arthas/site/docs/en/doc/profiler.md

profiler

profiler online tutorial

::: tip Generate a flame graph using async-profiler :::

The profiler command supports generate flame graph for application hotspots.

The basic usage of the profiler command is profiler action [actionArg]

Supported Options

Name	Specification
action	Action to execute
actionArg	Attribute name pattern
[i:]	sampling interval in ns (default: 10'000'000, i.e. 10 ms)
[f:]	dump output to specified directory
[d:]	run profiling for specified seconds
[e:]	which event to trace (cpu, alloc, lock, cache-misses etc.), default value is cpu

Start profiler

$ profiler start
Started [cpu] profiling

::: tip By default, the sample event is cpu. Can be specified with the --event parameter. :::

Get the number of samples collected

$ profiler getSamples
23

View profiler status

$ profiler status
[cpu] profiling is running for 4 seconds

Can view which event and sampling time.

Stop profiler

Generating html format results

By default, the result file is html format. You can also specify it with the --format parameter:

$ profiler stop --format html
profiler output file: /tmp/test/arthas-output/20211207-111550.html
OK

Or use the file name name format in the --file parameter. For example, --file /tmp/result.html.

View profiler results under arthas-output via browser

By default, arthas uses port 3658, which can be opened: http://localhost:3658/arthas-output/ View the arthas-output directory below Profiler results:

Click to view specific results:

::: tip If using the chrome browser, may need to be refreshed multiple times. :::

Profiler supported events

Under different platforms and different OSs, the supported events are different. For example, under macos:

$ profiler list
Basic events:
  cpu
  alloc
  lock
  wall
  itimer

Under linux

$ profiler list
Basic events:
  cpu
  alloc
  lock
  wall
  itimer
Perf events:
  page-faults
  context-switches
  cycles
  instructions
  cache-references
  cache-misses
  branches
  branch-misses
  bus-cycles
  L1-dcache-load-misses
  LLC-load-misses
  dTLB-load-misses
  mem:breakpoint
  trace:tracepoint

If you encounter the permissions/configuration issues of the OS itself and then missing some events, you can refer to the async-profiler documentation.

You can use the --event parameter to specify the event to sample, such as sampling the alloc event:

$ profiler start --event alloc

Resume sampling

$ profiler resume
Started [cpu] profiling

The difference between start and resume is: start is the new start sampling, resume will retain the data of the last stop.

You can verify the number of samples by executing profiler getSamples.

Use execute action to execute complex commands

For example, start sampling:

profiler execute 'start,framebuf=5000000'

Stop sampling and save to the specified file:

profiler execute 'stop,file=/tmp/result.html'

Specific format reference: arguments.cpp

View all supported actions

$ profiler actions
Supported Actions: [resume, dumpCollapsed, getSamples, start, list, execute, version, stop, load, dumpFlat, actions, dumpTraces, status]

View version

$ profiler version
Async-profiler 1.6 built on Sep  9 2019
Copyright 2019 Andrei Pangin

Configure framebuf option

::: tip you encounter [frame_buffer_overflow] in the generated result, you need to increase the framebuf (the default value is 1'000'000), which can be configured explicitly, such as: :::

profiler start --framebuf 5000000

Configure include/exclude to filter data

If the application is complex and generates a lot of content, and you want to focus on only part of stack traces, you can filter stack traces by --include/--exclude. --include defines the name pattern that must be present in the stack traces, while --exclude is the pattern that must not occur in any of stack traces in the output.A pattern may begin or end with a star * that denotes any (possibly empty) sequence of characters. such as

profiler start --include'java/*' --include 'com/demo/*' --exclude'*Unsafe.park*'

Both --include/--exclude support setting multiple values, but need to be configured at the end of the command line.

Specify execution time

For example, if you want the profiler to automatically end after 300 seconds, you can specify it with the -d/--duration parameter:

profiler start --duration 300

Generate jfr format result

Note that jfr only supports configuration at start. If it is specified at stop, it will not take effect.

profiler start --file /tmp/test.jfr

The file parameter supports some variables:

• Timestamp: --file /tmp/test-%t.jfr
• Process ID: --file /tmp/test-%p.jfr

The generated results can be viewed with tools that support the jfr format. such as:

• JDK Mission Control: https://github.com/openjdk/jmc
• JProfiler: https://github.com/alibaba/arthas/issues/1416

The 'unknown' in profiler result

• https://github.com/jvm-profiling-tools/async-profiler/discussions/409