19 Introduction
|
This document describes JMeter properties. The properties present in jmeter.properties or reportgenerator.properties should be set in the user.properties file.
These properties are only taken into account after restarting JMeter as they are usually resolved when the class is loaded.
|
|
19.1 Language
|
Parameters
Attribute | Description | Required |
language |
Preferred GUI language. Comment out to use the JVM default locale's language.
Example:
This property is the only one that must be set in jmeter.properties file |
|
No
|
locales.add |
Additional locale(s) to add to the displayed list.
The current default list is: en, fr,
de, no, es, tr,
ja, zh_CN,
zh_TW, pl, pt_BR.
See JMeterMenuBar#makeLanguageMenu()
The entries are a comma-separated list of language names.
Example:
|
No
|
|
|
19.3 SSL configuration
|
SSL (Java) System properties are now in system.properties
JMeter no longer converts javax.xxx property entries in
jmeter.properties into System properties. These must now be
defined in the system.properties file or on the command-line. The
system.properties file gives more flexibility. |
Parameters
Attribute | Description | Required |
https.sessioncontext.shared |
By default, SSL session contexts are now created per-thread, rather than being shared.
The old behaviour can be enabled by setting this property to true. Defaults to: false
|
No
|
https.default.protocol |
Be aware that https default protocol may vary depending on the version of JVM.
See Diagnosing TLS, SSL and HTTPS
and Bug 58236
.
Default HTTPS protocol level:
|
|
|
|
https.default.protocol=TLS |
|
|
|
|
This may need to be changed to:
|
|
|
|
https.default.protocol=SSLv3 |
|
|
|
|
|
No
|
https.socket.protocols |
List of protocols to enable. You may have to select only a subset if you find issues with target server.
This is needed when server does not support Socket version negotiation, this can lead to errors like:
javax.net.ssl.SSLPeerUnverifiedException: peer not authenticated
or java.net.SocketException: Connection reset.
See Bug 54759
, example:
|
|
|
|
https.socket.protocols=SSLv2Hello SSLv3 TLSv1 |
|
|
|
|
|
No
|
https.cipherSuites |
Comma-separated list of SSL cipher suites that may be used in HTTPS
connections. It may be desirable to use a subset of cipher suites in order
to match expected client behavior or to reduce encryption overhead in
JMeter when running with large numbers of users. Errors may occur if the
JVM does not support the specified cipher suites, or if the cipher suites
supported by the HTTPS server do not overlap this list. See the
JSSE
Reference Guide.
For example:
|
|
|
|
https.cipherSuites=TLS_ECDHE_ECDSA_WITH_AES_256_CBC_SHA384,TLS_RSA_WITH_AES_128_GCM_SHA256 |
|
|
|
|
If not specified, JMeter will use the default list of cipher suites
supported by the JVM.
|
No
|
httpclient.reset_state_on_thread_group_iteration |
Reset HTTP State when starting a new Thread Group iteration.
In summarytrue means next iteration is associated to a new user.
false means next iteration is associated to same user.
true involves:
- Closing opened connection
- resetting SSL State
Defaults to: true
|
No
|
https.use.cached.ssl.context |
Control if we allow reuse of cached SSL context between iterations.
Set the value to false to reset the SSL context each iteration.
Defaults to: true
DEPRECATED, you should use httpclient.reset_state_on_thread_group_iteration with correct value |
|
No
|
https.keyStoreStartIndex |
Start index to be used with keystores with many entries.
The default is to use entry 0, i.e. the first.
Defaults to: 0
|
No
|
https.keyStoreEndIndex |
End index to be used with keystores with many entries.
Defaults to: 0
|
No
|
|
|
19.5 Toolbar display
|
Parameters
Attribute | Description | Required |
jmeter.toolbar.icons |
Toolbar icon definitions.
Defaults to org/apache/jmeter/images/toolbar/icons-toolbar.properties
|
No
|
jmeter.toolbar |
Toolbar list.
Defaults to:
|
|
|
|
new,open,close,save,save_as_testplan,|,cut,copy,paste,|,expand,collapse,toggle,|,test_start,test_stop,test_shutdown,|,test_start_remote_all,test_stop_remote_all,test_shutdown_remote_all,|,test_clear,test_clear_all,|,search,search_reset,|,function_helper,help |
|
|
|
|
|
No
|
jmeter.toolbar.icons.size |
Available sizes are: 22x22, 32x32, 48x48. Suggested value for HiDPI mode is
|
|
|
|
jmeter.toolbar.icons.size=48x48 |
|
|
|
|
Defaults to: 22x22
|
No
|
jmeter.icons |
Icon definitions. Alternate set:
|
|
|
|
jmeter.icons=org/apache/jmeter/images/icon_1.properties |
|
|
|
|
Historical icon set (deprecated):
|
|
|
|
jmeter.icons=org/apache/jmeter/images/icon_old.properties |
|
|
|
|
Defaults to:org/apache/jmeter/images/icon.properties
|
No
|
jmeter.tree.icons.size |
Available sizes are: 19x19, 24x24, 32x32, 48x48.
Useful for HiDPI display (see below).
Defaults to: 19x19
Suggested value for HiDPI screen like 3200x1800 is: 32x32
|
No
|
jmeter.hidpi.mode |
HiDPI mode. Activate a 'pseudo'-HiDPI mode. Allows to increase size of some UI
elements which are not correctly managed by JVM with high resolution screens in Linux or Windows.
Defaults to: false |
No
|
jmeter.hidpi.scale.factor |
HiDPI scale factor. Suggested value for HiDPI: 2.0.
Defaults to: 1.0
|
No
|
not_in_menu |
Components to not display in JMeter GUI (GUI class name or static label).
These elements are deprecated and will be removed in next version:
|
|
|
|
MongoDB Script, MongoDB Source Config |
|
|
|
|
Defaults to:
|
|
|
|
org.apache.jmeter.protocol.mongodb.sampler.MongoScriptSampler, org.apache.jmeter.protocol.mongodb.config.MongoSourceElement |
|
|
|
|
|
No
|
undo.history.size |
Number of items in undo history.
Feature is disabled by default (0) due to known and not fixed bugs
Bug 57043
, Bug 57039
and Bug 57040
.
Set it to a number greater than zero (25 can be a good default).
The bigger it is, the more memory will be consumed.
Defaults to: 0
|
No
|
gui.quick_X |
Hotkeys to add JMeter components where X is the shortcut key, for example:
|
|
|
|
gui.quick_0=ThreadGroupGui
gui.quick_1=HttpTestSampleGui
gui.quick_2=RegexExtractorGui
gui.quick_3=AssertionGui
gui.quick_4=ConstantTimerGui
gui.quick_5=TestActionGui
gui.quick_6=JSR223PostProcessor
gui.quick_7=JSR223PreProcessor
gui.quick_8=DebugSampler
gui.quick_9=ViewResultsFullVisualizer
|
|
|
|
|
Above code will add the corresponding elements when you press
Ctrl + 0 …
Ctrl + 9
(⌘ + 0 …
⌘ + 9 on Mac)
|
No
|
|
|
19.6 JMX Backup configuration
|
Parameters
Attribute | Description | Required |
jmeter.gui.action.save.backup_on_save |
Enable auto backups of the .jmx file when a test plan is saved.
When enabled, before the .jmx is saved, it will be backed up to the directory
pointed to by the jmeter.gui.action.save.backup_directory property (see below).
Backup file names are built after the jmx file being saved. For example,
saving test-plan.jmx will create a test-plan-000012.jmx in the backup
directory provided that the last created backup file is test-plan-000011.jmx.
Default value is true indicating that auto backups are enabled.
Defaults to: true
|
No
|
jmeter.gui.action.save.backup_directory |
Set the backup directory path where JMX backups will be created upon save in the GUI.
If not set (what it defaults to) then backup files will be created in a sub-directory of
the JMeter base installation. If set and the directory does not exist, a corresponding directory will be created.
Defaults to: ${JMETER_HOME}/backups
|
No
|
jmeter.gui.action.save.keep_backup_max_hours |
Set the maximum time (in hours) that backup files should be preserved since the save time.
By default no expiration time is set which means we keep backups for ever.
Defaults to: 0 |
No
|
jmeter.gui.action.save.keep_backup_max_count |
Set the maximum number of backup files that should be preserved.
By default ten backups will be preserved.
Setting this to zero will cause the backups to not being deleted (unless keep_backup_max_hours
is set to a non zero value).
Defaults to: 10
|
No
|
save_automatically_before_run |
Enable auto saving of the .jmx file before start run a test plan
When enabled, before the run, the .jmx will be saved and also backed up to the directory pointed.
Defaults to: true
|
No
|
|
|
19.7 Remote hosts and RMI configuration
|
Parameters
Attribute | Description | Required |
remote_hosts |
Remote Hosts - comma delimited, for example
|
|
|
|
remote_hosts=localhost:1099,localhost:2010 |
|
|
|
|
Defaults to: 127.0.0.1 |
No
|
server_port |
RMI port to be used by the server (must start rmiregistry with same port).
To change the port to (say) 1234:
On the server(s):
- set server_port=1234
- start rmiregistry with port 1234
On Windows this can be done by:
|
|
|
|
SET SERVER_PORT=1234
JMETER-SERVER |
|
|
|
|
On Unix:
|
|
|
|
SERVER_PORT=1234 jmeter-server |
|
|
|
|
On the Windows client:
|
|
|
|
set remote_hosts=server:1234 |
|
|
|
|
On the Unix client:
|
|
|
|
export remote_hosts=server:1234 |
|
|
|
|
Defaults to: 1099
|
No
|
client.rmi.localport |
Parameter that controls the RMI port used by the RemoteSampleListenerImpl (The Controller)
Default value is 0, which means port is randomly assigned.
You may need to open a port in the firewall on the Controller machine. |
Defaults to: 0
|
No
|
client.tries |
When distributed test is starting, there may be several attempts to initialize remote engines.
By default, only a single try is made. Increase this property to make it retry additional times.
Defaults to: 1
|
No
|
client.retries_delay |
If initialization is retried, this property sets the delay between those attempts in milliseconds.
Defaults to: 5000
|
No
|
client.continue_on_fail |
When all initialization tries were made, the test will fail, if any remote engines are failed.
Set this property to true to ignore failed nodes and proceed with test.
Defaults to: false
|
No
|
server.rmi.port |
To change the default port (1099) used to access the server.
Defaults to: 1099
|
No
|
server.rmi.localport |
To use a specific port for the JMeter server engine, define this property before starting the server.
Defaults to: 4000
|
No
|
server.rmi.create |
From JMeter version 2.3.1, the JMeter server creates the RMI registry as part of the server process.
Set this property to false, to stop the server creating the RMI registry.
Defaults to: true
|
No
|
server.exitaftertest |
From JMeter version 2.3.1, define this property to cause JMeter to exit after the first test.
Defaults to: true
|
No
|
server.rmi.ssl.keystore.type |
Type of keystore for RMI connection security. Possible values are dependent on the JVM in use, but commonly supported are JKS and PKCS12.
Defaults to: JKS
|
No
|
server.rmi.ssl.keystore.file |
Keystore file that contains private key
Defaults to: rmi_keystore.jks
|
No
|
server.rmi.ssl.keystore.password |
Password of Keystore
Defaults to: changeit
|
No
|
server.rmi.ssl.keystore.alias |
Key alias
Defaults to: rmi
|
No
|
server.rmi.ssl.truststore.type |
Type of truststore for RMI connection security
Defaults to: the value of server.rmi.ssl.keystore.type, which is JKS
|
No
|
server.rmi.ssl.truststore.file |
Keystore file that contains certificate
Defaults to: the value of server.rmi.ssl.keystore.file, which is rmi_keystore.jks
|
No
|
server.rmi.ssl.truststore.password |
Password of Trust store
Defaults to: the value of server.rmi.ssl.keystore.password, which is changeit
|
No
|
server.rmi.ssl.disable |
Set this to true if you don't want to use SSL for RMI
Defaults to: false
|
No
|
|
|
19.8 Include Controller
|
Parameters
Attribute | Description | Required |
includecontroller.prefix |
Prefix used by IncludeController when building file names.
Defaults to empty value
|
No
|
|
|
19.9 HTTP Java configuration
|
Parameters
Attribute | Description | Required |
http.java.sampler.retries |
Number of connection retries performed by HTTP Java sampler before giving up.
0 means no retry since version 3.0.
Defaults to: 0
|
No
|
|
|
19.11 Kerberos properties
|
Parameters
Attribute | Description | Required |
kerberos_jaas_application |
AuthManager Kerberos configuration
Name of application module used in jaas.conf.
Defaults to: JMeter
|
No
|
kerberos.spnego.strip_port |
Should port be stripped from urls before constructing SPNs for SPNEGO authentication.
Defaults to: true
|
No
|
kerberos.spnego.delegate_cred |
Should SPNEGO authentication should use delegation of credentials.
Defaults to: false
|
No
|
|
|
19.13 Apache HttpComponents HTTPClient configuration (HTTPClient4)
|
Parameters
Attribute | Description | Required |
hc.parameters.file |
Define a properties file for overriding Apache HttpClient parameters.
Uncomment this line if you put anything in hc.parameters file.
Defaults to: hc.parameters
|
No
|
httpclient4.auth.preemptive |
Preemptively send Authorization Header when BASIC auth is used
Defaults to: true
|
No
|
httpclient4.retrycount |
Number of retries to attempt.
Retry will be done on Idempotent Http Methods by default.
If you want to retry for all methods, see property httpclient4.request_sent_retry_enabled
Defaults to: 0
|
No
|
httpclient4.request_sent_retry_enabled |
Set this property to true if it's OK to retry requests that have been sent.
This mean that both Idempotent and non Idempotent requests will be retried.
This should usually be false, but it can be useful when testing against some Load Balancers like Amazon ELB.
Defaults to: false
|
No
|
httpclient4.idletimeout |
Idle connection timeout (in milliseconds) to apply if the server does not send Keep-Alive timeout headers.
Defaults to: 0 (no suggested duration for Keep-Alived connections)
|
No
|
httpclient4.validate_after_inactivity |
Check connection if the elapsed time (in milliseconds) since the last use of the connection
exceeds this value. Ensure this value is always lower by at least 150 ms than httpclient4.time_to_live
Defaults to: 1700
|
No
|
httpclient4.time_to_live |
TTL (in milliseconds) represents an absolute value.
No matter what, the connection will not be re-used beyond its TTL.
Defaults to: 2000
|
No
|
httpclient4.max_body_retain_size |
Max size in bytes of PUT body to retain in result sampler.
Bigger results will be clipped.
Defaults to: 327678 (bytes)
|
No
|
httpclient4.deflate_relax_mode |
Ignore EOFException that some edgy application may emit to signal end of Deflated stream.
Defaults to: false
|
No
|
httpclient4.gzip_relax_mode |
Ignore EOFException that some edgy application may emit to signal end of GZipped stream.
Defaults to: false
|
No
|
httpclient4.default_user_agent_disabled |
If true, default HC4 User-Agent (Apache-HttpClient/X.Y.Z (Java/A.B.C_D)) will not be added.
Defaults to: false
|
No
|
|
|
19.14 HTTP Cache Manager configuration
|
Parameters
Attribute | Description | Required |
cacheable_methods |
Space or comma separated list of methods that can be cached.
Defaults to: GET
|
No
|
cache_manager.cached_resource_mode |
N.B. This property is currently a temporary solution for Bug 56162
. |
Since version 2.12, JMeter does not create anymore a Sample Result with a response
code of 204 for a resource found in cache. This is in line with what browser do.
You can choose between three modes:
- RETURN_NO_SAMPLE (default)
- this mode returns no Sample Result. It has no additional configuration.
- RETURN_200_CACHE
- this mode will return Sample Result with response code to 200 and
response message to "(ex cache)".
- RETURN_CUSTOM_STATUS
- choosing this mode, response code and message have to be set by specifying
RETURN_CUSTOM_STATUS.code and RETURN_CUSTOM_STATUS.message.
Defaults to: RETURN_NO_SAMPLE
|
No
|
RETURN_CUSTOM_STATUS.code |
This lets you select what response code you want to return if mode RETURN_CUSTOM_STATUS
is selected.
Defaults to empty value.
|
No
|
RETURN_CUSTOM_STATUS.message |
This lets you select what response message you want to return if mode RETURN_CUSTOM_STATUS
is selected.
Defaults to empty value
|
No
|
|
|
19.15 Results file configuration
|
Parameters
Attribute | Description | Required |
jmeter.save.saveservice.output_format |
This section helps determine how result data will be saved.
The commented out values are the defaults.
Legitimate values: xml, csv, db.
Only xml and csv are currently supported.
Defaults to: csv
|
No
|
jmeter.save.saveservice.assertion_results_failure_message |
true when field should be saved; false otherwise.
assertion_results_failure_message only affects CSV output.
Defaults to: true
|
No
|
jmeter.save.saveservice.assertion_results |
Legitimate values: none, first, all.
Defaults to: none
|
No
|
jmeter.save.saveservice.data_type |
Defaults to: true
|
No
|
jmeter.save.saveservice.label |
Defaults to: true
|
No
|
jmeter.save.saveservice.response_code |
Defaults to: true
|
No
|
jmeter.save.saveservice.response_data |
response_data is currently not supported for CSV output |
Defaults to: false
|
No
|
jmeter.save.saveservice.response_data.on_error |
Save ResponseData for failed samples.
Defaults to: false
|
No
|
jmeter.save.saveservice.response_message |
Defaults to: true
|
No
|
jmeter.save.saveservice.successful |
Defaults to: true
|
No
|
jmeter.save.saveservice.thread_name |
Defaults to: true
|
No
|
jmeter.save.saveservice.time |
Defaults to: true
|
No
|
jmeter.save.saveservice.subresults |
Defaults to: true
|
No
|
jmeter.save.saveservice.assertions |
Defaults to: true
|
No
|
jmeter.save.saveservice.latency |
Defaults to: true
|
No
|
jmeter.save.saveservice.connect_time |
Defaults to: false
|
No
|
jmeter.save.saveservice.samplerData |
Defaults to: false
|
No
|
jmeter.save.saveservice.responseHeaders |
Defaults to: false
|
No
|
jmeter.save.saveservice.requestHeaders |
Defaults to: false
|
No
|
jmeter.save.saveservice.encoding |
Defaults to: false
|
No
|
jmeter.save.saveservice.bytes |
Defaults to: true
|
No
|
jmeter.save.saveservice.url |
Defaults to: false
|
No
|
jmeter.save.saveservice.filename |
Defaults to: false
|
No
|
jmeter.save.saveservice.hostname |
Defaults to: false
|
No
|
jmeter.save.saveservice.thread_counts |
Defaults to: true
|
No
|
jmeter.save.saveservice.sample_count |
Defaults to: false
|
No
|
jmeter.save.saveservice.idle_time |
Defaults to: true
|
No
|
jmeter.save.saveservice.timestamp_format |
Timestamp format - this only affects CSV output files.
Legitimate values: none, ms, or a format suitable for SimpleDateFormat.
Defaults to: ms
|
No
|
jmeter.save.saveservice.timestamp_format |
Defaults to: yyyy/MM/dd HH:mm:ss.SSS
|
No
|
jmeter.save.saveservice.default_delimiter |
For use with Comma-separated value (CSV) files or other formats where the fields' values
are separated by specified delimiters.
Defaults to: ,
|
No
|
jmeter.save.saveservice.print_field_names |
Only applies to CSV format files:
Print field names as first line in CSV
Defaults to: true
|
No
|
sample_variables |
Optional list of JMeter variable names whose values are to be saved in the result data files.
Use commas to separate the names.
Defaults to: SESSION_ID,REFERENCE
|
No
|
jmeter.save.saveservice.xml_pi |
N.B. The current implementation saves the values in XML as attributes, so the names must be valid XML names. |
Versions of JMeter after 2.3.2 send the variable to all servers to ensure that the correct data is available at the client.
Optional XML processing instruction for line two of the file.
Defaults to empty value
|
No
|
jmeter.save.saveservice.base_prefix |
Prefix used to identify filenames that are relative to the current base.
Defaults to: ~/
|
No
|
jmeter.save.saveservice.autoflush |
AutoFlush on each line written in XML or CSV output.
Setting this to true will result in less test results data loss in case of a crash, but
with impact on performances, particularly for intensive tests (low or no pauses).
Since JMeter version 2.10, this is false by default.
Defaults to: false
|
No
|
|
|
19.16 Settings that affect SampleResults
|
Parameters
Attribute | Description | Required |
sampleresult.timestamp.start |
Save the start time stamp instead of the end.
This also affects the timestamp stored in result files.
Defaults to: true
|
No
|
sampleresult.useNanoTime |
Whether to use System.nanoTime() - otherwise only use System.currentTimeMillis().
Defaults to: true
|
No
|
sampleresult.nanoThreadSleep |
Use a background thread to calculate the nanoTime offset.
Set this to a value less than zero to disable the background thread.
Defaults to: 5000
|
No
|
subresults.disable_renaming |
Since version 5.0 JMeter has a new SubResult Naming Policy which numbers subresults by default
This property if set to true discards renaming policy. This can be required if you're using JMeter for functional testing.
Defaults to: false
|
No
|
|
|
19.17 Upgrade
|
Parameters
Attribute | Description | Required |
upgrade_properties |
File that holds a record of name changes for backward compatibility issues.
Defaults to: /bin/upgrade.properties
|
No
|
|
|
19.18 JMeter Test Script recorder configuration
|
Parameters
Attribute | Description | Required |
proxy.pause |
N.B. The element was originally called the Proxy recorder, which is why the properties
have the prefix "proxy". |
If the recorder detects a gap of at least 5s (default) between HTTP requests, it assumes
that the user has clicked a new URL. Specified in milliseconds.
Defaults to: 5000
|
No
|
proxy.number.requests |
Add numeric prefix to Sampler names.
defaults to: true
|
No
|
proxy.excludes.suggested |
List of URL patterns that will be added to URL Patterns to exclude.
Separate multiple lines with ;
Defaults to: .*\\.(bmp|css|js|gif|ico|jpe?g|png|swf|woff|woff2)
|
No
|
jmeter.httpsampler |
Change the default HTTP Sampler.
Can be one of
- HTTPSampler or Java
- Use the Java Sampler
- HTTPSampler2
- HttpClient4
- Use Apache HTTPClient version 4
Defaults to: HttpClient4
|
No
|
jmeter.httpclient.strict_rfc2616 |
By default JMeter tries to be more lenient with RFC 2616
redirects and allows relative paths.
If you want to test strict conformance, set this value to true.
When the property is true, JMeter follows
RFC 3986 section 5.2.
Defaults to: false
|
No
|
proxy.content_type_include |
Default content-type include filter to use. Specified as a regex.
Defaults to: text/html|text/plain|text/xml
|
No
|
proxy.content_type_exclude |
Default content-type exclude filter to use. Specified as a regex.
Defaults to: image/.*|text/css|application/.*
|
No
|
proxy.headers.remove |
Default headers to remove from Header Manager elements. Specified as comma separated list
The headers Cookie and Authorization are always removed. |
Defaults to: If-Modified-Since,If-None-Match,Host
|
No
|
proxy.binary.types |
Binary content-type handling.
These content-types will be handled by saving the request in a file.
Defaults to: application/x-amf,application/x-java-serialized-object,binary/octet-stream
|
No
|
proxy.binary.directory |
The files will be saved in this directory.
Defaults to: user.dir
|
No
|
proxy.binary.filesuffix |
The files will be created suffixed with this value.
Defaults to: .binary
|
No
|
proxy.redirect.disabling |
Whether to attempt disabling of samples that resulted from redirects where the
generated samples use auto-redirection.
Defaults to: true
|
No
|
proxy.ssl.protocol |
SSL configuration.
Defaults to: TLS
|
No
|
|
|
19.19 Test Script Recorder certificate configuration
|
Parameters
Attribute | Description | Required |
proxy.cert.directory |
Defaults to: JMeter bin directory
|
No
|
proxy.cert.file |
Defaults to: proxyserver.jks
|
No
|
proxy.cert.type |
Defaults to: JKS
|
No
|
proxy.cert.keystorepass |
Defaults to: password
|
No
|
proxy.cert.keypassword |
Defaults to: password
|
No
|
proxy.cert.factory |
Defaults to: SunX509
|
No
|
proxy.cert.alias |
Define this property if you wish to use a special entry from the keystore.
Defaults to empty value
|
No
|
proxy.cert.validity |
The default validity (in days) for certificates created by JMeter.
Defaults to: 7
|
No
|
proxy.cert.dynamic_keys |
Use dynamic key generation (if supported by JMeter/JVM).
If false, will revert to using a single key with no certificate.
Defaults to: true
|
No
|
|
|
19.20 JMeter Proxy configuration
|
Parameters
Attribute | Description | Required |
http.proxyDomain |
Use command-line flags for user-name and password.
Defaults to: NTLM domain, if required by HTTPClient sampler
|
No
|
|
|
19.21 HTML Parser configuration
|
Parameters
Attribute | Description | Required |
HTTPResponse.parsers |
Space-separated list of parser groups.
For each parser, there should be a parser.types and a parser.className property |
Defaults to: htmlParser wmlParser cssParser
|
No
|
cssParser.className |
CSS Parser based on ph-css.
Defaults to: org.apache.jmeter.protocol.http.parser.CssParser
|
No
|
cssParser.types |
Content types handled by cssParser.
Defaults to: text/css
|
No
|
css.parser.cache.size |
CSS parser LRU cache size. This cache stores the URLs found in a CSS to avoid continuously
parsing the CSS. By default the cache size is 400. It can be disabled by setting its value to 0.
Defaults to: 400
|
No
|
css.parser.ignore_all_css_errors |
Let the CSS Parser ignore all CSS errors.
Defaults to: true
|
No
|
htmlParser.className |
Define the HTML parser to be used.
Do not comment this property.
- org.apache.jmeter.protocol.http.parser.LagartoBasedHtmlParser
- This new parser (since 2.10) should perform better than all others. See Bug 55632
.
- org.apache.jmeter.protocol.http.parser.JTidyHTMLParser
- Default parser before JMeter version 2.10
- org.apache.jmeter.protocol.http.parser.RegexpHTMLParser
Note that Regexp extractor may detect references that have been commented out. |
In many cases it will work OK, but you should be aware that it may generate additional references.
- org.apache.jmeter.protocol.http.parser.JsoupBasedHtmlParser
- This parser is based on JSoup. It should be the most accurate parser,
but it is less performant than LagartoBasedHtmlParser
Defaults to: org.apache.jmeter.protocol.http.parser.LagartoBasedHtmlParser
|
Yes
|
htmlParser.types |
Used by HTTPSamplerBase to associate htmlParser with content types below.
Defaults to: text/html application/xhtml+xml application/xml text/xml
|
No
|
wmlParser.className |
Defaults to: org.apache.jmeter.protocol.http.parser.RegexpHTMLParser
|
No
|
wmlParser.types |
Used by HTTPSamplerBase to associate wmlParser with content types below.
Defaults to: text/vnd.wap.wml
|
No
|
|
|
19.22 Remote batching configuration
|
Configure how SampleResults are sent from server to client when using distributed testing.
Note that the mode is currently resolved on the client, while other properties
(e.g. time_threshold) are resolved on the server. |
Since JMeter version 2.9, default is StrippedBatch, which returns samples in
batch mode (every 100 samples or every minute by default).
You can set mode by configuring:
|
|
|
|
mode=one of the possible modes below |
|
|
|
|
StrippedBatch strips response data from SampleResult, so if you need the response data, change to another mode. |
Possible modes are:
- Standard
- Sends SampleResult one by one
- Batch
- Accumulates SampleResults before sending them. Configured by
properties num_sample_threshold and time_threshold
- Statistical
- returns sample summary statistics. Configured by
properties key_on_threadname and time_threshold
- Stripped
- Similar to Standard mode but strips Response from SampleResult.
Configured by property sample_sender_strip_also_on_error
- StrippedBatch
- Same as Batch but strips Response from SampleResult.
Configured by properties num_sample_threshold, time_threshold
and sample_sender_strip_also_on_error
- Asynch
- Asynchronous sender; uses a queue and background worker process to return the samples.
Configured by property asynch.batch.queue.size
- StrippedAsynch
- Same as Asynch but strips response data from SampleResult.
Configured by properties asynch.batch.queue.size
and sample_sender_strip_also_on_error
- StrippedDiskStore
- Same as DiskStore but strips response data from SampleResult
- Class extending AbstractSampleSender (org.example.load.MySampleSender for example)
- A custom implementation of your choice
Parameters
Attribute | Description | Required |
sample_sender_client_configured |
How is Sample sender implementations configured:
- true
- (default) means client configuration will be used
- false
- means server configuration will be used
Defaults to: true
|
No
|
sample_sender_strip_also_on_error |
By default when Stripping modes are used JMeter since version 3.1 will strip
response even for SampleResults in error. If you want to revert to previous
behaviour (no stripping of Responses in error) set this property to false
Defaults to: true
|
No
|
mode |
Remote batching support.
Since JMeter version 2.9, default is StrippedBatch, which returns samples in
batch mode (every 100 samples or every minute by default).
Note also that StrippedBatch strips response data from SampleResult, so if you need
the response data, change to another mode. |
|
No
|
key_on_threadname |
Set to true to key statistical samples on threadName rather than threadGroup.
Defaults to: false
|
No
|
num_sample_threshold |
Number of SampleResults to accumulate before sending to client.
Defaults to: 100
|
No
|
time_threshold |
Time to retain SampleResults before sending them to client.
Value is in milliseconds.
Defaults to: 60000
|
No
|
asynch.batch.queue.size |
Default queue size used by Async mode.
Defaults to: 100
|
No
|
|
|
19.24 OS Process Sampler configuration
|
Parameters
Attribute | Description | Required |
os_sampler.poll_for_timeout |
Polling to see if process has finished its work, used when a timeout
is configured on sampler. Specified in milliseconds.
Defaults to: 100
|
No
|
|
|
19.25 TCP Sampler configuration
|
Parameters
Attribute | Description | Required |
tcp.handler |
The default handler class.
Defaults to: TCPClientImpl
|
No
|
tcp.eolByte |
Set this to a value outside the range -128 to +127 to skip
eol checking.
Defaults to byte value for end of line: 1000
|
No
|
tcp.charset |
TCP Charset, used by org.apache.jmeter.protocol.tcp.sampler.TCPClientImpl.
Defaults to platforms default charset as returned by Charset.defaultCharset().name()
|
No
|
tcp.status.prefix |
String at the beginning of the status response code.
Defaults to: Status |
No
|
tcp.status.suffix |
String at the end of the status response code.
defaults to: .
|
No
|
tcp.status.properties |
Property file to convert codes to messages.
Defaults to: mytestfiles/tcpstatus.properties
|
No
|
tcp.binarylength.prefix.length |
The length prefix used by LengthPrefixedBinaryTCPClientImpl implementation (in bytes).
Defaults to: 2 |
No
|
|
|
19.26 Summariser - Generate Summary Results - configuration (mainly applies to CLI mode)
|
Parameters
Attribute | Description | Required |
summariser.name |
Comment the following property to disable the default CLI mode summariser.
[or change the value to rename it]
Defaults to: summary
|
No
|
summariser.interval |
Interval between summaries (in seconds).
Defaults to: 30
|
No
|
summariser.log |
Write messages to log file.
Defaults to: true
|
No
|
summariser.out |
Write messages to System.out.
Defaults to: true
|
No
|
summariser.ignore_transaction_controller_sample_result |
Ignore SampleResults generated by TransactionControllers.
Defaults to: true
|
No
|
|
|
19.27 Aggregate Report and Aggregate Graph - configuration
|
Parameters
Attribute | Description | Required |
aggregate_rpt_pct1 |
Percentiles to display in reports.
Given as a float value between 0 and 100 (means percent).
First percentile to display.
Defaults to: 90
|
No
|
aggregate_rpt_pct2 |
Second percentile to display.
Given as a float value between 0 and 100 (means percent).
Defaults to: 95
|
No
|
aggregate_rpt_pct3 |
Second percentile to display.
Given as a float value between 0 and 100 (means percent).
Defaults to: 99
|
No
|
|
|
19.28 BackendListener - configuration
|
Parameters
Attribute | Description | Required |
backend_graphite.send_interval |
Send interval in seconds.
Defaults to: 1 second
|
No
|
backend_influxdb.send_interval |
Send interval in seconds.
Defaults to: 5 seconds
|
No
|
backend_influxdb.connection_timeout |
InfluxDB connection timeout.
Defaults to: 1000 millis
|
No
|
backend_influxdb.socket_timeout |
InfluxDB socket read timeout.
Defaults to: 3000 millis
|
No
|
backend_influxdb.connection_request_timeout |
InfluxDB timeout to get a connection.
Defaults to: 100 millis
|
No
|
backend_metrics_window |
Backend metrics sliding window size for Percentiles, Min
and Max.
Defaults to: 100
|
No
|
backend_metrics_large_window |
Backend metrics sliding window size for Percentiles, Min
and Max. when backend_metrics_window_mode=timed
Setting this value too high can lead to OOM Backend metrics sliding window size
Defaults to: 5000
|
No
|
backend_metrics_window_mode |
Backend metrics window mode.
Possible values:
- fixed : fixed-size window
- timed : time boxed
Defaults to: fixed
|
No
|
|
|
19.29 BeanShell configuration
|
Parameters
Attribute | Description | Required |
beanshell.server.port |
BeanShell Server properties.
Define the port number as non-zero to start the http server on that port.
The telnet server will be started on the next port.
Defaults to: 0 (i.e. don't start the server)
There is no security. Anyone who can connect to the port can issue any BeanShell commands.
These can provide unrestricted access to the JMeter application and the host.
Do not enable the server unless the ports are protected against access, e.g. by a firewall.
|
|
No
|
beanshell.server.file |
Define the server initialisation file.
Defaults to: ../extras/startup.bsh
|
No
|
beanshell.init.file |
Define a file to be processed at startup.
This is processed using its own interpreter.
Defaults to empty value.
|
No
|
beanshell.sampler.init |
Define the initialisation files for BeanShell Sampler, Function and
other BeanShell elements.
N.B. Beanshell test elements do not share interpreters.
Each element in each thread has its own interpreter.
This is retained between samples. |
Defaults to empty value.
|
No
|
beanshell.function.init |
Defaults to empty value.
|
No
|
beanshell.assertion.init |
Defaults to empty value.
|
No
|
beanshell.listener.init |
Defaults to empty value.
|
No
|
beanshell.postprocessor.init |
Defaults to empty value.
|
No
|
beanshell.preprocessor.init |
Defaults to empty value.
|
No
|
beanshell.timer.init |
Defaults to empty value.
|
No
|
The file BeanShellListeners.bshrc contains sample definitions
of Test and Thread Listeners.
|
|
19.30 MailerModel configuration
|
Parameters
Attribute | Description | Required |
mailer.successlimit |
Number of successful samples before a message is sent.
Defaults to: 2
|
No
|
mailer.failurelimit |
Number of failed samples before a message is sent.
Defaults to: 2
|
No
|
|
|
19.31 CSVRead configuration
|
Parameters
Attribute | Description | Required |
csvread.delimiter |
CSVRead delimiter setting (default ",").
Make sure that there are no trailing spaces or tabs after the delimiter
characters, or these will be included in the list of valid delimiters. |
Defaults to: ,
|
No
|
|
|
19.32 __time() function configuration
|
Parameters
Attribute | Description | Required |
time.YMD |
This and the following properties can be used to redefine the default time formats.
Defaults to: yyyyMMdd
|
No
|
time.HMS |
Defaults to: HHmmss
|
No
|
time.YMDHMS |
Defaults to: yyyyMMdd-HHmmss
|
No
|
time.USER1 |
Defaults to empty value
|
No
|
time.USER2 |
Defaults to empty value
|
No
|
|
|
19.33 CSV DataSet configuration
|
Parameters
Attribute | Description | Required |
csvdataset.eofstring |
String to return at EOF (if recycle not used).
Defaults to: <EOF>
|
No
|
csvdataset.file.encoding_list |
List of file encoding values
Defaults to: platform default
|
No
|
|
|
19.34 LDAP Sampler configuration
|
Parameters
Attribute | Description | Required |
ldapsampler.max_sorted_results |
Maximum number of search results returned by a search that will be sorted
to guarantee a stable ordering (if more results then this limit are returned
then no sorting is done).
Set to zero to turn off all sorting, in which case "Equals" response assertions
will be very likely to fail against search results.
Defaults to: 1000
|
No
|
assertion.equals_section_diff_len |
Number of characters to log for each of three sections (starting matching section,
diff section, ending matching section where not all sections will appear for all diffs)
diff display when an Equals assertion fails. So a value of 100 means a
maximum of 300 characters of diff text will be displayed (plus a number
of extra characters like "..." and "[[["/"]]]"
which are used to decorate it).
Defaults to: 100
|
No
|
assertion.equals_diff_delta_start |
Test written out to log to signify start/end of diff delta.
Defaults to: [[[
|
No
|
assertion.equals_diff_delta_end |
Defaults to: ]]]
|
No
|
|
|
19.35 Miscellaneous configuration
|
Parameters
Attribute | Description | Required |
cssselector.parser.cache.size |
Size of cache used by CSS Selector Extractor (for JODD implementation only) to store parsed CSS Selector expressions.
Defaults to: 400
|
No
|
resultcollector.action_if_file_exists |
Used to control what happens when you start a test and have listeners that could overwrite existing result files.
Possible values:
- ASK : Ask user
- APPEND : Append results to existing file
- DELETE : Delete existing file and start a new file
|
No
|
mirror.server.port |
If defined and greater then zero, then start the mirror server on the port.
Defaults to: 0
|
No
|
oro.patterncache.size |
ORO PatternCacheLRU size.
Defaults to: 1000
|
No
|
propertyEditorSearchPath |
TestBeanGui
Defaults to: null
|
No
|
jmeter.expertMode |
Turn expert mode on/off: expert mode will show expert-mode beans and properties.
Defaults to: true
|
No
|
httpsampler.max_bytes_to_store_per_request |
Max size of bytes stored in memory per SampleResult. Ensure that you
don't exceed the maximum capacity of a Java Array and remember that the higher you
set this value, the more memory JMeter will consume.
Defaults to: 0 bytes which means no truncation will occur
|
No
|
httpsampler.max_buffer_size |
Max size of buffer in bytes used when reading responses.
Defaults to: 66560 bytes
|
No
|
httpsampler.max_redirects |
Maximum redirects to follow in a single sequence.
Defaults to: 20
|
No
|
httpsampler.max_frame_depth |
Maximum frame/iframe nesting depth.
defaults to: 5
|
No
|
httpsampler.separate.container |
Revert to Bug 51939
behaviour (no separate container for embedded resources)
by setting the following false.
defaults to: true
|
No
|
httpsampler.ignore_failed_embedded_resources |
If embedded resources download fails due to missing resources or other reasons,
if this property is true, Parent sample will not be marked as failed.
Defaults to: false
|
No
|
httpsampler.parallel_download_thread_keepalive_inseconds |
Keep-alive time for the parallel download threads (in seconds).
Defaults to: 60
|
No
|
httpsampler.embedded_resources_use_md5 |
Don't keep the embedded resources response data; just keep the size and the MD5 sum.
Defaults to: false
|
No
|
httpsampler.user_defined_methods |
List of extra HTTP methods that should be available in select box.
Defaults to:
|
|
|
|
VERSION-CONTROL,REPORT,CHECKOUT,CHECKIN,UNCHECKOUT,MKWORKSPACE,UPDATE,LABEL,MERGE,BASELINE-CONTROL,MKACTIVITY |
|
|
|
|
|
No
|
sampleresult.default.encoding |
The encoding to be used if none is provided.
Defaults to: ISO-8859-1
|
No
|
CookieManager.delete_null_cookies |
CookieManager behaviour - should cookies with null/empty values be deleted?
Use false to revert to original behaviour.
Defaults to: true
|
No
|
CookieManager.allow_variable_cookies |
CookieManager behaviour - should variable cookies be allowed?
Use false to revert to original behaviour.
Defaults to: true
|
No
|
CookieManager.save.cookies |
CookieManager behaviour - should Cookies be stored as variables?
Default to: false
|
No
|
CookieManager.name.prefix |
CookieManager behaviour - prefix to add to cookie name before storing it as a variable.
Default is COOKIE_; to remove the prefix, define it as one or more spaces.
Defaults to: COOKIE_ |
No
|
CookieManager.check.cookies |
CookieManager behaviour - check received cookies are valid before storing them?
Use false to revert to previous behaviour.
Defaults to: true
|
No
|
cookies |
Netscape HTTP Cookie file.
Defaults to: cookies
|
No
|
javascript.use_rhino |
Ability to switch to Rhino as default Javascript Engine used by IfController
and __javaScript function.
JMeter uses Nashorn since 3.2 version.
If you want to use Rhino, set this value to true |
Defaults to: false
|
No
|
jmeterengine.threadstop.wait |
Number of milliseconds to wait for a thread to stop.
Defaults to: 5000
|
No
|
jmeterengine.remote.system.exit |
Whether to invoke System.exit(0) in server exit code after
stopping RMI.
Defaults to: false
|
No
|
jmeterengine.stopfail.system.exit |
Whether to call System.exit(1) on failure to stop threads in CLI mode.
This only takes effect if the test was explicitly requested to stop.
If this is disabled, it may be necessary to kill the JVM externally.
Defaults to: true
|
No
|
jmeterengine.force.system.exit |
Whether to force call System.exit(0) at end of test in CLI mode,
even if there were no failures and the test was not explicitly asked to stop.
Without this, the JVM may never exit if there are other threads spawned by
the test which never exit.
Defaults to: false
|
No
|
jmeter.exit.check.pause |
How long to pause (in ms) in the daemon thread before reporting that the JVM has
failed to exit.
If the value is less than zero, the JMeter does not start the daemon thread
Defaults to: 2000
|
No
|
jmeterengine.nongui.port |
If running CLI mode, then JMeter listens on the following port for a shutdown message.
To disable, set the port to 1000 or less.
Defaults to: 4445
|
No
|
jmeterengine.nongui.maxport |
If the initial port is busy, keep trying until this port is reached
(to disable searching, set the value less than or equal to the .port property).
Defaults to: 4455
|
No
|
jmeterthread.rampup.granularity |
How often to check for shutdown during ramp-up (milliseconds).
Defaults to: 1000
|
No
|
onload.expandtree |
Should JMeter expand the tree when loading a test plan?
Default value is false since JMeter 2.7
Defaults to: false
|
No
|
jsyntaxtextarea.wrapstyleword |
JSyntaxTextArea configuration.
Defaults to: true
|
No
|
jsyntaxtextarea.linewrap |
Defaults to: true
|
No
|
jsyntaxtextarea.codefolding |
Defaults to: true
|
No
|
jsyntaxtextarea.maxundos |
Set to zero to disable undo feature in JSyntaxTextArea.
Defaults to: 50
|
No
|
jsyntaxtextarea.font.family |
Change the font on the (JSyntax) Text Areas. (Useful for HiDPI screens).
Defaults to empty value, which means platform default monospaced font
|
No
|
jsyntaxtextarea.font.size |
Change the size of the (JSyntax) Text Areas. Will be used only,
when jsyntaxtextarea.font.family is set.
Defaults to: -1 |
No
|
loggerpanel.usejsyntaxtext |
Set this to false to disable the use of JSyntaxTextArea
for the Console Logger panel.
Defaults to: true
|
No
|
view.results.tree.max_results |
Maximum number of main samples, that should be stored and displayed.
A value of 0 will store all results. This might consume a lot of memory.
Defaults to: 500
|
No
|
view.results.tree.max_size |
Maximum size (in bytes) of HTML page that can be displayed.
Set to zero to disable the size check and display the whole response.
Defaults to: 10485760
|
No
|
view.results.tree.renderers_order |
Order of Renderers in View Results Tree.
Note full class names should be used for non JMeter core renderers |
For JMeter core renderers, class names start with . and are automatically
prefixed with org.apache.jmeter.visualizers
Defaults to:
|
|
|
|
.RenderAsText,.RenderAsRegexp,.RenderAsCssJQuery,.RenderAsXPath,.RenderAsHTML,.RenderAsHTMLWithEmbedded,.RenderAsDocument,.RenderAsJSON,.RenderAsXML |
|
|
|
|
|
No
|
document.max_size |
Maximum size (in bytes) of Document that can be parsed by Tika engine
Set to zero to disable the size check.
Defaults to: 10485760
|
No
|
JMSSampler.useSecurity.properties |
JMS options.
Enable the following property to stop JMS Point-to-Point Sampler from using
the properties java.naming.security.[principal|credentials] when
creating the queue connection.
Defaults to: false
|
No
|
confirm.delete.skip |
Set the following value to true in order to skip the delete
confirmation dialogue.
Defaults to: false
|
No
|
|
|
19.36 Classpath configuration
|
Parameters
Attribute | Description | Required |
search_paths |
List of directories (separated by ;) to search for additional
JMeter plugin classes, for example new GUI elements and samplers.
Any jar file in such a directory will be automatically included; jar files in sub directories are ignored.
The given value is in addition to any jars found in the lib/ext directory.
Do not use this for utility or plugin dependency jars.
Defaults to empty value.
|
No
|
user.classpath |
List of directories that JMeter will search for utility and plugin dependency classes.
Use your platform path separator (java.io.File.pathSeparatorChar in Java) to separate multiple paths.
Any jar file in such a directory will be automatically included; jar files in sub directories are ignored.
The given value is in addition to any jars found in the lib directory.
All entries will be added to the class path of the system class loader and also to the path
of the JMeter internal loader.
Paths with spaces may cause problems for the JVM.
Defaults to empty value.
|
No
|
plugin_dependency_paths |
List of directories (separated by ;) that JMeter will search for utility
and plugin dependency classes.
Any jar file in such a directory will be automatically included; jar files in sub directories are ignored.
The given value is in addition to any jars found in the lib directory
or given by the user.classpath property.
All entries will be added to the path of the JMeter internal loader only.
For plugin dependencies this property should be used instead of user.classpath.
Defaults to empty value.
|
No
|
classfinder.functions.contain |
The classpath finder currently needs to load every single JMeter class to find the classes it needs.
For CLI mode, it's only necessary to scan for Function classes, but all classes are still loaded.
All current Function classes include ".function." in their name, and none include
".gui." in the name, so the number of unwanted classes loaded can be reduced by
checking for these. However, if a valid function class name does not match these restrictions,
it will not be loaded. If problems are encountered, then comment or change this or the following property.
Defaults to: .functions.
|
No
|
classfinder.functions.notContain |
Defaults to: .gui.
|
No
|
|
|
19.38 Additional property files to load
|
Parameters
Attribute | Description | Required |
user.properties |
Should JMeter automatically load additional JMeter properties?
File name to look for (comment to disable)
Defaults to: user.properties
|
No
|
system.properties |
Should JMeter automatically load additional system properties?
File name to look for (comment to disable)
Defaults to: system.properties
|
No
|
template.files |
Comma separated list of files that contain reference to templates and their description.
Path must be relative to JMeter root folder
Defaults to: /bin/templates/templates.xml
|
No
|
|
|
19.42 Help
|
Controls how documentation in JMeter is displayed
Parameters
Attribute | Description | Required |
help.local |
Switch that allows using Local documentation opened in JMeter GUI.
By default we use Online documentation opened in Browser.
Defaults to false
|
No
|
|
|
19.43 Advanced Groovy Scripting configuration
|
Advanced properties for configuration of scripting in Groovy
Parameters
Attribute | Description | Required |
groovy.utilities |
Path to Groovy file containing utility functions to make available to __groovy function.
Defaults to bin/utility.groovy
|
No
|
|
|
19.44 Advanced JSR-223 Scripting configuration
|
Advanced properties for configuration of scripting in JSR-223
Parameters
Attribute | Description | Required |
jsr223.init.file |
Path to JSR-223 file containing script to call on JMeter startup.
The actual scripting engine to use will be determined by the extension
of the init file name. If the file name has no extension, or no scripting
engine could be found for that extension, Groovy will be used.
This script can use pre-defined variables:
- log: Logger to log any message, uses SLF4J library
- props: JMeter Properties
- OUT: System.OUT, useful to write in the console
No script is defined by default.
|
No
|
jsr223.compiled_scripts_cache_size |
Used by JSR-223 elements.
Size of compiled scripts cache.
Defaults to: 100 |
No
|
|
|
19.45 Documentation generation
|
Advanced properties for documentation generation
Parameters
Attribute | Description | Required |
docgeneration.schematic_xsl |
Path to XSL file used to generate Schematic View of Test Plan.
When empty, JMeter will use the embedded one in src/core/org/apache/jmeter/gui/action/schematic.xsl
No default value |
No
|
|
|
|