headers
Tom | 1 Sep 2010 01:01
Picon

Re: s="false" lb="HTTP Request HTTPClient" rc="200" in the generated jtl file


http://jmeter.512774.n5.nabble.com/file/n2798784/test-plan-template.jmx
test-plan-template.jmx 

Here’s the test plan. ‘XXXX’ means I can replace it with the proper value.
In Response Assertion, besides checking for code 200, can you please show a
general way to verify a page return besides code 200?

--

-- 
View this message in context: http://jmeter.512774.n5.nabble.com/s-false-lb-HTTP-Request-HTTPClient-rc-200-in-the-generated-jtl-file-tp2795583p2798784.html
Sent from the JMeter - User mailing list archive at Nabble.com.
Permalink | Reply | 0 comments |
headers
sebb | 1 Sep 2010 01:34
Picon

Re: Fails to run a proxy-recorded test plan without GUI

On 31 August 2010 14:23, Wilhelmsen Tor Iver <TorIverW <at> arrive.no> wrote:
>> We are trying to run a test plan recorded using the JMeter 2.4 GUI and
>> HTTP Proxy, but are running into problems: Only a few of the HTTP
>> samplers in the plan actually work!
>
> Seems the cause was using the HTTPClient sampler
(org.apache.jmeter.protocol.http.sampler.HTTPSampler2), switching back to the "default" sampler
makes the tests run fine.

That's a bit odd.

The HttpClient sampler does use different code to process the response
headers, but it should not be failing like this.

Can you provide a sample (public) URL that shows the problem?
[I tried the URL in the previous posting and it worked OK for me]

> - Tor Iver
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe <at> jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help <at> jakarta.apache.org
>
>
Permalink | Reply | 0 comments |
headers
Himanshu Ghai | 1 Sep 2010 01:48
Picon

Re: problem with remote beanshell configuration

will try it..thanks for quick responses

On Tue, Aug 31, 2010 at 3:37 PM, sebb <sebbaz <at> gmail.com> wrote:

> On 31 August 2010 23:31, Himanshu Ghai <himanshughai <at> gmail.com> wrote:
> > Thanks Sebb, is there a way that I can get it to execute when I run
> > remotely..
>
> No.
>
> It is a startup script, it runs when you start JMeter, not when you
> start a test.
>
> However you can use a BeanShell test element wherever you like in the test
> plan.
>
> >
> > On Tue, Aug 31, 2010 at 3:15 PM, sebb <sebbaz <at> gmail.com> wrote:
> >
> >> On 31 August 2010 21:11, Himanshu Ghai <himanshughai <at> gmail.com> wrote:
> >> > It executes on the controller machine when I launch jmeter with this
> >> command
> >> >
> >> > ./jmeter -t plan_xxx.jmx -r
> >> >
> >> > then I run it on the configured remote hosts using menu option..it
> >> doesn't
> >> > execute there
> >> > on the remote hosts. one thing i noticed is that this code(.bsh) gets
> >> > executed not when
(Continue reading)

Permalink | Reply | 2 comments |
headers
sebb | 1 Sep 2010 01:52
Picon

Re: StandardSamplerSender MarshalException during ftp get test

On 27 August 2010 17:07, shigella Gazeta.pl <shigella <at> gazeta.pl> wrote:
> Dear Team
> I run distributed test of FTP.
> Test case is simple ftp plan, it contains thread config, ftp request and
> Aggregate Report and View Results Tree listeners.
> When I test "put", I could send even large, 20GB files and I get results in
> Aggregate Report and View Results Tree.
> When I tried to test "get" in the same way, I can transfer only small, 50 MB
> files only.
> For 100MB file I get exception listed below and in aggregate report are no
> results (from time to time result of one single transfer is visible in view
> results tree.
> Summary report behaves as Aggregate report.
> Test file was transferred.
> I'm able to get it through FTP manually
> Do you know the reason and remedy to my problem?

Did you select the checkbox "Save File in Response?" ?
If so, them try de-selecting it.

Does the FTP get work OK if you don't use client-server mode?
> Sincerely yours
> Zofia Nowinska
>
> Exception:
> 2010/08/27 07:57:10 INFO  - jmeter.engine.RemoteJMeterEngineImpl: Creating
> JMeter engine on host 10.124.124.56
> 2010/08/27 07:57:10 INFO  - jmeter.engine.RemoteJMeterEngineImpl: running
> test
> Starting the test on host 10.124.131.56  <at>  Fri Aug 27 07:57:10 CDT 2010
(Continue reading)

Permalink | Reply | 0 comments |
headers
sebb | 1 Sep 2010 02:10
Picon

Re: How to add 'control' field to an LDAP modify request

On 24 August 2010 18:06, Sorin Nicolae Ghebarau
<nicolae.ghebarau <at> computaris.com> wrote:
> Hi,
>
> I have problems in creating a specific ldapmodify in Jmeter.
>
> This is the LDAP modify that I need to make in Jmeter:
>
>
>
> ldapmodify -x -h localhost -p 389 -D cn=auser -w apassword -f ldiffile.ldif
>
>
>
> ----ldiffile.ldif start-----
>
> dn: uid=automation <at> test.com,key=auseruid,o=test,o=test.com
>
> control: 1.3.6.1.4.1.15573.1.2.1.2
>
> changetype: modify
>
> replace: ipV4Address
>
> ipV4Address: F1F1F1F1
>
> ----ldiffile.ldif end-----
>
>
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Wilhelmsen Tor Iver | 1 Sep 2010 11:06
Picon
Favicon

jmeter-server and already busy 1099

It seems that jmeter-server (jmeter -s) fails if some other Java code has fired up RMI server on the official
port. Could it instead test and if registry already runs instead bind to it instead of just assume it does
not run? (I am not sure if it actually is RMI that has grabbed the port though, or if it just has been assigned
to some other socket;  Windows etc/services file does not "reserve" the port, I will try to do that and see if
it helps.)

