Commit 5a9b154d authored by Sasha Goldshtein's avatar Sasha Goldshtein

ustat: Add man page and companion example file

parent fb3c471f
.TH ustat 8 "2016-11-07" "USER COMMANDS"
.SH NAME
ustat \- Activity stats from high-level languages.
.SH SYNOPSIS
.B ustat [-l {java,python,ruby,node}] [-C] [-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d] [interval [count]]
.SH DESCRIPTION
This is "top" for high-level language events, such as garbage collections,
exceptions, thread creations, object allocations, method calls, and more. The
events are aggregated for each process and printed in a top-like table, which
can be sorted by various fields.
This uses in-kernel eBPF maps to store per process summaries for efficiency.
This tool relies on USDT probes embedded in many high-level languages, such as
Node, Java, Python, and Ruby. It requires a runtime instrumented with these
probes, which in some cases requires building from source with a USDT-specific
flag, such as "--enable-dtrace" or "--with-dtrace". For Java, some probes are
not enabled by default, and can be turned on by running the Java process with
the "-XX:+ExtendedDTraceProbes" flag.
Newly-created processes will only be traced at the next interval. If you run
this tool with a short interval (say, 1-5 seconds), this should be virtually
unnoticeable. For longer intervals, you might miss processes that were started
and terminated during the interval window.
Since this uses BPF, only the root user can use this tool.
.SH REQUIREMENTS
CONFIG_BPF and bcc.
.SH OPTIONS
.TP
\-l {java,python,ruby,node}
The language to trace. By default, all languages are traced.
.TP
\-C
Do not clear the screen between updates.
.TP
\-S {cload,excp,gc,method,objnew,thread}
Sort the output by the specified field.
.TP
\-r MAXROWS
Do not print more than this number of rows.
.TP
\-d
Print the resulting BPF program, for debugging purposes.
.TP
interval
Interval between updates, seconds.
.TP
count
Number of interval summaries.
.SH EXAMPLES
.TP
Summarize activity in high-level languages, 1 second refresh:
#
.B ustat
.TP
Don't clear the screen, and top 8 rows only:
#
.B ustat -Cr 8
.TP
5 second summaries, 10 times only:
#
.B ustat 5 10
.SH FIELDS
.TP
loadavg
The contents of /proc/loadavg
.TP
PID
Process ID.
.TP
CMDLINE
Process command line (often the second and following arguments will give you a
hint as to which application is being run.
.TP
METHOD/s
Count of method invocations during interval.
.TP
GC/s
Count of garbage collections during interval.
.TP
OBJNEW/s
Count of objects allocated during interval.
.TP
CLOAD/s
Count of classes loaded during interval.
.TP
EXC/s
Count of exceptions thrown during interval.
.TP
THR/s
Count of threads created during interval.
.SH OVERHEAD
When using this tool with high-frequency events, such as method calls, a very
significant slow-down can be expected. However, many of the high-level
languages covered by this tool already have a fairly high per-method invocation
cost, especially when running in interpreted mode. For the lower-frequency
events, such as garbage collections or thread creations, the overhead should
not be significant. Specifically, when probing Java processes and not using the
"-XX:+ExtendedDTraceProbes" flag, the most expensive probes are not emitted,
and the overhead should be acceptable.
.SH SOURCE
This is from bcc.
.IP
https://github.com/iovisor/bcc
.PP
Also look in the bcc distribution for a companion _example.txt file containing
example usage, output, and commentary for this tool.
.SH OS
Linux
.SH STABILITY
Unstable - in development.
.SH AUTHOR
Sasha Goldshtein
.SH SEE ALSO
trace(8), argdist(8), tplist(8)
Demonstrations of ustat.
ustat is a "top"-like tool for monitoring events in high-level languages. It
prints statistics about garbage collections, method calls, object allocations,
and various other events for every process that it recognizes with a Java,
Python, Ruby, or Node runtime.
For example:
# ./ustat.py
Tracing... Output every 10 secs. Hit Ctrl-C to end
12:17:17 loadavg: 0.33 0.08 0.02 5/211 26284
PID CMDLINE METHOD/s GC/s OBJNEW/s CLOAD/s EXC/s THR/s
3018 node/node 0 3 0 0 0 0
^C
Detaching...
If desired, you can instruct ustat to print a certain number of entries and
exit, which can be useful to get a quick picture on what's happening on the
system over a short time interval. Here, we ask ustat to print 5-second
summaries 12 times (for a total time of 1 minute):
# ./ustat.py -C 5 12
Tracing... Output every 5 secs. Hit Ctrl-C to end
12:18:26 loadavg: 0.27 0.11 0.04 2/336 26455
PID CMDLINE METHOD/s GC/s OBJNEW/s CLOAD/s EXC/s THR/s
3018 node/node 0 1 0 0 0 0
12:18:31 loadavg: 0.33 0.12 0.04 2/336 26456
PID CMDLINE METHOD/s GC/s OBJNEW/s CLOAD/s EXC/s THR/s
3018 node/node 0 0 0 0 0 0
26439 java -XX:+ExtendedDT 2776045 0 0 0 0 0
12:18:37 loadavg: 0.38 0.14 0.05 2/336 26457
PID CMDLINE METHOD/s GC/s OBJNEW/s CLOAD/s EXC/s THR/s
3018 node/node 0 0 0 0 0 0
26439 java -XX:+ExtendedDT 2804378 0 0 0 0 0
(...more output omitted for brevity)
USAGE message:
# ./ustat.py -h
usage: ustat.py [-h] [-l {java,python,ruby,node}] [-C]
[-S {cload,excp,gc,method,objnew,thread}] [-r MAXROWS] [-d]
[interval] [count]
Activity stats from high-level languages.
positional arguments:
interval output interval, in seconds
count number of outputs
optional arguments:
-h, --help show this help message and exit
-l {java,python,ruby,node}, --language {java,python,ruby,node}
language to trace (default: all languages)
-C, --noclear don't clear the screen
-S {cload,excp,gc,method,objnew,thread}, --sort {cload,excp,gc,method,objnew,thread}
sort by this field (descending order)
-r MAXROWS, --maxrows MAXROWS
maximum rows to print, default 20
-d, --debug Print the resulting BPF program (for debugging
purposes)
examples:
./ustat # stats for all languages, 1 second refresh
./ustat -C # don't clear the screen
./ustat -l java # Java processes only
./ustat 5 # 5 second summaries
./ustat 5 10 # 5 second summaries, 10 times only
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment