[#1716]
-
Add the (current) transaction name to the error (when using APM Server 8.0) - {pull}2235[#2235]
-
The JVM/JMX metrics are reported for each service name individually (when using APM Server 8.0) - {pull}2233[#2233]
-
Added
span_stack_trace_min_duration
option. This replaces the now deprecatedspan_frames_min_duration
option. The difference is that the new option has more intuitive semantics for negative values (never collect stack trace) and zero (always collect stack trace). - {pull}2220[#2220] -
Add support to Jakarta EE for JAX-WS - {pull}2247[#2247]
-
Add support to Jakarta EE for JAX-RS - {pull}2248[#2248]
-
Add support for Jakarta EE EJB annotations
@Schedule
,@Schedules
- {pull}2250[#2250] -
Add support to Jakarta EE for Servlets - {pull}1912[#1912]
-
Added support to Quartz 1.x - {pull}2219[#2219]
-
Disable compression when sending data to a local APM Server
-
Reducing startup contention related to instrumentation through
ensureInstrumented
- {pull}2150[#2150]
-
Fix k8s metadata discovery for containerd-cri envs - {pull}2126[#2126]
-
Fixing/reducing startup delays related to
ensureInstrumented
- {pull}2150[#2150] -
Fix runtime attach when bytebuddy is in application classpath - {pull}2116[#2116]
-
Fix failed integration between agent traces and host metrics coming from Beats/Elastic-Agent due to incorrect hostname discovery - {pull}2205[#2205]
-
Fix infinitely kept-alive transactions in Hikari connection pool - {pull}2210[#2210]
-
Fix few Webflux exceptions and missing reactor module - {pull}2207[#2207]
-
Loading the agent from an isolated class loader - {pull}2109[#2109]
-
Refactorings in the
apm-agent-plugin-sdk
that may imply breaking changes for beta users of the external plugin mechanism-
WeakMapSupplier.createMap()
is nowWeakConcurrent.buildMap()
and contains more builders - {pull}2136[#2136] -
GlobalThreadLocal
has been removed in favor ofDetachedThreadLocal
. To make it global, useGlobalVariables
- {pull}2136[#2136] -
DynamicTransformer.Accessor.get().ensureInstrumented
is nowDynamicTransformer.ensureInstrumented
- {pull}2164[#2164] -
The
@AssignTo.
annotations have been removed. Use the@Advice.AssignReturned.
annotations that come with the latest version of Byte Buddy. If your plugin uses the old annotations, it will be skipped. {pull}2171[#2171]
-
-
Switching last instrumentations (
trace_methods
, sparkjava, JDKHttpServer
and Struts 2) toTracerAwareInstrumentation
- {pull}2170[#2170] -
Replace concurrency plugin maps to
SpanConcurrentHashMap
ones - {pull}2173[#2173] -
Align User-Agent HTTP header with other APM agents - {pull}2177[#2177]
-
Update Log4j to 2.12.4 and log4j2-ecs-layout to 1.3.2 - {pull}2378[#2378]
-
If you rely on Database span subtype and use Microsoft SQL Server, the span subtype has been changed from
sqlserver
tomssql
to align with other agents.
-
Stop collecting the field
http.request.socket.encrypted
in http requests - {pull}2136[#2136]
-
Improved naming for Spring controllers - {pull}1906[#1906]
-
ECS log reformatting improvements - {pull}1910[#1910]
-
Automatically sets
service.node.name
in all log events if set through agent configuration -
Add
log_ecs_reformatting_additional_fields
option to support arbitrary fields in logs -
Automatically serialize markers as tags where relevant (log4j2 and logback)
-
-
gRPC spans (client and server) can detect errors or cancellation through custom listeners - {pull}2067[#2067]
-
Add
-download-agent-version
to the agent attach CLI tool options, allowing the user to configure an arbitrary agent version that will be downloaded from maven and attached - {pull}1959[#1959] -
Add extra check to detect improper agent setup - {pull}2076[#2076]
-
In redis tests - embedded RedisServer is replaced by testcontainers - {pull}2221[#2221]
-
Reduce GC time overhead caused by WeakReferences - {pull}2086[#2086], {pull}2081[#2081]
-
Reduced memory overhead by a smarter type pool caching strategy - {pull}2102[#2102].
The type pool cache improves the startup times by speeding up type matching (determining whether a class that’s about to be loaded should be instrumented). Generally, the more types that are cached, the faster the startup.
The old strategy did not impose a limit to the cache but cleared it after it hasn’t been accessed in a while. However, load test have discovered that the cache may never be cleared and leave a permanent overhead of 23mb. The actual size of the cache highly depends on the application and loosely correlates with the number of loaded classes.
The new caching strategy targets to allocate 1% of the committed heap, at least 0.5mb and max 10mb. If a particular entry hasn’t been accessed within 20s, it will be removed from the cache.
The results based on load testing are very positive:-
Equivalent startup times (within the margins of error of the previous strategy)
-
Equivalent allocation rate (within the margins of error of the previous strategy)
-
Reduced avg heap utilization from 10%/15mb (previous strategy) to within margins of error without the agent
-
Reduced GC time due to the additional headroom that the application can utilize.
-
Based on heap dump analysis, after warmup, the cache size is now around 59kb (down from 23mb with the previous strategy).
-
-
Fix failure to parse some forms of the
Implementation-Version
property from jar manifest files - {pull}1931[#1931] -
Ensure single value for context-propagation header - {pull}1937[#1937]
-
Fix gRPC non-terminated (therefore non-reported) client spans - {pull}2067[#2067]
-
Fix Webflux response status code - {pull}1948[#1948]
-
Ensure path filtering is applied when Servlet path is not available - {pull}2099[#2099]
-
Align span subtype for MS SqlServer - {pull}2112[#2112]
-
Fix potential destination host name corruption in OkHttp client spans - {pull}2118[#2118]
-
Migrate several plugins to indy dispatcher {pull}2087[#2087], {pull}2088[#2088], {pull}2090[#2090], {pull}2094[#2094], {pull}2095[#2095]
-
If you rely on instrumentations that are in the
experimental
group, you must now setenable_experimental_instrumentations=true
otherwise the experimental instrumentations will be disabled by default. Up to version1.24.0
using an empty value fordisable_instrumentations
was the recommended way to override the defaultdisable_instrumentations=experimental
.