CMD>jmeter-server
Changing to JMeter home directory
Could not find ApacheJmeter_core.jar ...
... Trying JMETER_HOME=..
Found ApacheJMeter_core.jar
Created remote object: UnicastServerRef [liveRef: [endpoint:[....
Problem creating registry: java.rmi.server.ExportException: Port already in use: 1099; nested
exception is:
        java.net.BindException: Address already in use: JVM_Bind
Continuing...
Server failed to start: java.rmi.RemoteException: Cannot start. See server log file.
errorlevel=1

Med vennlig hilsen

TOR IVER WILHELMSEN
Senior systemutvikler
Arrive AS
T (+47) 48 16 06 18
E-post: toriverw <at> arrive.no
http://www.arrive.no
http://servicedesk.arrive.no
(Continue reading)

Permalink | Reply | 5 comments |
headers
sebb | 1 Sep 2010 12:41
Picon

Re: jmeter-server and already busy 1099

On 1 September 2010 10:06, Wilhelmsen Tor Iver <TorIverW <at> arrive.no> wrote:
> It seems that jmeter-server (jmeter -s) fails if some other Java code has fired up RMI server on the
official port. Could it instead test and if registry already runs instead bind to it instead of just assume
it does not run? (I am not sure if it actually is RMI that has grabbed the port though, or if it just has been
assigned to some other socket;  Windows etc/services file does not "reserve" the port, I will try to do
that and see if it helps.)

The code running the RMI server needs to have certain jars on the
classpath, otherwise it will not be able to find JMeter.
So just running the RMI server on its own won't work.
I don't know if it is possible to code around this.

If you define the property:

server.rmi.create=false

the JMeter won't try to create the server, so you can try it out.

> CMD>jmeter-server
> Changing to JMeter home directory
> Could not find ApacheJmeter_core.jar ...
> ... Trying JMETER_HOME=..
> Found ApacheJMeter_core.jar
> Created remote object: UnicastServerRef [liveRef: [endpoint:[....
> Problem creating registry: java.rmi.server.ExportException: Port already in use: 1099; nested
exception is:
>        java.net.BindException: Address already in use: JVM_Bind
> Continuing...
> Server failed to start: java.rmi.RemoteException: Cannot start. See server log file.
> errorlevel=1
(Continue reading)

Permalink | Reply | 4 comments |
headers
Wilhelmsen Tor Iver | 1 Sep 2010 14:01
Picon
Favicon

SV: jmeter-server and already busy 1099

> The code running the RMI server needs to have certain jars on the
> classpath, otherwise it will not be able to find JMeter.

As I understood the use of rmiregistry/rmid was that the serialization and deserialization happened in
the (running) VM that had bound a name in the registry to an object, and in the client making the remote call.
Thus, as far as the registry was concerned it is just shuffling some binary data over a socket or two. 

Using server.rmi.create=false and running rmiregistry separately (as per the JRE executable) seems to
work fine, but just to be on the safe side my services file now includes

java-rmi	 1099/tcp			    #Java RJMP

- Tor Iver
Permalink | Reply | 3 comments |
headers
sebb | 1 Sep 2010 14:41
Picon

Re: jmeter-server and already busy 1099

On 1 September 2010 13:01, Wilhelmsen Tor Iver <TorIverW <at> arrive.no> wrote:
>> The code running the RMI server needs to have certain jars on the
>> classpath, otherwise it will not be able to find JMeter.
>
> As I understood the use of rmiregistry/rmid was that the serialization and deserialization happened in
the (running) VM that had bound a name in the registry to an object, and in the client making the remote call.
Thus, as far as the registry was concerned it is just shuffling some binary data over a socket or two.

The classpath setting for the RMI server has been present since JMeter 1.9.
Perhaps it was necessary then - or perhaps it was necessary with early JVMs.

>
> Using server.rmi.create=false and running rmiregistry separately (as per the JRE executable) seems to
work fine, but just to be on the safe side my services file now includes
>
> java-rmi         1099/tcp                           #Java RJMP

OK, in that case the server code could be amended to carry on if there
was already a server running on the port.

Please raise a Bugzilla enhancement issue.

> - Tor Iver
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe <at> jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help <at> jakarta.apache.org
>
>
(Continue reading)

Permalink | Reply | 2 comments |
headers
Wilhelmsen Tor Iver | 1 Sep 2010 15:13
Picon
Favicon

Re: jmeter-server and already busy 1099

> OK, in that case the server code could be amended to carry on if there
> was already a server running on the port.
> 
> Please raise a Bugzilla enhancement issue.

Done: https://issues.apache.org/bugzilla/show_bug.cgi?id=49858

- Tor Iver
Permalink | Reply | 1 comment |

Gmane