You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by "Christensen, Alan" <al...@hp.com> on 2005/11/28 17:11:32 UTC
Testing the effect of page compression with Jmeter; Any gotchas?
Has anyone used Jmeter to test the effects of using page compression vs
not using page compression? Any gotchas that should be avoided, or best
practices that will make this easier? What would you suggest be used to
mimic slower connections?
RE: Point-To-Point JMS Test Plan
Posted by Kim Loh <ki...@pd.jaring.my>.
I meant am I able to do this using JMeter to test the processing time of
submitting a message onto a queue and the time reach the database.
Thanks.
Regards,
Kim
-----Original Message-----
From: Kim Loh [mailto:kimlkn@pd.jaring.my]
Sent: Wednesday, November 30, 2005 12:03 AM
To: 'Peter Lin'; 'JMeter Users List'
Subject: RE: Point-To-Point JMS Test Plan
I am facing a trouble here where I put a request to the queue but receive
reply from another component (Not from ActiveMQ provider). In other words,
if I will need to measure the processing time submitting a message into a
queue and the time it reached to a database (eg. postgresql). Am I able to
do this via jndi settings to lookup for the database location or any
recommendation?
Thanks.
Regards,
Kim
_____
From: Peter Lin [mailto:woolfel@gmail.com]
Sent: Tuesday, November 29, 2005 11:50 PM
To: JMeter Users List; kimlkn@pd.jaring.my
Subject: Re: Point-To-Point JMS Test Plan
the current implementation for JMS samplers has been tested against
ActiveMQ, OpenJMS, Orion Jms, IBM MQSeries and Weblogic.
you will need to copy the necessary jar files to jmeter/lib directory, or
use the new classpath feature in the nightly build. After that, you'll need
to provide the correct jndi settings to lookup the provider. depending on
the JMS implementation, it may use jndi.properties instead of a full blown
jndi server.
there's a short explanation of the jms samplers in the user manual.
hope that helps
peter
On 11/29/05, Kim Loh <ki...@pd.jaring.my> wrote:
Hi,
I am trying to test point-to-point JMS messaging (ActiveMQ).
I wonder how JMeter can help me to test on the components which have no
direct interfaces to ActiveMQ.
Does anybody have done so?
Thanks.
Kim
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
RE: Point-To-Point JMS Test Plan
Posted by Kim Loh <ki...@pd.jaring.my>.
I am facing a trouble here where I put a request to the queue but receive
reply from another component (Not from ActiveMQ provider). In other words,
if I will need to measure the processing time submitting a message into a
queue and the time it reached to a database (eg. postgresql). Am I able to
do this via jndi settings to lookup for the database location or any
recommendation?
Thanks.
Regards,
Kim
_____
From: Peter Lin [mailto:woolfel@gmail.com]
Sent: Tuesday, November 29, 2005 11:50 PM
To: JMeter Users List; kimlkn@pd.jaring.my
Subject: Re: Point-To-Point JMS Test Plan
the current implementation for JMS samplers has been tested against
ActiveMQ, OpenJMS, Orion Jms, IBM MQSeries and Weblogic.
you will need to copy the necessary jar files to jmeter/lib directory, or
use the new classpath feature in the nightly build. After that, you'll need
to provide the correct jndi settings to lookup the provider. depending on
the JMS implementation, it may use jndi.properties instead of a full blown
jndi server.
there's a short explanation of the jms samplers in the user manual.
hope that helps
peter
On 11/29/05, Kim Loh <ki...@pd.jaring.my> wrote:
Hi,
I am trying to test point-to-point JMS messaging (ActiveMQ).
I wonder how JMeter can help me to test on the components which have no
direct interfaces to ActiveMQ.
Does anybody have done so?
Thanks.
Kim
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: Point-To-Point JMS Test Plan
Posted by Peter Lin <wo...@gmail.com>.
the current implementation for JMS samplers has been tested against
ActiveMQ, OpenJMS, Orion Jms, IBM MQSeries and Weblogic.
you will need to copy the necessary jar files to jmeter/lib directory, or
use the new classpath feature in the nightly build. After that, you'll need
to provide the correct jndi settings to lookup the provider. depending on
the JMS implementation, it may use jndi.properties instead of a full blown
jndi server.
there's a short explanation of the jms samplers in the user manual.
hope that helps
peter
On 11/29/05, Kim Loh <ki...@pd.jaring.my> wrote:
>
> Hi,
> I am trying to test point-to-point JMS messaging (ActiveMQ).
>
> I wonder how JMeter can help me to test on the components which have no
> direct interfaces to ActiveMQ.
>
> Does anybody have done so?
>
> Thanks.
>
> Kim
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
Re: XML output
Posted by Morten Kristiansen <mo...@tracetracker.com>.
Perhaps somewhere in section 18.3 in the Component Reference?
http://jakarta.apache.org/jmeter/usermanual/component_reference.html
Morten
On Tue, 2005-11-29 at 15:00 +0000, sebb wrote:
> Of course, but we would have to create one first...
>
> I just meant - if you looked for the meaning of the fields somewhere
> in the existing documentation, where would be the place(s) you looked?
> There are quite a few files the comprise the current documentation.
>
> S
> On 29/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> >
> > Couldn't you just reference a DTD or and xml schema or something? That's
> > good enough documentation for me.
> >
> > Morten
> >
> >
> > On Tue, 2005-11-29 at 00:24 +0000, sebb wrote:
> > > From reading the code:
> > >
> > > t = elapsed time for whole sample
> > > lt = latency = time to first response
> > >
> > > It looks as though the latency only applies to the HttpSampler (plus
> > > its derived classes AccessLogSampler and SoapSampler) at present.
> > >
> > > I guess it ought to be added to the others (if appropriate).
> > >
> > > And the mnemonics ought to be documented ... not sure exactly where though.
> > > Where would you have looked for them to be documented?
> > >
> > > S.
> > >
> > > On 28/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> > > >
> > > > Hmm. The label says "lt" and not "it" doesn't it? And each element in the <httpSample> represents one request, so I don't see how "iteration" fits in this case. The number 329 says nothing to me. My test was 1 thread with 5 iterations and I got 5 <httpSample> results (like the one below) in my result file. Anyone?
> > > >
> > > > Morten
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: Peter Lin [mailto:woolfel@gmail.com]
> > > > Sent: Mon 11/28/2005 5:40 PM
> > > > To: JMeter Users List; Morten Kristiansen
> > > > Subject: Re: XML output
> > > >
> > > > t is time
> > > > it is iteration
> > > > ts is timestamp
> > > > s is success
> > > > rc is response code
> > > >
> > > > hope that answers your questions
> > > >
> > > > peter
> > > >
> > > > On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> > > > >
> > > > >
> > > > > An example from an xml output from JMeter. What is "lt" here? I guess
> > > > > "t" is time (or total time)?
> > > > >
> > > > >
> > > > > <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> > > > > rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
> > > > >
> > > > > Morten
> > > > >
> > > > >
> > > > > ---------------------------------------------------------------------
> > > > > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > > > > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> > > > >
> > > > >
> > > >
> > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> >
> >
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: XML output
Posted by sebb <se...@gmail.com>.
Of course, but we would have to create one first...
I just meant - if you looked for the meaning of the fields somewhere
in the existing documentation, where would be the place(s) you looked?
There are quite a few files the comprise the current documentation.
S
On 29/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
>
> Couldn't you just reference a DTD or and xml schema or something? That's
> good enough documentation for me.
>
> Morten
>
>
> On Tue, 2005-11-29 at 00:24 +0000, sebb wrote:
> > From reading the code:
> >
> > t = elapsed time for whole sample
> > lt = latency = time to first response
> >
> > It looks as though the latency only applies to the HttpSampler (plus
> > its derived classes AccessLogSampler and SoapSampler) at present.
> >
> > I guess it ought to be added to the others (if appropriate).
> >
> > And the mnemonics ought to be documented ... not sure exactly where though.
> > Where would you have looked for them to be documented?
> >
> > S.
> >
> > On 28/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> > >
> > > Hmm. The label says "lt" and not "it" doesn't it? And each element in the <httpSample> represents one request, so I don't see how "iteration" fits in this case. The number 329 says nothing to me. My test was 1 thread with 5 iterations and I got 5 <httpSample> results (like the one below) in my result file. Anyone?
> > >
> > > Morten
> > >
> > >
> > > -----Original Message-----
> > > From: Peter Lin [mailto:woolfel@gmail.com]
> > > Sent: Mon 11/28/2005 5:40 PM
> > > To: JMeter Users List; Morten Kristiansen
> > > Subject: Re: XML output
> > >
> > > t is time
> > > it is iteration
> > > ts is timestamp
> > > s is success
> > > rc is response code
> > >
> > > hope that answers your questions
> > >
> > > peter
> > >
> > > On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> > > >
> > > >
> > > > An example from an xml output from JMeter. What is "lt" here? I guess
> > > > "t" is time (or total time)?
> > > >
> > > >
> > > > <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> > > > rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
> > > >
> > > > Morten
> > > >
> > > >
> > > > ---------------------------------------------------------------------
> > > > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > > > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> > > >
> > > >
> > >
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: XML output
Posted by Morten Kristiansen <mo...@tracetracker.com>.
Couldn't you just reference a DTD or and xml schema or something? That's
good enough documentation for me.
Morten
On Tue, 2005-11-29 at 00:24 +0000, sebb wrote:
> From reading the code:
>
> t = elapsed time for whole sample
> lt = latency = time to first response
>
> It looks as though the latency only applies to the HttpSampler (plus
> its derived classes AccessLogSampler and SoapSampler) at present.
>
> I guess it ought to be added to the others (if appropriate).
>
> And the mnemonics ought to be documented ... not sure exactly where though.
> Where would you have looked for them to be documented?
>
> S.
>
> On 28/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> >
> > Hmm. The label says "lt" and not "it" doesn't it? And each element in the <httpSample> represents one request, so I don't see how "iteration" fits in this case. The number 329 says nothing to me. My test was 1 thread with 5 iterations and I got 5 <httpSample> results (like the one below) in my result file. Anyone?
> >
> > Morten
> >
> >
> > -----Original Message-----
> > From: Peter Lin [mailto:woolfel@gmail.com]
> > Sent: Mon 11/28/2005 5:40 PM
> > To: JMeter Users List; Morten Kristiansen
> > Subject: Re: XML output
> >
> > t is time
> > it is iteration
> > ts is timestamp
> > s is success
> > rc is response code
> >
> > hope that answers your questions
> >
> > peter
> >
> > On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> > >
> > >
> > > An example from an xml output from JMeter. What is "lt" here? I guess
> > > "t" is time (or total time)?
> > >
> > >
> > > <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> > > rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
> > >
> > > Morten
> > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> > >
> > >
> >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
RE: Efficient Messages Loading Method
Posted by Łukasz Żebrowski <lu...@rodan.pl>.
Hi, I think _StringFromFile would by much better:
http://jakarta.apache.org/jmeter/usermanual/functions.html#__StringFromF
ile
-----Original Message-----
From: Kim Loh [mailto:kimlkn@pd.jaring.my]
Sent: Monday, December 19, 2005 4:26 PM
To: 'JMeter Users List'
Subject: Efficient Messages Loading Method
Hello,
My performance objective is to use JMeter to measure the scalability of
my
application deals with increased load of messages.
I use JMeter to load the large volume of messages to my application.
However, I am facing problem where I need to create my messages from
somewhere else, then I need to copy and paste those created messages one
by
one to JMeter "Message Area". That's means if I will to load 5,000
messages
then I have to copy and paste 5,000 messages ONE BY ONE on Jmeter Test
Plan.
That is how I was doing so far, I was copy & paste 250 messages manually
but
it is impossible for me if continue doing so for more and more load.
Does anybody can advise me am I using the correct way of what Jmeter
provided or may be I miss out something that Jmeter can automate the
messages loading process? Or may be anybody can share a more productive
way
to load many messages for messaging processing load test in Jmeter?
Thanks.
Holy X'mas.
Regards,
Kim
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Efficient Messages Loading Method
Posted by Kim Loh <ki...@pd.jaring.my>.
Hello,
My performance objective is to use JMeter to measure the scalability of my
application deals with increased load of messages.
I use JMeter to load the large volume of messages to my application.
However, I am facing problem where I need to create my messages from
somewhere else, then I need to copy and paste those created messages one by
one to JMeter "Message Area". That's means if I will to load 5,000 messages
then I have to copy and paste 5,000 messages ONE BY ONE on Jmeter Test Plan.
That is how I was doing so far, I was copy & paste 250 messages manually but
it is impossible for me if continue doing so for more and more load.
Does anybody can advise me am I using the correct way of what Jmeter
provided or may be I miss out something that Jmeter can automate the
messages loading process? Or may be anybody can share a more productive way
to load many messages for messaging processing load test in Jmeter?
Thanks.
Holy X'mas.
Regards,
Kim
Point-To-Point JMS Test Plan
Posted by Kim Loh <ki...@pd.jaring.my>.
Hi,
I am trying to test point-to-point JMS messaging (ActiveMQ).
I wonder how JMeter can help me to test on the components which have no
direct interfaces to ActiveMQ.
Does anybody have done so?
Thanks.
Kim
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: XML output
Posted by sebb <se...@gmail.com>.
>From reading the code:
t = elapsed time for whole sample
lt = latency = time to first response
It looks as though the latency only applies to the HttpSampler (plus
its derived classes AccessLogSampler and SoapSampler) at present.
I guess it ought to be added to the others (if appropriate).
And the mnemonics ought to be documented ... not sure exactly where though.
Where would you have looked for them to be documented?
S.
On 28/11/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
>
> Hmm. The label says "lt" and not "it" doesn't it? And each element in the <httpSample> represents one request, so I don't see how "iteration" fits in this case. The number 329 says nothing to me. My test was 1 thread with 5 iterations and I got 5 <httpSample> results (like the one below) in my result file. Anyone?
>
> Morten
>
>
> -----Original Message-----
> From: Peter Lin [mailto:woolfel@gmail.com]
> Sent: Mon 11/28/2005 5:40 PM
> To: JMeter Users List; Morten Kristiansen
> Subject: Re: XML output
>
> t is time
> it is iteration
> ts is timestamp
> s is success
> rc is response code
>
> hope that answers your questions
>
> peter
>
> On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
> >
> >
> > An example from an xml output from JMeter. What is "lt" here? I guess
> > "t" is time (or total time)?
> >
> >
> > <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> > rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
> >
> > Morten
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
> >
> >
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
RE: XML output
Posted by Morten Kristiansen <mo...@tracetracker.com>.
Hmm. The label says "lt" and not "it" doesn't it? And each element in the <httpSample> represents one request, so I don't see how "iteration" fits in this case. The number 329 says nothing to me. My test was 1 thread with 5 iterations and I got 5 <httpSample> results (like the one below) in my result file. Anyone?
Morten
-----Original Message-----
From: Peter Lin [mailto:woolfel@gmail.com]
Sent: Mon 11/28/2005 5:40 PM
To: JMeter Users List; Morten Kristiansen
Subject: Re: XML output
t is time
it is iteration
ts is timestamp
s is success
rc is response code
hope that answers your questions
peter
On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
>
>
> An example from an xml output from JMeter. What is "lt" here? I guess
> "t" is time (or total time)?
>
>
> <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
>
> Morten
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
Re: XML output
Posted by Peter Lin <wo...@gmail.com>.
t is time
it is iteration
ts is timestamp
s is success
rc is response code
hope that answers your questions
peter
On 11/28/05, Morten Kristiansen <mo...@tracetracker.com> wrote:
>
>
> An example from an xml output from JMeter. What is "lt" here? I guess
> "t" is time (or total time)?
>
>
> <httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
> rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
>
> Morten
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
XML output
Posted by Morten Kristiansen <mo...@tracetracker.com>.
An example from an xml output from JMeter. What is "lt" here? I guess
"t" is time (or total time)?
<httpSample t="608" lt="329" ts="1133195467745" s="true" lb="Request 1"
rc="200" rm="OK" tn="Thread 1 1-1" dt="text"/>
Morten
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: Testing the effect of page compression with Jmeter; Any gotchas?
Posted by sebb <se...@gmail.com>.
As to slower connections, the latest nightly builds include a means of
simulating slow connections for the Apache HttpClient sampler.
See jmeter.properties:
#httpclient.socket.http.cps=0
#httpclient.socket.https.cps=0
==
I looked at extending this to the default Http implementation, but it
is non-trivial, and requires overriding the boot classpath, as one has
to create a java.net class.
S.
On 28/11/05, Christensen, Alan <al...@hp.com> wrote:
>
> Has anyone used Jmeter to test the effects of using page compression vs
> not using page compression? Any gotchas that should be avoided, or best
> practices that will make this easier? What would you suggest be used to
> mimic slower connections?
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org