-
Support for inheritance of public API annotations - {pull}1805[#1805]
-
JDBC instrumentation sets
context.db.instance
- {pull}1820[#1820] -
Add support for Vert.x web client- {pull}1824[#1824]
-
Avoid recycling of spans and transactions that are using through the public API, so to avoid reference-counting-related errors - {pull}1859[#1859]
-
Add [config-enable-experimental-instrumentations] configuration option to enable experimental features - {pull}1863[#1863]
-
Previously, when adding an instrumentation group to
disable_instrumentations
, we had to make sure to not forget the defaultexperimental
value, for example when disablingjdbc
instrumentation we had to setdisable_instrumentations=experimental,jdbc
otherwise settingdisable_instrumentations=jdbc
would disable jdbc and also enable experimental features, which would not be the desired effect. -
Previously, by default
disable_instrumentations
containedexperimental
-
Now by default
disable_instrumentations
is empty andenable_experimental_instrumentations=false
-
Set
enable_experimental_instrumentations=true
to enable experimental instrumentations
-
-
Eliminating concerns related to log4j2 vulnerability - https://nvd.nist.gov/vuln/detail/CVE-2020-9488#vulnCurrentDescriptionTitle. We cannot upgrade to version above 2.12.1 because this is the last version of log4j that is compatible with Java 7. Instead, we exclude the SMTP appender (which is the vulnerable one) from our artifacts. Note that older versions of our agent are not vulnerable as well, as the SMTP appender was never used, this is only to further reduce our users' concerns.
-
Adding public APIs for setting
destination.service.resource
,destination.address
anddestination.port
fields for exit spans - {pull}1788[#1788] -
Only use emulated runtime attachment as fallback, remove the
--without-emulated-attach
option - {pull}1865[#1865] -
Instrument
javax.servlet.Filter
the same way asjavax.servlet.FilterChain
- {pull}1858[#1858] -
Propagate trace context headers in HTTP calls occurring from within traced exit points, for example - when using Elasticsearch’s REST client - {pull}1883[#1883]
-
Added support for naming sparkjava (not Apache Spark) transactions {pull}1894[#1894]
-
Added the ability to manually create exit spans, which will result with the auto creation of service nodes in the service map and downstream service in the dependencies table - {pull}1898[#1898]
-
Basic support for
com.sun.net.httpserver.HttpServer
- {pull}1854[#1854] -
Update to async-profiler 1.8.6 {pull}1907[#1907]
-
Added support for setting the framework using the public api (#1908) - {pull}1909[#1909]
-
Fix NPE with
null
binary header values + properly serialize them - {pull}1842[#1842] -
Fix
ListenerExecutionFailedException
when using Spring AMQP’s ReplyTo container - {pull}1872[#1872] -
Enabling log ECS reformatting when using Logback configured with
LayoutWrappingEncoder
and a pattern layout - {pull}1879[#1879] -
Fix NPE with Webflux + context propagation headers - {pull}1871[#1871]
-
Fix
ClassCastException
withConnnectionMetaData
and multiple classloaders - {pull}1864[#1864] -
Fix NPE in
co.elastic.apm.agent.servlet.helper.ServletTransactionCreationHelper.getClassloader
- {pull}1861[#1861] -
Fix for Jboss JMX unexpected notifications - {pull}1895[#1895]
-
Basic support for Apache Struts 2 {pull}1763[#1763]
-
Extending the [config-log-ecs-reformatting] config option to enable the overriding of logs with ECS-reformatted events. With the new
OVERRIDE
option, non-file logs can be ECS-reformatted automatically as well - {pull}1793[#1793] -
Instrumentation for Vert.x Web {pull}1697[#1697]
-
Changed log level of vm arguments to debug
-
Giving precedence for the W3C
tracecontext
header over theelastic-apm-traceparent
header - {pull}1821[#1821] -
Add instrumentation for Webflux - {pull}1305[#1305]
-
Add instrumentation for Javalin {pull}1822[#1822]
-
Fix another error related to instrumentation plugins loading on Windows - {pull}1785[#1785]
-
Load Spring AMQP plugin- {pull}1784[#1784]
-
Avoid
IllegalStateException
when multipletracestate
headers are used - {pull}1808[#1808] -
Ensure CLI attach avoids
sudo
only when required and avoid blocking - {pull}1819[#1819] -
Avoid sending metric-sets without samples, so to adhere to the intake API - {pull}1826[#1826]
-
Fixing our type-pool cache, so that it can’t cause OOM (softly-referenced), and it gets cleared when not used for a while - {pull}1828[#1828]
-
Remove single-package limitation for embedded plugins - {pull}1780[#1780]
-
There are breaking changes in the attacher cli. See the Features section for more information.
-
Overhaul of the attacher cli application that allows to attach the agent to running JVMs - {pull}1667[#1667]
-
The artifact of the standalone cli application is now called
apm-agent-attach-cli
. The attacher API is still calledapm-agent-attach
. -
There is also a slim version of the cli application that does not bundle the Java agent. It requires the
--agent-jar
option to be set. -
Improved logging
The application uses {ecs-logging-java-ref}/intro.html[Java ECS logging] to emit JSON logs. The log level can be configured with the--log-level
option. By default, the program is logging to the console but using the--log-file
option, it can also log to a file. -
Attach to JVMs running under a different user (unix only)
The JVM requires the attacher to be running under the same user as the target VM (the attachee). Theapm-agent-attach-standalone.jar
can now be run with a user that has permissions to switch to the user that runs the target VM. On Windows, the attacher can still only attach to JVMs that are running with under the same user. -
New include/exclude discovery rules
-
--include-all
: Attach to all discovered JVMs. If no matchers are provided, it will not attach to any JVMs. -
--include-user
/--exclude-user
: Attach to all JVMs of a given operating system user. -
--include-main
/--exclude-main
: Attach to all JVMs that whose main class/jar name, or system properties match the provided regex. -
--include-vmargs
/--exclude-vmargs
: Attach to all JVMs that whose main class/jar name, or system properties match the provided regex.
-
-
Removal of options
-
The deprecated
--arg
option has been removed. -
The
-i
/--include
,-e
/exclude
options have been removed in favor of the--<include|exclude>-<main|vmargs>
options. -
The
-p
/--pid
options have been removed in favor of the--include-pid
option.
-
-
Changed behavior of the
-l
/--list
option
The option now only lists JVMs that match the include/exclude discovery rules. Thus, it can be used to do a dry-run of the matchers without actually performing an attachment. It even works in combination with--continuous
now. By default, the VM arguments are not printed, but only when the-a
/--list-vmargs
option is set. -
Remove dependency on
jps
Even when matching on the main class name or on system properties, -
Checks the Java version before attaching to avoid attachment on unsupported JVMs.
-
-
Cassandra instrumentation - {pull}1712[#1712]
-
Log correlation supports JBoss Logging - {pull}1737[#1737]
-
Update Byte-buddy to
1.11.0
- {pull}1769[#1769] -
Support for user.domain {pull}1756[#1756]
-
JAX-RS supports javax.ws.rs.PATCH
-
Enabling build and unit tests on Windows - {pull}1671[#1671]
-
Fixed log correlation for log4j2 - {pull}1720[#1720]
-
Fix apm-log4j1-plugin and apm-log4j2-plugin dependency on slf4j - {pull}1723[#1723]
-
Avoid systematic
MessageNotWriteableException
error logging, now only visible indebug
- {pull}1715[#1715] and {pull}1730[#1730] -
Fix rounded number format for non-english locales - {pull}1728[#1728]
-
Fix
NullPointerException
on legacy Apache client instrumentation when host isnull
- {pull}1746[#1746] -
Apply consistent proxy class exclusion heuristic - {pull}1738[#1738]
-
Fix micrometer serialization error - {pull}1741[#1741]
-
Optimize & avoid
ensureInstrumented
deadlock by skipping stack-frame computation for Java7+ bytecode - {pull}1758[#1758] -
Fix instrumentation plugins loading on Windows - {pull}1671[#1671]
-
Migrate some plugins to indy dispatcher {pull}1369[#1369] {pull}1410[#1410] {pull}1374[#1374]
-
Dots in metric names of Micrometer metrics get replaced with underscores to avoid mapping conflicts. De-dotting be disabled via
dedot_custom_metrics
. - {pull}1700[#1700]
-
Introducing a new mechanism to ease the development of community instrumentation plugins. See [config-plugins-dir] for more details. This configuration was already added in 1.18.0, but more extensive and continuous integration testing allows us to expose it now. It is still marked as "experimental" though, meaning that future changes in the mechanism may break early contributed plugins. However, we highly encourage our community to try it out and we will do our best to assist with such efforts.
-
Deprecating
ignore_user_agents
in favour oftransaction_ignore_user_agents
, maintaining the same functionality - {pull}1644[#1644] -
Update existing Hibernate Search 6 instrumentation to the final relase
-
The
use_path_as_transaction_name
option is now dynamic -
Flushing internal and micrometer metrics before the agent shuts down - {pull}1658[#1658]
-
Support for OkHttp 4.4+ - {pull}1672[#1672]
-
Adding capability to automatically create ECS-JSON-formatted version of the original application log files, through the [config-log-ecs-reformatting] config option. This allows effortless ingestion of logs to Elasticsearch without any further configuration. Supports log4j1, log4j2 and Logback. {pull}1261[#1261]
-
Add support to Spring AMQP - {pull}1657[#1657]
-
Adds the ability to automatically configure usage of the OpenTracing bridge in systems using ServiceLoader - {pull}1708[#1708]
-
Update to async-profiler 1.8.5 - includes a fix to a Java 7 crash and enhanced safe mode to better deal with corrupted stack frames.
-
Add a warning on startup when
-Xverify:none
or-noverify
flags are set as this can lead to crashes that are very difficult to debug - {pull}1593[#1593]. In an upcoming version, the agent will not start when these flags are set, unless the system propertyelastic.apm.disable_bootstrap_checks
is set to true.
-
fix sample rate rounded to zero when lower than precision - {pull}1655[#1655]
-
fixed a couple of bugs with the external plugin mechanism (not documented until now) - {pull}1660[#1660]
-
Fix runtime attach conflict with multiple users - {pull}1704[#1704]
-
Following PR {pull}1650[#1650], there are two slight changes with the [config-server-url] and [config-server-urls] configuration options:
-
So far, setting
server_urls
with an empty string would allow the agent to work normally, apart from any action that requires communication with the APM Server, including the attempt to fetch a central configuration. Starting in this agent version, settingserver_urls
to empty string doesn’t have any special meaning, it is the default expected configuration, whereserver_url
will be used instead. In order to achieve the same behaviour, use the new [config-disable-send] configuration. -
Up to this version,
server_url
was used as an alias toserver_urls
, meaning that one could potentially set theserver_url
config with a comma-separated list of multiple APM Server addresses, and that would have been a valid configuration. Starting in this agent version,server_url
is a separate configuration, and it only accepts Strings that represent a single valid URL. Specifically, empty strings and commas are invalid.
-
-
Add cloud provider metadata to reported events, see spec for details. By default, the agent will try to automatically detect the cloud provider on startup, but this can be configured through the
cloud_provider
config option - {pull}1599[#1599] -
Add span & transaction
outcome
field to improve error rate calculations - {pull}1613[#1613]
-
Fixing crashes observed in Java 7 at sporadic timing by applying a few seconds delay on bootstrap - {pull}1594[#1594]
-
Fallback to using "TLS"
SSLContext
when "SSL" is not available - {pull}1633[#1633] -
Fixing agent startup failure with
NullPointerException
thrown by Byte-buddy’sMultipleParentClassLoader
- {pull}1647[#1647] -
Fix cached type resolution triggering
ClassCastException
- {pull}1649[#1649]
-
The following public API types were
public
so far and became package-private:NoopScope
,ScopeImpl
andAbstractSpanImpl
. If your code is using them, you will need to change that when upgrading to this version. Related PR: {pull}1532[#1532]
-
Add support for RabbitMQ clients - {pull}1328[#1328]
-
Fix small memory allocation regression introduced with tracestate header {pull}1508[#1508]
-
Fix
NullPointerException
fromWeakConcurrentMap.put
through the Elasticsearch client instrumentation - {pull}1531[#1531] -
Sending
transaction_id
andparent_id
only for events that contain a validtrace_id
as well - {pull}1537[#1537] -
Fix
ClassNotFoundError
with old versions of Spring resttemplate {pull}1524[#1524] -
Fix Micrometer-driven metrics validation errors by the APM Server when sending with illegal values - {pull}1559[#1559]
-
Serialize all stack trace frames when setting
stack_trace_limit=-1
instead of none - {pull}1571[#1571] -
Fix
UnsupportedOperationException
when callingServletContext.getClassLoader()
- {pull}1576[#1576] -
Fix improper request body capturing - {pull}1579[#1579]
-
Avoid
NullPointerException
due to null return values instrumentation advices - {pull}1601[#1601] -
Update async-profiler to 1.8.3 {pull}1602[1602]
-
Use null-safe data structures to avoid
NullPointerException
{pull}1597[1597] -
Fix memory leak in sampling profiler mechanism - {pull}1592[#1592]
-
Migrate some plugins to indy dispatcher {pull}1405[#1405] {pull}1394[#1394]
-
The agent version now includes a git hash if it’s a snapshot version. This makes it easier to differ distinct snapshot builds of the same version. Example:
1.18.1-SNAPSHOT.4655910
-
Add support for sampling weight with propagation in
tracestate
W3C header {pull}1384[#1384] -
Adding two more valid options to the
log_level
config:WARNING
(equivalent toWARN
) andCRITICAL
(will be treated asERROR
) - {pull}1431[1431] -
Add the ability to disable Servlet-related spans for
INCLUDE
,FORWARD
andERROR
dispatches (without affecting basic Servlet capturing) by addingservlet-api-dispatch
to [config-disable-instrumentations] - {pull}1448[1448] -
Add Sampling Profiler support for AArch64 architectures - {pull}1443[1443]
-
Support proper transaction naming when using Spring’s
ServletWrappingController
- {pull}1461[#1461] -
Update async-profiler to 1.8.2 {pull}1471[1471]
-
Update existing Hibernate Search 6 instrumentation to work with the latest CR1 release
-
Deprecating the
addLabel
public API in favor ofsetLabel
(still supportingaddLabel
) - {pull}1449[#1449]
-
Fix
HttpUrlConnection
instrumentation issue (affecting distributed tracing as well) when using HTTPS without usingjava.net.HttpURLConnection#disconnect
- {pull}1447[1447] -
Fixes class loading issue that can occur when deploying multiple applications to the same application server - {pull}1458[#1458]
-
Fix ability to disable agent on startup wasn’t working for runtime attach {pull}1444[1444]
-
Avoid
UnsupportedOperationException
on some spring application startup {pull}1464[1464] -
Fix ignored runtime attach
config_file
{pull}1469[1469] -
Fix
IllegalAccessError: Module 'java.base' no access to: package 'java.lang'…
in J9 VMs of Java version >= 9 - {pull}1468[#1468] -
Fix JVM version parsing on HP-UX {pull}1477[#1477]
-
Fix Spring-JMS transactions lifecycle management when using multiple concurrent consumers - {pull}1496[#1496]
-
Migrate some plugins to indy dispatcher {pull}1404[1404] {pull}1411[1411]
-
Replace System Rules with System Lambda {pull}1434[#1434]
-
Migrate some plugins to indy dispatcher {pull}1362[1362] {pull}1366[1366] {pull}1363[1363] {pull}1383[1383] {pull}1368[1368] {pull}1364[1364] {pull}1365[1365] {pull}1367[1367] {pull}1371[1371]
-
Fix instrumentation error for HttpClient - {pull}1402[#1402]
-
Eliminate
unsupported class version error
messages related to loading the Java 11 HttpClient plugin in pre-Java-11 JVMs {pull}1397[1397] -
Fix rejected metric events by APM Server with response code 400 due to data validation error - sanitizing Micrometer metricset tag keys - {pull}1413[1413]
-
Fix invalid micrometer metrics with non-numeric values {pull}1419[1419]
-
Fix
NoClassDefFoundError
with JDBC instrumentation plugin {pull}1409[1409] -
Apply
disable_metrics
config to Micrometer metrics - {pull}1421[1421] -
Remove cgroup
inactive_file.bytes
metric according to spec {pull}1422[1422]
-
Deprecating
ignore_urls
config in favour oftransaction_ignore_urls
to align with other agents, while still allowing the old config name for backward compatibility - {pull}1315[#1315] -
Enabling instrumentation of classes compiled with Java 1.4. This is reverting the restriction of instrumenting only bytecode of Java 1.5 or higher ({pull}320[#320]), which was added due to potential
VerifyError
. Such errors should be avoided now by the usage ofTypeConstantAdjustment
- {pull}1317[#1317] -
Enabling agent to work without attempting any communication with APM server, by allowing setting
server_urls
with an empty string - {pull}1295[#1295] -
Add micrometer support - {pull}1303[#1303]
-
Add
profiling_inferred_spans_lib_directory
option to override the default temp directory used for exporting the async-profiler library. This is useful for server-hardened environments where/tmp
is often configured withnoexec
, leading tojava.lang.UnsatisfiedLinkError
errors - {pull}1350[#1350] -
Create spans for Servlet dispatches to FORWARD, INCLUDE and ERROR - {pull}1212[#1212]
-
Support JDK 11 HTTPClient - {pull}1307[#1307]
-
Lazily create profiler temporary files {pull}1360[#1360]
-
Convert the followings to Indy Plugins (see details in 1.18.0-rc1 relase notes): gRPC, AsyncHttpClient, Apache HttpClient
-
The agent now collects cgroup memory metrics (see details in Metrics page)
-
Update async-profiler to 1.8.1 {pull}1382[#1382]
-
Runtime attach install option is promoted to 'beta' status (was experimental).
-
Fixes a
NoClassDefFoundError
in the JMS instrumentation ofMessageListener
- {pull}1287[#1287] -
Fix
/ by zero
error message when settingserver_urls
with an empty string - {pull}1295[#1295] -
Fix
ClassNotFoundException
orClassCastException
in some cases where special log4j configurations are used - {pull}1322[#1322] -
Fix
NumberFormatException
when using early access Java version - {pull}1325[#1325] -
Fix
service_name
config being ignored when set to the same auto-discovered default value - {pull}1324[#1324] -
Fix service name error when updating a web app on a Servlet container - {pull}1326[#1326]
-
Fix remote attach 'jps' executable not found when 'java' binary is symlinked ot a JRE - {pull}1352[#1352]
This release candidate adds some highly anticipated features: It’s now possible to attach the agent at runtime in more cases than before. Most notably, it enables runtime attachment on JBoss, WildFly, Glassfish/Payara, and other OSGi runtimes such as Atlassian Jira and Confluence.
To make this and other significant features, such as external plugins, possible,
we have implemented major changes to the architecture of the agent.
The agent now relies on the invokedynamic
bytecode instruction to make plugin development easier, safer, and more efficient.
As early versions of Java 7 and Java 8 have unreliable support for invokedynamic,
we now require a minimum update level of 60 for Java 7 (7u60+) in addition to the existing minimum update level of 40 for Java 8 (8u40+).
We’re looking for users who would like to try this out to give feedback.
If we see that the invokedynamic
-based approach (indy plugins) works well, we can continue and migrate the rest of the plugins.
After the migration has completed, we can move forward with external plugins and remove the experimental label from runtime attachment.
If all works like in our testing, you would not see NoClassDefFoundError
s anymore when, for example, trying to attach the agent at runtime to an OSGi container or a JBoss server.
Also, non-standard OSGi containers, such as Atlassian Jira and other technologies with restrictive class loading policies, such as MuleSoft ESB, will benefit from this change.
In the worst case, there might be JVM crashes due to invokedynamic
-related JVM bugs.
However, we already disable the agent when attached to JVM versions that are known to be problematic.
Another potentially problematic area is that we now dynamically raise the bytecode version of instrumented classes to be at least bytecode version 51 (Java 7).
This is needed in order to be able to use the invokedynamic
instruction.
This requires re-computation of stack map frames which makes instrumentation a bit slower.
We don’t anticipate notable slowdowns unless you extensively (over-)use trace_methods
.
-
Early Java 7 versions, prior to update 60, are not supported anymore. When trying to attach to a non-supported version, the agent will disable itself and not apply any instrumentations.
-
Experimental support for runtime attachment now also for OSGi containers, JBoss, and WildFly
-
New mitigation of OSGi bootdelegation errors (
NoClassDefFoundError
). You can remove anyorg.osgi.framework.bootdelegation
related configuration. This release also removes the configuration optionboot_delegation_packages
. -
Overhaul of the
ExecutorService
instrumentation that avoidsClassCastException
issues - {pull}1206[#1206] -
Support for
ForkJoinPool
andScheduledExecutorService
(see [supported-async-frameworks]) -
Support for
ExecutorService#invokeAny
andExecutorService#invokeAll
-
Added support for
java.util.TimerTask
- {pull}1235[#1235] -
Add capturing of request body in Elasticsearch queries:
_msearch
,_count
,_msearch/template
,_search/template
,_rollup_search
- {pull}1222[#1222] -
Add
enabled
flag -
Add experimental support for Scala Futures
-
The agent now collects heap memory pools metrics - {pull}1228[#1228]
-
Fixes error capturing for log4j2 loggers. Version 1.17.0 introduced a regression.
-
Fixes
NullPointerException
related to JAX-RS and Quartz instrumentation - {pull}1249[#1249] -
Expanding k8s pod ID discovery to some formerly non-supported environments
-
When
recording
is set tofalse
, the agent will not send captured errors anymore. -
Fixes NPE in Dubbo instrumentation that occurs when the application is acting both as a provider and as a consumer - {pull}1260[#1260]
-
Adding a delay by default what attaching the agent to Tomcat using the premain route to work around the JUL deadlock issue - {pull}1262[#1262]
-
Fixes missing
jboss.as:*
MBeans on JBoss - {pull}1257[#1257]
-
Log files are now rotated after they reach [config-log-file-size]. There will always be one history file
${log_file}.1
. -
Add [config-log-format-sout] and [config-log-format-file] with the options
PLAIN_TEXT
andJSON
. The latter uses ecs-logging-java to format the logs. -
Exposing [config-classes-excluded-from-instrumentation] config - {pull}1187[#1187]
-
Add support for naming transactions based on Grails controllers. Supports Grails 3+ - {pull}1171[#1171]
-
Add support for the Apache/Alibaba Dubbo RPC framework
-
Async Profiler version upgraded to 1.7.1, with a new debugging flag for the stack frame recovery mechanism - {pull}1173[#1173]
-
Fixes
IndexOutOfBoundsException
that can occur when profiler-inferred spans are enabled. This also makes the profiler more resilient by just removing the call tree related to the exception (which might be in an invalid state) as opposed to stopping the profiler when an exception occurs. -
Fix
NumberFormatException
when parsing Ingres/Actian JDBC connection strings - {pull}1198[#1198] -
Prevent agent from overriding JVM configured truststore when not using HTTPS for communication with APM server - {pull}1203[#1203]
-
Fix
java.lang.IllegalStateException
withjps
JVM when using continuous runtime attach - {pull}1205[1205] -
Fix agent trying to load log4j2 plugins from application - {pull}1214[1214]
-
Fix memory leak in gRPC instrumentation plugin - {pull}1196[1196]
-
Fix HTTPS connection failures when agent is configured to use HTTPS to communicate with APM server {pull}1209[1209]
-
The log correlation feature now adds
error.id
to the MDC. See [supported-logging-frameworks] for details. - {pull}1050[#1050] -
Deprecating the
incubating
tag in favour of theexperimental
tag. This is not a breaking change, so formerdisable_instrumentation
configuration containing theincubating
tag will still be respected - {pull}1123[#1123] -
Add a
--without-emulated-attach
option for runtime attachment to allow disabling this feature as a workaround. -
Add workaround for JDK bug JDK-8236039 with TLS 1.3 {pull}1149[#1149]
-
Add log level
OFF
to silence agent logging -
Adds
span_min_duration
option to exclude fast executing spans. When set together with one of the more specific thresholds -trace_methods_duration_threshold
orprofiling_inferred_spans_min_duration
, the higher threshold will determine which spans will be discarded. -
Automatically instrument quartz jobs from the quartz-jobs artifact {pull}1170[#1170]
-
Perform re-parenting of regular spans to be a child of profiler-inferred spans. Requires APM Server and Kibana 7.8.0. {pull}1117[#1117]
-
Upgrade Async Profiler version to 1.7.0
-
When Servlet-related Exceptions are handled through exception handlers that return a 200 status code, agent shouldn’t override with 500 - {pull}1103[#1103]
-
Exclude Quartz 1 from instrumentation to avoid
IncompatibleClassChangeError: Found class org.quartz.JobExecutionContext, but interface was expected
- {pull}1108[#1108] -
Fix breakdown metrics span sub-types {pull}1113[#1113]
-
Fix flaky gRPC server instrumentation {pull}1122[#1122]
-
Fix side effect of calling
Statement.getUpdateCount
more than once {pull}1139[#1139] -
Stop capturing JDBC affected rows count using
Statement.getUpdateCount
to prevent unreliable side-effects {pull}1147[#1147] -
Fix OpenTracing error tag handling (set transaction error result when tag value is
true
) {pull}1159[#1159] -
Due to a bug in the build we didn’t include the gRPC plugin in the build so far
-
java.lang.ClassNotFoundException: Unable to load class 'jdk.internal…'
is thrown when tracing specific versions of Atlassian systems {pull}1168[#1168] -
Make sure spans are kept active during
AsyncHandler
methods in theAsyncHttpClient
-
CPU and memory metrics are sometimes not reported properly when using IBM J9 {pull}1148[#1148]
-
NullPointerException
thrown by the agent on WebLogic {pull}1142[#1142]
-
Ordering of configuration sources has slightly changed, please review [configuration]:
-
elasticapm.properties
file now has higher priority over java system properties and environment variables,
This change allows to change dynamic options values at runtime by editing file, previously values set in java properties or environment variables could not be overridden, even if they were dynamic.
-
-
Renamed some configuration options related to the experimental profiler-inferred spans feature ({pull}1084[#1084]):
-
profiling_spans_enabled
→profiling_inferred_spans_enabled
-
profiling_sampling_interval
→profiling_inferred_spans_sampling_interval
-
profiling_spans_min_duration
→profiling_inferred_spans_min_duration
-
profiling_included_classes
→profiling_inferred_spans_included_classes
-
profiling_excluded_classes
→profiling_inferred_spans_excluded_classes
-
Removed
profiling_interval
andprofiling_duration
(both are fixed to 5s now)
-
-
Gracefully abort agent init when running on a known Java 8 buggy JVM {pull}1075[#1075].
-
Add support for Redis Redisson client
-
Makes [config-instrument], [config-trace-methods], and [config-disable-instrumentations] dynamic. Note that changing these values at runtime can slow down the application temporarily.
-
Do not instrument Servlet API before 3.0 {pull}1077[#1077]
-
Add support for API keys for apm backend authentication {pull}1083[#1083]
-
Add support for gRPC client & server instrumentation {pull}1019[#1019]
-
Deprecating
active
configuration option in favor ofrecording
. Settingactive
still works as it’s now an alias forrecording
.
-
When JAX-RS-annotated method delegates to another JAX-RS-annotated method, transaction name should include method A - {pull}1062[#1062]
-
Fixed bug that prevented an APM Error from being created when calling
org.slf4j.Logger#error
- {pull}1049[#1049] -
Wrong address in JDBC spans for Oracle, MySQL and MariaDB when multiple hosts are configured - {pull}1082[#1082]
-
Document and re-order configuration priorities {pull}1087[#1087]
-
Improve heuristic for
service_name
when not set through config {pull}1097[#1097]
-
Support for the official W3C
traceparent
andtracestate
headers.
The agent now accepts both theelastic-apm-traceparent
and the officialtraceparent
header. By default, it sends both headers on outgoing requests, unlessuse_elastic_traceparent_header
is set to false. -
Creating spans for slow methods with the help of the sampling profiler async-profiler. This is a low-overhead way of seeing which methods make your transactions slow and a replacement for the
trace_methods
configuration option. See [supported-java-methods] for more details -
Adding a Circuit Breaker to pause the agent when stress is detected on the system and resume when the stress is relieved. See [circuit-breaker] and {pull}1040[#1040] for more info.
