curl-loader-devel

Re: Curl-loader results vary from time to time

[alo s. <asi...@ic...>]

Hi Robert,

Done some more analysis on it.

Error: (in one linux 64 system, it works fine. in another linux 64 
system, it throws the following error)
------
setup_curl_handle_appl - error: post_data is NULL.
setup_curl_handle_init - error: setup_curl_handle_appl () failed .

Cause: (when changing the parameters under #POST-part)
-----------------------------------------------------------
case 1: *works fine*
URL=""
URL_USE_CURRENT= 1

case 2: *throws error*
URL= http://xx.x.x.xxx:pppp/main
#URL_USE_CURRENT= 1

case 3: *throws error*
URL= http://xx.x.x.xxx:pppp/login
#URL_USE_CURRENT= 1


Conf file:
---------

# GET-part
URL= http://xx.x.x.xxx:pppp/main
URL_SHORT_NAME="user-GET"
#URL_DONT_CYCLE = 1
REQUEST_TYPE=GET
TIMER_URL_COMPLETION = 0 # In msec. Now it is enforced by cancelling url 
fetch on timeout
TIMER_AFTER_URL_SLEEP =0

# POST-part
URL=""
URL_USE_CURRENT= 1
URL_SHORT_NAME="user-POST"
#URL_DONT_CYCLE = 1
USERNAME=xxxxx
PASSWORD=xxxxx
REQUEST_TYPE=POST
FORM_USAGE_TYPE= SINGLE_USER
FORM_STRING= USERNAME=%s&PASSWORD=%s # Means the same credentials for 
all clients/users
TIMER_URL_COMPLETION = 0    # In msec. When positive, Now it is enforced 
by cancelling url fetch on timeout
TIMER_AFTER_URL_SLEEP =0

New Question:
---------------

Is there any way to test the load with only two or three different type 
of user ids and passwords?. For example, i want to test 1000 login 
attempts with three different user names and passwords. If there is a 
way, please render your advice..

Greatly appreciate your help.

Thanks,
Alo



Robert Iakobashvili wrote:
> Hi Alo,
>
> On Tue, Nov 11, 2008 at 4:13 PM, alo sinnathamby 
> <asi...@ic... <mailto:asi...@ic...>> wrote:
>
>     Hi Robert,
>
>     I wish to do that but i m not sure as how to do it. Anyway i will
>     try it
>     and if there is a success, i will let you know.
>
>     Thanks,
>     Alo
>
>  
> OK.
> If you can share the numbers of concurrent clients and CAPS, that you
> are getting to, it could be also of interest for the list.
>  
> Thanks,
> Robert

Re: Curl-loader results vary from time to time

[Gary F. <ga...@in...>]

Here's a suggestion. (I haven't tried this myself, and I'm no expert,  
but I think will work.) Use FORM_RECORDS_RANDOM = 1, and only provide  
three different records in the FORM_RECORDS_FILE. Ordinarily, there  
must be as many form records as clients, but with FORM_RECORDS_RANDOM,  
there may be any number of records, and they are assigned to clients  
at random. Alternatively, you could load the form records file with  
just three different sets of credentials, each repeated 333 times.


On Nov 12, 2008, at 10:59 AM, alo sinnathamby wrote:

> Is there any way to test the load with only two or three different  
> type
> of user ids and passwords?. For example, i want to test 1000 login
> attempts with three different user names and passwords. If there is a
> way, please render your advice..

Re: Curl-loader results vary from time to time

[Robert I. <cor...@gm...>]

Hi folks,

On Wed, Nov 12, 2008 at 9:30 PM, Gary Fitts <ga...@in...> wrote:

> Here's a suggestion. (I haven't tried this myself, and I'm no expert,
> but I think will work.) Use FORM_RECORDS_RANDOM = 1, and only provide
> three different records in the FORM_RECORDS_FILE. Ordinarily, there
> must be as many form records as clients, but with FORM_RECORDS_RANDOM,
> there may be any number of records, and they are assigned to clients
> at random. Alternatively, you could load the form records file with
> just three different sets of credentials, each repeated 333 times.
>
>
> On Nov 12, 2008, at 10:59 AM, alo sinnathamby wrote:
>
> > Is there any way to test the load with only two or three different
> > type
> > of user ids and passwords?. For example, i want to test 1000 login
> > attempts with three different user names and passwords. If there is a
> > way, please render your advice..
>
>
Thanks, Gary, very good explanation.

-- 
Truly,
Robert Iakobashvili, Ph.D.
......................................................................
www.ghotit.com
Assistive technology that understands you
......................................................................

Re: problem when using RECORDS_FROM_FILE

[Gary F. <ga...@in...>]

Yes, this seems to resolve it.
I'm still planning to submit this and other
changes soon.

Thanks,
Gary


On Nov 11, 2008, at 9:46 AM, Robert Iakobashvili wrote:

> Hi Matt,
>
> On Tue, Nov 4, 2008 at 6:45 PM, Gary Fitts <ga...@in...>  
> wrote:
> Matt,
>
> I've recently started using curl-loader, and I ran into this too.
> I think we're extending the use of FORM_STRING beyond what was
> originally intended. I fixed this in the source file parse_config.c
> by changing the line
>
> if (ftype != FORM_USAGETYPE_AS_IS)
>
> to
>
> if (ftype != FORM_USAGETYPE_AS_IS && ftype !=
> FORM_USAGETYPE_RECORDS_FROM_FILE)
>
> Maybe you could give this a try. It seems to work for me, although
> I'm still testing this along with a few other changes.
>
> Gary Fitts
>
> Is the issue solved, or you need me to investigate it?
>
> Sincerely,
> Robert
> <ATT00001.txt><ATT00002.txt>

Re: problem when using RECORDS_FROM_FILE

[Robert I. <cor...@gm...>]

On Tue, Nov 11, 2008 at 8:00 PM, Gary Fitts <ga...@in...> wrote:

> Yes, this seems to resolve it.
> I'm still planning to submit this and other
> changes soon.
>
> Thanks,
> Gary
>

Thank you very much!

-- 
Truly,
Robert Iakobashvili, Ph.D.
......................................................................
www.ghotit.com
Assistive technology that understands you
......................................................................

building urls from server responses

[Gary F. <Ga...@in...>]

Oops, I forgot to include one file..


Hello Robert,

Here are the additions to curl-loader that I've been working on. The  
new tags are documented in an attached file. They allow clients to  
fetch urls built from prior server responses, or from a token file.

There are also a few bug fixes (at least I think they were bugs):  
multiple clients can now upload the same file repeatedly without  
interference, form_strings can now have many %s markers, and  
whitespace is removed from form_string tokens. (The code always  
removed whitespace, but a minor bug caused the original token to be  
used instead.)

I tried to keep all of my changes in one new source file, url_set.c,  
but I had to put hooks and small changes into 5 other files: url.h,  
client.h, parse_conf.c, loader.c, and loader_fsm.c. These are all  
included. I've marked all the changes with GF in a comment to make  
them easy to find.

I've tested all this to my own satisfaction, but of course there will  
be bugs that I haven't uncovered yet.

If you want to adopt any of these changes, it might make sense to move  
the url_set.c code back into parse_conf.c and loader.c -- whatever is  
most convenient.

Gary Fitts

Re: building urls from server responses

[Robert I. <cor...@gm...>]

Hi Gary,

On Wed, Nov 12, 2008 at 11:51 PM, Gary Fitts <Ga...@in...> wrote:

> Oops, I forgot to include one file..
>
>
>
> Hello Robert,
>
> Here are the additions to curl-loader that I've been working on. The new
> tags are documented in an attached file. They allow clients to fetch urls
> built from prior server responses, or from a token file.
>
> There are also a few bug fixes (at least I think they were bugs): multiple
> clients can now upload the same file repeatedly without interference,
> form_strings can now have many %s markers, and whitespace is removed from
> form_string tokens. (The code always removed whitespace, but a minor bug
> caused the original token to be used instead.)
>
> I tried to keep all of my changes in one new source file, url_set.c, but I
> had to put hooks and small changes into 5 other files: url.h, client.h,
> parse_conf.c, loader.c, and loader_fsm.c. These are all included. I've
> marked all the changes with GF in a comment to make them easy to find.
>
> I've tested all this to my own satisfaction, but of course there will be
> bugs that I haven't uncovered yet.
>
> If you want to adopt any of these changes, it might make sense to move the
> url_set.c code back into parse_conf.c and loader.c -- whatever is most
> convenient.
>
> Gary Fitts
>


Thank you very much for your contribution.
I will look at it next week.

Sounds great!


-- 
Truly,
Robert Iakobashvili, Ph.D.
......................................................................
Assistive technology that understands you
......................................................................

RE: problem when using RECORDS_FROM_FILE

[Matt L. <mat...@av...>]

Robert,

 

Yes, this issue was resolved by using Gary's workaround below.  

 

Thanks,

Matt Love

 

From: Robert Iakobashvili [mailto:cor...@gm...] 
Sent: Tuesday, November 11, 2008 6:47 PM
To: curl-loader-devel
Subject: Re: problem when using RECORDS_FROM_FILE

 

Hi Matt,

On Tue, Nov 4, 2008 at 6:45 PM, Gary Fitts <ga...@in...>
wrote:

Matt,

I've recently started using curl-loader, and I ran into this too.
I think we're extending the use of FORM_STRING beyond what was
originally intended. I fixed this in the source file parse_config.c
by changing the line

if (ftype != FORM_USAGETYPE_AS_IS)

to

if (ftype != FORM_USAGETYPE_AS_IS && ftype !=
FORM_USAGETYPE_RECORDS_FROM_FILE)

Maybe you could give this a try. It seems to work for me, although
I'm still testing this along with a few other changes.

Gary Fitts

 

Is the issue solved, or you need me to investigate it?

 

Sincerely,

Robert

Re: problem when using RECORDS_FROM_FILE

[Robert I. <cor...@gm...>]

Hi Matt,

On Thu, Nov 13, 2008 at 5:38 PM, Matt Love <mat...@av...> wrote:

>  Robert,
>
>
>
> Yes, this issue was resolved by using Gary's workaround below.
>
>
>
Thanks, I will commit the fix.
Thank, Gary!



-- 
Truly,
Robert Iakobashvili, Ph.D.
......................................................................
Assistive technology that understands you
......................................................................

Curl-loader - login with three different user ids and passwords

[alo s. <asi...@ic...>]

Hi Robert,

Thanks for your reply and Greatly appreciate it. You suggestion worked 
fine and this is how i used it.

URL= http://xx.x.x.xxx:pppp/main
URL_SHORT_NAME="login-GET"
REQUEST_TYPE=GET
TIMER_URL_COMPLETION = 0 # In msec. Now it is enforced by cancelling url 
fetch on timeout
TIMER_AFTER_URL_SLEEP =0

URL=""
URL_USE_CURRENT= 1
URL_SHORT_NAME="login-POST"
REQUEST_TYPE=POST
FORM_USAGE_TYPE= RECORDS_FROM_FILE
FORM_STRING= USERNAME=%s&PASSWORD=%s # Means the same credentials for 
all clients/users
FORM_RECORDS_RANDOM=1
FORM_RECORDS_FILE_MAX_NUM=3
FORM_RECORDS_FILE= ./conf-stagetest/credentials.cred # Path relative to 
curl-loader location
TIMER_URL_COMPLETION = 0    # In msec. When positive, Now it is enforced 
by cancelling url fetch on timeout
TIMER_AFTER_URL_SLEEP =0

i have provided three different records in the credentials.cred file and 
it worked fine.

New Question:
---------------
When i look at the results summary at the console, there are two columns 
under success. What does the first column mean?. For example in the 
given below results, what do 89 and 70 mean?

Test total duration was 15 seconds and CAPS average 13:
H/F   
Req:728,1xx:0,2xx:667,3xx:0,4xx:0,5xx:0,Err:0,T-Err:0,D:1195ms,D-2xx:1195ms,Ti:379752B/s,To:15756B/s
H/F/S 
Req:0,1xx:0,2xx:0,3xx:0,4xx:0,5xx:0,Err:0,T-Err:0,D:0ms,D-2xx:0ms,Ti:0B/s,To:0B/s
 Operations:             Success                 Failed                 
Timed out
URL0:login-GET          89     250              0      0                
0      0      
URL1:login-POST         70     215              0      0                
0      0      


Thanks and Appreciate your help.

Thanks,
Alo



Robert Iakobashvili wrote:
> Hi folks,
>
> On Wed, Nov 12, 2008 at 9:30 PM, Gary Fitts <ga...@in... 
> <mailto:ga...@in...>> wrote:
>
>     Here's a suggestion. (I haven't tried this myself, and I'm no expert,
>     but I think will work.) Use FORM_RECORDS_RANDOM = 1, and only provide
>     three different records in the FORM_RECORDS_FILE. Ordinarily, there
>     must be as many form records as clients, but with FORM_RECORDS_RANDOM,
>     there may be any number of records, and they are assigned to clients
>     at random. Alternatively, you could load the form records file with
>     just three different sets of credentials, each repeated 333 times.
>
>
>     On Nov 12, 2008, at 10:59 AM, alo sinnathamby wrote:
>
>     > Is there any way to test the load with only two or three different
>     > type
>     > of user ids and passwords?. For example, i want to test 1000 login
>     > attempts with three different user names and passwords. If there
>     is a
>     > way, please render your advice..
>
>
> Thanks, Gary, very good explanation.
>
> -- 
> Truly,
> Robert Iakobashvili, Ph.D.
> ......................................................................
> www.ghotit.com <http://www.ghotit.com>
> Assistive technology that understands you
> ......................................................................

NOTE: This message, and any attached files, contain information that is privileged, confidential, proprietary or otherwise protected from disclosure.  Any disclosure, copying or distribution of, or reliance upon, this message by anyone else is strictly prohibited.  If you received this communication in error, please notify the sender immediately by reply e-mail message or by telephone to one of the numbers above and deleting it from your computer.  Thank you.

Re: Curl-loader - login with three different user ids and passwords

[Robert I. <cor...@gm...>]

Hi alo

On Mon, Nov 17, 2008 at 6:19 PM, alo sinnathamby <asi...@ic...>wrote:

> Hi Robert,
>
> Thanks for your reply and Greatly appreciate it. You suggestion worked
> fine and this is how i used it.
>


Re-addressing your thanks to Gary, who was the man,
who actually advised you.



>
>
>
> New Question:
> ---------------
> When i look at the results summary at the console, there are two columns
> under success. What does the first column mean?. For example in the
> given below results, what do 89 and 70 mean?
>
> Test total duration was 15 seconds and CAPS average 13:
> H/F
>
> Req:728,1xx:0,2xx:667,3xx:0,4xx:0,5xx:0,Err:0,T-Err:0,D:1195ms,D-2xx:1195ms,Ti:379752B/s,To:15756B/s
> H/F/S
>
> Req:0,1xx:0,2xx:0,3xx:0,4xx:0,5xx:0,Err:0,T-Err:0,D:0ms,D-2xx:0ms,Ti:0B/s,To:0B/s
>  Operations:             Success                 Failed
> Timed out
> URL0:login-GET          89     250              0      0
> 0      0
> URL1:login-POST         70     215              0      0
> 0      0
>
> The above questions are FAQs:
http://curl-loader.sourceforge.net/doc/faq.html

Sincerely,
Robert

Re: building urls from server responses

[Robert I. <cor...@gm...>]

Hi Gary,

On Wed, Nov 12, 2008 at 11:51 PM, Gary Fitts <Ga...@in...> wrote:

> Oops, I forgot to include one file..
>
>
>
> Hello Robert,
>
> Here are the additions to curl-loader that I've been working on. The new
> tags are documented in an attached file. They allow clients to fetch urls
> built from prior server responses, or from a token file.
>
> There are also a few bug fixes (at least I think they were bugs): multiple
> clients can now upload the same file repeatedly without interference,
> form_strings can now have many %s markers, and whitespace is removed from
> form_string tokens. (The code always removed whitespace, but a minor bug
> caused the original token to be used instead.)
>
> I tried to keep all of my changes in one new source file, url_set.c, but I
> had to put hooks and small changes into 5 other files: url.h, client.h,
> parse_conf.c, loader.c, and loader_fsm.c. These are all included. I've
> marked all the changes with GF in a comment to make them easy to find.
>
> I've tested all this to my own satisfaction, but of course there will be
> bugs that I haven't uncovered yet.
>
> If you want to adopt any of these changes, it might make sense to move the
> url_set.c code back into parse_conf.c and loader.c -- whatever is most
> convenient.
>
> Gary Fitts
>

Thank you for your great contribution to the project.
I liked your design decisions.
Confident, that many people will enjoy the features, that you developed.

It was integrated to the existing files deliberately without serious
changes.
Could you, please, checkout from svn the latest version and run
as much tests in your conditions to see, that nothing was broken upon
integration?
Thank you in advance.

We will still need to polish the code, bring coding conventions and
comments,
 README, FAQs and man pages. Later I will submit my suggestions.

Once more, thank you and my appreciations for the good job!

Sincerely,
Robert Iakobashvili, Ph.D.
......................................................................
www.ghotit.com
Assistive technology that understands you
......................................................................

Re: building urls from server responses

[Gary F. <ga...@in...>]

Unfortunately I'm not very familiar with svn. I checked out the latest  
version with
svn co https://curl-loader.svn.sourceforge.net/svnroot/curl-loader  
curl-loader

Then I did this
cd curl-loader/trunk/curl-loader
make

The make run ended in this error
/bin/sh: ../../packages/curl/configure: No such file or directory

There are lots of references in the Makefile to "packages" directories,
but no such directory came with the checkout. Undoubtedly I'm doing  
something wrong.


On Nov 20, 2008, at 9:11 AM, Robert Iakobashvili wrote:

> Thank you for your great contribution to the project.
> I liked your design decisions.
> Confident, that many people will enjoy the features, that you  
> developed.
>
> It was integrated to the existing files deliberately without serious  
> changes.
> Could you, please, checkout from svn the latest version and run
> as much tests in your conditions to see, that nothing was broken  
> upon integration?
> Thank you in advance.
>
> We will still need to polish the code, bring coding conventions and  
> comments,
>  README, FAQs and man pages. Later I will submit my suggestions.
>
> Once more, thank you and my appreciations for the good job!
>
> Sincerely,
> Robert Iakobashvili, Ph.D.
> ......................................................................
> www.ghotit.com
> Assistive technology that understands you
> ......................................................................
> <ATT00001.txt><ATT00002.txt>

Re: building urls from server responses

[Robert I. <cor...@gm...>]

Hi Gary,

On Thu, Nov 20, 2008 at 7:47 PM, Gary Fitts <ga...@in...> wrote:

> Unfortunately I'm not very familiar with svn. I checked out the latest
> version with
> svn co https://curl-loader.svn.sourceforge.net/svnroot/curl-loader
> curl-loader
>
> Then I did this
> cd curl-loader/trunk/curl-loader
> make
>
> The make run ended in this error
> /bin/sh: ../../packages/curl/configure: No such file or directory
>
> There are lots of references in the Makefile to "packages" directories,
> but no such directory came with the checkout. Undoubtedly I'm doing
> something wrong.
>

It may be some problem with firewall or etc devices. One more
issue can be not enough space at filesystem.

Clean all and try once more:

$svn co https://curl-loader.svn.sourceforge.net/svnroot/curl-loadercurl-loader

$cd curl-loader/trunk/curl-loader
$ls -alF
to observe packages directory
$make

Take care,

Truly,
Robert Iakobashvili, Ph.D.
......................................................................
www.ghotit.com
Assistive technology that understands you
......................................................................

Re: building urls from server responses

[Gary F. <ga...@in...>]

Thanks! It worked this time.
I'll test as soon as I get a chance.


On Nov 20, 2008, at 9:57 AM, Robert Iakobashvili wrote:

> It may be some problem with firewall or etc devices. One more
> issue can be not enough space at filesystem.
>
> Clean all and try once more:
>
> $svn co https://curl-loader.svn.sourceforge.net/svnroot/curl-loader  
> curl-loader
>
> $cd curl-loader/trunk/curl-loader
> $ls -alF
> to observe packages directory
> $make
>
> Take care,
>
> Truly,
> Robert Iakobashvili, Ph.D.

Re: building urls from server responses

[Gary F. <ga...@in...>]

Everything tests out just fine, Robert.

One question: you put most of the new code (that was in a separate  
file url_set.c)
into parse_conf.c. A lot of that is "run time" code that is executed  
while the load
is running, while parse_conf.c used to contain only parse-time code.  
Is that OK?
Or do you want to put run-time code into loader.c or a separate file?  
Either is
fine -- depends on how you want to organize it.

Since I submitted these changes I've found the need for a couple more  
tags.
I'll send those along in a day or so when I've been able to test them.


On Nov 20, 2008, at 9:11 AM, Robert Iakobashvili wrote:

> Thank you for your great contribution to the project.
> I liked your design decisions.
> Confident, that many people will enjoy the features, that you  
> developed.
>
> It was integrated to the existing files deliberately without serious  
> changes.
> Could you, please, checkout from svn the latest version and run
> as much tests in your conditions to see, that nothing was broken  
> upon integration?
> Thank you in advance.
>
> We will still need to polish the code, bring coding conventions and  
> comments,
>  README, FAQs and man pages. Later I will submit my suggestions.
>
> Once more, thank you and my appreciations for the good job!

Re: building urls from server responses

[Gary F. <ga...@in...>]

Hello Robert,

Here is the latest (hopefully last) patch file, except for bug fixes.
And everything else -- the README and the sample files -- looks OK.

I've added a bit to the README to explain two new tags that I'm using
(ERROR_TOKEN and PRINT_TIMING_STATS) in case others want to use them  
too.

Gary

Re: building urls from server responses

[Robert I. <cor...@gm...>]

Hi Gary,

On Mon, Dec 1, 2008 at 10:11 PM, Gary Fitts <ga...@in...> wrote:

> Hello Robert,
>
> Here is the latest (hopefully last) patch file, except for bug fixes.
> And everything else -- the README and the sample files -- looks OK.
>
> I've added a bit to the README to explain two new tags that I'm using
> (ERROR_TOKEN and PRINT_TIMING_STATS) in case others want to use them too.
>
> Gary
>
>

Thank you very much.

I will be doing release very soon, therefore, the new tags
will be added after release. We are also postponing code improvements, etc,
since this release will incorporate a number of good bug-fixes, that are
better
to release ASAP.


-- 
Truly,
Robert Iakobashvili, Ph.D.
......................................................................
www.ghotit.com
Assistive technology that understands you
......................................................................

Re: Curl-loader results vary from time to time

[alo s. <asi...@ic...>]

Hi Robert,

Thanks for your reply.

We have two test clients and the environment in both machines are same 
except the client 2 has 8 Gig RAM and client 1 has 4 Gig RAM. The test 
script works fine with test client 1 and it gives the following error 
with test client 2.

setup_curl_handle_appl - error: post_data is NULL.
setup_curl_handle_init - error: setup_curl_handle_appl () failed .
setup_url error: setup_curl_handle_init - failed
setup_curl_handle_appl - error: post_data is NULL.
setup_curl_handle_init - error: setup_curl_handle_appl () failed .
setup_url error: setup_curl_handle_init - failed
setup_curl_handle_appl - error: post_data is NULL.
setup_curl_handle_init - error: setup_curl_handle_appl () failed .
setup_url error: setup_curl_handle_init - failed
setup_curl_handle_appl - error: post_data is NULL.
setup_curl_handle_init - error: setup_curl_handle_appl () failed .


This is my test script:
--------------------
########### GENERAL SECTION ################################
BATCH_NAME=NOV10_SA_login
CLIENTS_NUM_MAX = 10000
CLIENTS_NUM_START=50
CLIENTS_RAMPUP_INC=50
INTERFACE=eth0   
NETMASK=24 
IP_ADDR_MIN=xx.x.x.xxx
IP_ADDR_MAX=xx.x.x.xxx
CYCLES_NUM= -1
URLS_NUM=2

########### URL SECTION ##################################

# GET-part
URL=http://xx.x.x.xxx/main
URL_SHORT_NAME="Login-GET"
#URL_DONT_CYCLE = 1
REQUEST_TYPE=GET
TIMER_URL_COMPLETION = 0 # In msec. Now it is enforced by cancelling url 
fetch on timeout
TIMER_AFTER_URL_SLEEP =0

# POST-part
URL=http://xx.x.x.xxx/login
#URL_USE_CURRENT= 1
URL_SHORT_NAME="Login-POST"
#URL_DONT_CYCLE = 1
USERNAME=test
PASSWORD=test
REQUEST_TYPE=POST
FORM_USAGE_TYPE= SINGLE_USER
FORM_STRING= USERNAME=%s&PASSWORD=%s # Means the same credentials for 
all clients/users
TIMER_URL_COMPLETION = 0    # In msec. When positive, Now it is enforced 
by cancelling url fetch on timeout
TIMER_AFTER_URL_SLEEP =0


These are my client machines:
-----------------------------

Test client 1:
Operating System:    CentOS release 5.2 (Final)
CPU:    Intel(R) Core(TM)2 Duo CPU     E4600  @ 2.40GHz
Ram:    4 Gigabytes DDR2

Test client 2:
Operating System:    CentOS release 5.2 (Final)
CPU:    Intel(R) Core(TM)2 Duo CPU     E4600  @ 2.40GHz
Ram:    8 Gigabytes DDR2


I have also followed the recommendations given in the following links,

section 7.2 in http://curl-loader.sourceforge.net/doc/faq.html
http://curl-loader.sourceforge.net/high-load-hw/index.html


If you have any clue as why we get this type of error in only one 
client, please advice us and Greatly appreciate your help.

Thanks,
Alo






Robert Iakobashvili wrote:
> Hi Alo,
>
> On Tue, Nov 4, 2008 at 4:18 PM, alo sinnathamby 
> <asi...@ic... <mailto:asi...@ic...>> wrote:
>
>     Hi Robert,
>
>     Because of these different results, i am not able to determine the
>     correct run time for any given number of requests. Our objective is to
>     determine the load capability and run time for any given number of
>     requests, say for an example 20000, 50000, 70000, 100000 requests.
>
>     Please render your valuable advice and Greatly appreciate your help.
>
>      
>
>  
> The suggestion is always to measure such things not at a ramp-up
> time (75 seems to be the max for your machine), but at a steady state.
>  
> Y may wish to look at the steady run-time slot and to adjust
> the number of cycles accordingly.
> Besides that, you can see CAPS number, which is at a steady stage
> the number of cycles, that all your virtual clients are doing each second.
>  
> Besides that, the following conversation may be of your interest:
> http://sourceforge.net/mailarchive/forum.php?thread_name=BAY131-W4271C5395CD9FFBBC04BF8D9C80%40phx.gbl&forum_name=curl-loader-devel 
> <http://sourceforge.net/mailarchive/forum.php?thread_name=BAY131-W4271C5395CD9FFBBC04BF8D9C80%40phx.gbl&forum_name=curl-loader-devel>
>  
> Sincerely,
> Robert