-
Span#captureException
andTransaction#captureException
in public API return reported error id - {pull}1015[#1015]
-
java.lang.IllegalStateException: Cannot resolve type description for <com.another.commercial.apm.agent.Class> - {pull}1037[#1037]
-
properly handle
java.sql.SQLException
for unsupported JDBC features {pull}[#1035] #1025
-
Add support for Redis Lettuce client
-
Add
context.message.age.ms
field for JMS message receiving spans and transactions - {pull}970[#970] -
Instrument log4j2 Logger#error(String, Throwable) ({pull}919[#919]) Automatically captures exceptions when calling
logger.error("message", exception)
-
Add instrumentation for external process execution through
java.lang.Process
and Apachecommons-exec
- {pull}903[#903] -
Add
destination
fields to exit span contexts - {pull}976[#976] -
Removed
context.message.topic.name
field - {pull}993[#993] -
Add support for Kafka clients - {pull}981[#981]
-
Add support for binary
traceparent
header format (see the spec for more details) - {pull}1009[#1009] -
Add support for log correlation for log4j and log4j2, even when not used in combination with slf4j. See [supported-logging-frameworks] for details.
-
Fix parsing value of
trace_methods
configuration property {pull}930[#930] -
Workaround for
java.util.logging
deadlock {pull}965[#965] -
JMS should propagate traceparent header when transactions are not sampled {pull}999[#999]
-
Spans are not closed if JDBC implementation does not support
getUpdateCount
{pull}1008[#1008]
-
JMS Enhancements {pull}911[#911]:
-
Add special handling for temporary queues/topics
-
Capture message bodies of text Messages
-
Rely on the existing
ELASTIC_APM_CAPTURE_BODY
agent config option (off by default). -
Send as
context.message.body
-
Limit size to 10000 characters. If longer than this size, trim to 9999 and append with ellipsis
-
-
Introduce the
ignore_message_queues
configuration to disable instrumentation (message tagging) for specific queues/topics as suggested in {pull}710[#710] -
Capture predefined message headers and all properties
-
Rely on the existing
ELASTIC_APM_CAPTURE_HEADERS
agent config option. -
Send as
context.message.headers
-
Sanitize sensitive headers/properties based on the
sanitize_field_names
config option
-
-
-
Added support for the MongoDB sync driver. See supported data stores.
-
JDBC regression-
PreparedStatement#executeUpdate()
andPreparedStatement#executeLargeUpdate()
are not traced {pull}918[#918] -
When systemd cgroup driver is used, the discovered Kubernetes pod UID contains "_" instead of "-" {pull}920[#920]
-
DB2 jcc4 driver is not traced properly {pull}926[#926]
-
Add the ability to configure a unique name for a JVM within a service through the
service_node_name
config option] -
Add ability to ignore some exceptions to be reported as errors <<config-ignore-exceptions[ignore_exceptions]
-
Applying new logic for JMS
javax.jms.MessageConsumer#receive
so that, instead of the transaction created for the polling method itself (ie fromreceive
start to end), the agent will create a transaction attempting to capture the code executed during actual message handling. This logic is suitable for environments where polling APIs are invoked within dedicated polling threads. This polling transaction creation strategy can be reversed through a configuration option (message_polling_transaction_strategy
) that is not exposed in the properties file by default. -
Send IP obtained through
javax.servlet.ServletRequest#getRemoteAddr()
incontext.request.socket.remote_address
instead of parsing from headers {pull}889[#889] -
Added
ElasticApmAttacher.attach(String propertiesLocation)
to specify a custom properties location -
Logs message when
transaction_max_spans
has been exceeded {pull}849[#849] -
Report the number of affected rows by a SQL statement (UPDATE,DELETE,INSERT) in 'affected_rows' span attribute {pull}707[#707]
-
Add
@Traced
annotation which either creates a span or a transaction, depending on the context -
Report JMS destination as a span/transaction context field {pull}906[#906]
-
Added
capture_jmx_metrics
configuration option
-
JMS creates polling transactions even when the API invocations return without a message
-
Support registering MBeans which are added after agent startup
-
Add ability to manually specify reported hostname
-
Add support for Redis Jedis client.
-
Add support for identifying target JVM to attach apm agent to using JVM property. See also the documentation of the
--include
and--exclude
flags -
Added
capture_jmx_metrics
configuration option -
Improve servlet error capture {pull}812[#812] Among others, now also takes Spring MVC `@ExceptionHandler`s into account
-
Instrument Logger#error(String, Throwable) {pull}821[#821] Automatically captures exceptions when calling
logger.error("message", exception)
-
Easier log correlation with https://github.com/elastic/java-ecs-logging. See docs.
-
Avoid creating a temp agent file for each attachment {pull}859[#859]
-
Instrument
View#render
instead ofDispatcherServlet#render
{pull}829[#829] This makes the transaction breakdown graph more useful. Instead ofdispatcher-servlet
, the graph now shows a type which is based on the view name, for example,FreeMarker
orThymeleaf
.
-
Error in log when setting server_urls to an empty string -
co.elastic.apm.agent.configuration.ApmServerConfigurationSource - Expected previousException not to be null
-
Avoid terminating the TCP connection to APM Server when polling for configuration updates {pull}823[#823]
-
Upgrading supported OpenTracing version from 0.31 to 0.33
-
Added annotation and meta-annotation matching support for
trace_methods
, for example:-
public @java.inject.* org.example.*
(for annotation) -
public @@javax.enterprise.context.NormalScope org.example.*
(for meta-annotation)
-
-
The runtime attachment now also works when the
tools.jar
or thejdk.attach
module is not available. This means you don’t need a full JDK installation - the JRE is sufficient. This makes the runtime attachment work in more environments such as minimal Docker containers. Note that the runtime attachment currently does not work for OSGi containers like those used in many application servers such as JBoss and WildFly. See the documentation for more information. -
Support for Hibernate Search
-
A warning in logs saying APM server is not available when using 1.8 with APM server 6.x. Due to that, agent 1.8.0 will silently ignore non-string labels, even if used with APM server of versions 6.7.x or 6.8.x that support such. If APM server version is <6.7 or 7.0+, this should have no effect. Otherwise, upgrade the Java agent to 1.9.0+.
-
ApacheHttpAsyncClientInstrumentation
matching increases startup time considerably -
Log correlation feature is active when
active==false
-
Tomcat’s memory leak prevention mechanism is causing a… memory leak. JDBC statement map is leaking in Tomcat if the application that first used it is undeployed/redeployed. See this related discussion.
-
The
apm-agent-attach.jar
is not executable anymore. Useapm-agent-attach-standalone.jar
instead.
-
Added support for tracking time spent by span type. Can be disabled by setting
breakdown_metrics
tofalse
. -
Added support for central configuration. Can be disabled by setting
central_config
tofalse
. -
Added support for Spring’s JMS flavor - instrumenting
org.springframework.jms.listener.SessionAwareMessageListener
-
Added support to legacy ApacheHttpClient APIs (which adds support to Axis2 configured to use ApacheHttpClient)
-
Added support for setting
server_urls
dynamically via properties file {pull}723[#723] -
Added
config_file
option -
Added option to use
@javax.ws.rs.Path
value as transaction nameuse_jaxrs_path_as_transaction_name
-
Instrument quartz jobs docs
-
SQL parsing improvements {pull}696[#696]
-
Introduce priorities for transaction name {pull}748[#748]. Now uses the path as transaction name if
use_path_as_transaction_name
is set totrue
rather thanServletClass#doGet
. But if a name can be determined from a high level framework, like Spring MVC, that takes precedence. User-supplied names from the API always take precedence over any others. -
Use JSP path name as transaction name as opposed to the generated servlet class name {pull}751[#751]
-
Some JMS Consumers and Producers are filtered due to class name filtering in instrumentation matching
-
Jetty: When no display name is set and context path is "/" transaction service names will now correctly fall back to configured values
-
JDBC’s
executeBatch
is not traced -
Drops non-String labels when connected to APM Server < 6.7 to avoid validation errors {pull}687[#687]
-
Parsing container ID in cloud foundry garden {pull}695[#695]
-
Automatic instrumentation should not override manual results {pull}752[#752]
-
The log correlation feature does not add
span.id
to the MDC anymore but onlytrace.id
andtransaction.id
{pull}742[#742].
-
Added the
trace_methods_duration_threshold
config option. When using thetrace_methods
config option with wild cards, this enables considerable reduction of overhead by limiting the number of spans captured and reported (see more details in config documentation). NOTE: Using wildcards is still not the recommended approach for thetrace_methods
feature. -
Add
Transaction#addCustomContext(String key, String|Number|boolean value)
to public API -
Added support for AsyncHttpClient 2.x
-
Added
global_labels
configuration option. This requires APM Server 7.2+. -
Added basic support for JMS- distributed tracing for basic scenarios of
send
,receive
,receiveNoWait
andonMessage
. Both Queues and Topics are supported. Asyncsend
APIs are not supported in this version. NOTE: This feature is currently marked as "experimental" and is disabled by default. In order to enable, it is required to set thedisable_instrumentations
configuration property to an empty string. -
Improved OSGi support: added a configuration option for
bootdelegation
packages {pull}641[#641] -
Better span names for SQL spans. For example,
SELECT FROM user
instead of justSELECT
{pull}633[#633]
-
ClassCastException related to async instrumentation of Pilotfish Executor causing thread hang (applied workaround)
-
NullPointerException when computing Servlet transaction name with null HTTP method name
-
FileNotFoundException when trying to find implementation version of jar with encoded URL
-
NullPointerException when closing Apache AsyncHttpClient request producer
-
Fixes loading of
elasticapm.properties
for Spring Boot applications -
Fix startup error on WebLogic 12.2.1.2.0 {pull}649[#649]
-
Disable metrics reporting and APM Server health check when active=false {pull}653[#653]
-
Fixes transaction name for non-sampled transactions #581
-
Makes log_file option work again #594
-
Async context propagation fixes
-
Fixes Manifest version reading error to support
jar:file
protocol #601 -
Fixes transaction name for non-sampled transactions #597
-
Fixes potential classloader deadlock by preloading
FileSystems.getDefault()
#603
-
Java APM Agent became part of the Cloud Foundry Java Buildpack as of Release v4.19
-
Support Apache HttpAsyncClient - span creation and cross-service trace context propagation
-
Added the
jvm.thread.count
metric, indicating the number of live threads in the JVM (daemon and non-daemon) -
Added support for WebLogic
-
Added support for Spring
@Scheduled
and EJB@Schedule
annotations - #569
-
If you didn’t explicitly set the
service_name
previously and you are dealing with a servlet-based application (including Spring Boot), yourservice_name
will change. See the documentation for <<config-service-name[service_name
] and the corresponding section in Features for more information. Note: this requires APM Server 7.0+. If using previous versions, nothing will change.
-
Added property
"allow_path_on_hierarchy"
to JAX-RS plugin, to lookup inherited usage of@path
-
Support for number and boolean labels in the public API {pull}497[497]. This change also renames
tag
tolabel
on the API level to be compliant with the Elastic Common Schema (ECS). TheaddTag(String, String)
method is still supported but deprecated in favor ofaddLabel(String, String)
. As of version 7.x of the stack, labels will be stored underlabels
in Elasticsearch. Previously, they were stored undercontext.tags
. -
Support async queries made by Elasticsearch REST client
-
Added
setStartTimestamp(long epochMicros)
andend(long epochMicros)
API methods toSpan
andTransaction
, allowing to set custom start and end timestamps. -
Auto-detection of the
service_name
based on the<display-name>
element of theweb.xml
with a fallback to the servlet context path. If you are using a spring-based application, the agent will use the setting forspring.application.name
for itsservice_name
. See the documentation forservice_name
for more information. Note: this requires APM Server 7.0+. If using previous versions, nothing will change. -
Previously, enabling
capture_body
could only capture form parameters. Now it supports all UTF-8 encoded plain-text content types. The optioncapture_body_content_types
controls which `Content-Type`s should be captured. -
Support async calls made by OkHttp client (
Call#enqueue
) -
Added support for providing config options on agent attach.
-
CLI example:
--config server_urls=http://localhost:8200,http://localhost:8201
-
API example:
ElasticApmAttacher.attach(Map.of("server_urls", "http://localhost:8200,http://localhost:8201"));
-
-
Logging integration through MDC is not working properly - #499
-
ClassCastException with adoptopenjdk/openjdk11-openj9 - #505
-
Span count limitation is not working properly - reported in our forum
-
Java agent causes Exceptions in Alfresco cluster environment due to failure in the instrumentation of Hazelcast `Executor`s - reported in our forum
-
Added support for sync calls of OkHttp client
-
Added support for context propagation for `java.util.concurrent.ExecutorService`s
-
The
trace_methods
configuration now allows to omit the method matcher. Example:com.example.*
traces all classes and methods within thecom.example
package and sub-packages. -
Added support for JSF. Tested on WildFly, WebSphere Liberty and Payara with embedded JSF implementation and on Tomcat and Jetty with MyFaces 2.2 and 2.3
-
Introduces a new configuration option
disable_metrics
which disables the collection of metrics via a wildcard expression. -
Support for HttpUrlConnection
-
Adds
subtype
andaction
to spans. This replaces former typing mechanism where type, subtype and action were all set through the type in an hierarchical dotted-syntax. In order to support existing API usages, dotted types are parsed into subtype and action, howeverSpan.createSpan
andSpan.setType
are deprecated starting this version. Instead, type-less spans can be created using the newSpan.startSpan
API and typed spans can be created using the newSpan.startSpan(String type, String subtype, String action)
API -
Support for JBoss EAP 6.4, 7.0, 7.1 and 7.2
-
Improved startup times
-
Support for SOAP (JAX-WS). SOAP client create spans and propagate context. Transactions are created for
@WebService
classes and@WebMethod
methods.
-
Fixes a failure in BitBucket when agent deployed #349
-
Fixed some OpenTracing bridge functionalities that were not working when auto-instrumentation is disabled
-
Fixed an error occurring when ending an OpenTracing span before deactivating
-
Sending proper
null
for metrics that have a NaN value -
Fixes JVM crash with Java 7 #458
-
Fixes an application deployment failure when using EclipseLink and
trace_methods
configuration #474
-
The agent now collects system and JVM metrics #360
-
Add API methods
ElasticApm#startTransactionWithRemoteParent
andSpan#injectTraceHeaders
to allow for manual context propagation #396. -
Added
trace_methods
configuration option which lets you define which methods in your project or 3rd party libraries should be traced. To create spans for allpublic
methods of classes whose name ends inService
which are in a sub-package oforg.example.services
use this matcher:public org.example.services..*Service#
#398 -
Added span for
DispatcherServlet#render
#409. -
Flush reporter on shutdown to make sure all recorded Spans are sent to the server before the program exits #397
-
Adds Kubernetes #383 and Docker metadata to, enabling correlation with the Kibana Infra UI.
-
Improved error handling of the Servlet Async API #399
-
Support async API’s used with AsyncContext.start #388
-
Fixing a potential memory leak when there is no connection with APM server
-
Fixes NoSuchMethodError CharBuffer.flip() which occurs when using the Elasticsearch RestClient and Java 7 or 8 #401
-
Added
capture_headers
configuration option. Set tofalse
to disable capturing request and response headers. This will reduce the allocation rate of the agent and can save you network bandwidth and disk space. -
Makes the API methods
addTag
,setName
,setType
,setUser
andsetResult
fluent, so that calls can be chained.
-
Catch all errors thrown within agent injected code
-
Enable public APIs and OpenTracing bridge to work properly in OSGi systems, fixes this WildFly issue
-
Remove module-info.java to enable agent working on early Tomcat 8.5 versions
-
Some memory allocation improvements
-
Enabling bootdelegation for agent classes in Atlassian OSGI systems
-
Update dsl-json which fixes a memory leak. See ngs-doo/dsl-json#102 for details.
-
Avoid `VerifyError`s by non instrumenting classes compiled for Java 4 or earlier
-
Enable APM Server URL configuration with path (fixes #339)
-
Reverse
system.hostname
andsystem.platform
order sent to APM server
-
Fixes NoSuchMethodError CharBuffer.flip() which occurs when using the Elasticsearch RestClient and Java 7 or 8 {pull}313[#313]
-
Remove intake v1 support. This version requires APM Server 6.5.0+ which supports the intake api v2. Until the time the APM Server 6.5.0 is officially released, you can test with docker by pulling the APM Server image via
docker pull docker.elastic.co/apm/apm-server:6.5.0-SNAPSHOT
.
-
Adds
@CaptureTransaction
and@CaptureSpan
annotations which let you declaratively add custom transactions and spans. Note that it is required to configure theapplication_packages
for this to work. See the documentation for more information. -
The public API now supports to activate a span on the current thread. This makes the span available via
ElasticApm#currentSpan()
Refer to the documentation for more details. -
Capturing of Elasticsearch RestClient 5.0.2+ calls. Currently, the
*Async
methods are not supported, only their synchronous counterparts. -
Added API methods to enable correlating the spans created from the JavaScrip Real User Monitoring agent with the Java agent transaction. More information can be found in the documentation.
-
Added
Transaction.isSampled()
andSpan.isSampled()
methods to the public API -
Added
Transaction#setResult
to the public API {pull}293[#293]
-
Fix for situations where status code is reported as
200
, even though it actually was500
{pull}225[#225] -
Capturing the username now properly works when using Spring security {pull}183[#183]
-
Remove intake v1 support. This version requires APM Server 6.5.0+ which supports the intake api v2. Until the time the APM Server 6.5.0 is officially released, you can test with docker by pulling the APM Server image via
docker pull docker.elastic.co/apm/apm-server:6.5.0-SNAPSHOT
. -
Wildcard patterns are case insensitive by default. Prepend
(?-i)
to make the matching case sensitive.
-
Support for Distributed Tracing
-
Adds
@CaptureTransaction
and@CaptureSpan
annotations which let you declaratively add custom transactions and spans. Note that it is required to configure theapplication_packages
for this to work. See the documentation for more information. -
The public API now supports to activate a span on the current thread. This makes the span available via
ElasticApm#currentSpan()
Refer to the documentation for more details. -
Capturing of Elasticsearch RestClient 5.0.2+ calls. Currently, the
*Async
methods are not supported, only their synchronous counterparts. -
Added API methods to enable correlating the spans created from the JavaScrip Real User Monitoring agent with the Java agent transaction. More information can be found in the documentation.
-
Microsecond accurate timestamps {pull}261[#261]
-
Support for JAX-RS annotations. Transactions are named based on your resources (
ResourceClass#resourceMethod
).
-
Fix for situations where status code is reported as
200
, even though it actually was500
{pull}225[#225]
-
Wildcard patterns are case insensitive by default. Prepend
(?-i)
to make the matching case sensitive.
-
Wildcard patterns are now not limited to only one wildcard in the middle and can be arbitrarily complex now. Example:
*foo*bar*baz
. -
Support for JAX-RS annotations. Transactions are named based on your resources (
ResourceClass#resourceMethod
).
-
Removed
ElasticApm.startSpan
. Spans can now only be created from their transactions viaTransaction#createSpan
. -
ElasticApm.startTransaction
andTransaction#createSpan
don’t activate the transaction and spans and are thus not available viaElasticApm.activeTransaction
andElasticApm.activeSpan
.
-
Public API
-
Add
Span#captureException
andTransaction#captureException
to public API.ElasticApm.captureException
is deprecated now. UseElasticApm.currentSpan().captureException(exception)
instead. -
Added
Transaction.getId
andSpan.getId
methods
-
-
Added support for async servlet requests
-
Added support for Payara/Glassfish
-
Incubating support for Apache HttpClient
-
Support for Spring RestTemplate
-
Added configuration options
use_path_as_transaction_name
andurl_groups
, which allow to use the URL path as the transaction name. As that could contain path parameters, like/user/$userId
however, You can set theurl_groups
option to define a wildcard pattern, like/user/*
, to group those paths together. This is especially helpful when using an unsupported Servlet API-based framework. -
Support duration suffixes (
ms
,s
andm
) for duration configuration options. Not using the duration suffix logs out a deprecation warning and will not be supported in future versions. -
Add ability to add multiple APM server URLs, which enables client-side load balancing. The configuration option
server_url
has been renamed toserver_urls
to reflect this change. However,server_url
still works for backwards compatibility. -
The configuration option
service_name
is now optional. It defaults to the main class name, the name of the executed jar file (removing the version number), or the application server name (for exampletomcat-application
). In a lot of cases, you will still want to set theservice_name
explicitly. But it helps getting started and seeing data easier, as there are no required configuration options anymore. In the future we will most likely determine more useful application names for Servlet API-based applications.