jmeter-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Philippe Mouawad <philippe.moua...@gmail.com>
Subject Re: getting error when loading a results file; how do I get more details?
Date Wed, 08 Mar 2017 20:43:53 GMT
Hello,
How does the corresponding Transaction Controller look like ? Which fields
did you check ?
Thanks

On Tue, Mar 7, 2017 at 10:57 PM, Scott Dean <psdean@ncsu.edu> wrote:

> In line
> >
> > 1487361334636,301358,Voucher nav/create/submit,,"Number of samples in
> > transaction : 23, number of failing samples : 0",,false,0,0,0,0,0,1,1,
> > NCSU202LRG11,0,0
>
>
> after "number of failing samples : 0", the next field is empty followed by
> false then a bunch of 0s.  The empty field should contain the thread name
> value which should  be something like "voucher fluid TG 4-14".  In addition
> the values after "false" should be numbers other than zero.
>
> Thanks
>
> -- Scott Dean
>
>
> On Tue, Mar 7, 2017 at 3:01 PM, Philippe Mouawad <
> philippe.mouawad@gmail.com
> > wrote:
>
> > Hi Scott,
> > What is missing in the lines below ?
> >
> > Can you show a "complete" line and an "incomplete" line from your point
> of
> > view ?
> >
> > Thank you
> > Regards
> >
> > On Tue, Mar 7, 2017 at 2:46 PM, Scott Dean <psdean@ncsu.edu> wrote:
> >
> > > My corruption occurred in the middle of the file and the test was not
> > > interrupted.  It ran continuously for the 3hr 36 minutes that it was
> > > scheduled for.  As for interrupting a test with Ctrl-C and restarting
> it
> > > using the same CSV results file, I would not be surprised to have
> > > corruption, and even expect it, under those conditions.
> > >
> > > Regarding the corruption, the line appears to be complete in that it
> has
> > > all the fields represented.  What is missing is the values as shown
> > below.
> > >   This is an excerpt from the results file with the two lines that
> > preceded
> > > it and the two lines that followed:
> > >
> > > 1487361635421,107,Portal display login page 240,200,"Number of samples
> in
> > > >> transaction : 9, number of failing samples : 0",background TG
> > > >> 1-10,true,98122,9884,80,185,0,1,0,NCSU202LRG11,3,0
> > > >
> > > > 1487361632433,3189,Voucher Navbar 415,200,"Number of samples in
> > > >> transaction : 23, number of failing samples : 0",voucher fluid TG
> > > >> 4-4,true,106343,114891,15,185,0,1,0,NCSU202LRG11,6,0
> > > >
> > > > 1487361334636,301358,Voucher nav/create/submit,,"Number of samples in
> > > >> transaction : 23, number of failing samples :
> > > >> 0",,false,0,0,0,0,0,1,1,NCSU202LRG11,0,0
> > > >
> > > > 1487361289503,321134,Voucher fluid package,,"Number of samples in
> > > >> transaction : 132, number of failing samples : 0",voucher fluid TG
> > > >> 4-14,false,413053,404133,15,185,133,1,1,NCSU202LRG11,25357,1
> > > >
> > > > 1487361334636,301352,Voucher Navbar 415,502,"Number of samples in
> > > >> transaction : 23, number of failing samples : 1",voucher fluid TG
> > > >> 4-14,false,93924,112116,15,185,0,1,1,NCSU202LRG11,6,0
> > > >
> > > >
> > > > All the lines came from the same remote server (NCSU202LRG11).
> > >
> > > Hope this helps!
> > >
> > > -- Scott Dean
> > >
> > >
> > > On Tue, Mar 7, 2017 at 2:46 AM, Felix Schumacher <
> > > felix.schumacher@internetallee.de> wrote:
> > >
> > > >
> > > >
> > > > Am 7. März 2017 05:03:58 MEZ schrieb Scott Dean <psdean@ncsu.edu>:
> > > > >I am using java 1.8.0_121 on Windows 2008r2 x64.
> > > > >
> > > > >Regarding the test plan, I need to scrub the plan of sensitive data
> > > > >before
> > > > >I can upload it.
> > > >
> > > > I have looked into my test results, that showed CSV corruption.
> > > >
> > > > The easiest and only way I could reproduce those was by stopping a
> > > running
> > > > test with Ctrl-c and restarting them using the same CSV file.
> > > >
> > > > The stopping could leave the last line incomplete and the next run
> > would
> > > > fill it with new results, leaving the line still corrupt.
> > > >
> > > > So on my end there was no concurrency issue, but rather a test end
> > issue.
> > > >
> > > > Could your corruption have the same origin?
> > > >
> > > > Felix
> > > >
> > > > >
> > > > >Just to be clear, the test plan was actually executed on four remote
> > > > >servers via the command line with the name of the results file
> > > > >specified as
> > > > >a commandline parameter and located on a network file share.  It is
> > not
> > > > >clear to me if each remote server instance wrote its own data to the
> > > > >results file or if the data was passed back to the jmeter master
> > > > >instance
> > > > >for the master to actually write to the results file.  In addition,
> I
> > > > >was
> > > > >loadtesting as part of an application upgrade project where we are
> > also
> > > > >moving from Solaris-based servers to linux-based servers.  I was
> > > > >loadtesting against upgraded application and my time windows for
> > > > >running
> > > > >tests has expired.  The system has wiped and rebuilt in preparation
> > for
> > > > >going live later this week.  I cannot go back and retest anything
at
> > > > >this
> > > > >point.
> > > > >
> > > > >Thanks!
> > > > >
> > > > >-- Scott Dean
> > > > >   NC State University
> > > > >
> > > > >On Sat, Mar 4, 2017 at 4:34 PM, Philippe Mouawad
> > > > ><philippe.mouawad@gmail.com
> > > > >> wrote:
> > > > >
> > > > >> Hi Again,
> > > > >> Could you also tell which JVM you're using and which version
?
> > > > >> Thank you
> > > > >> Regards
> > > > >>
> > > > >> On Sat, Mar 4, 2017 at 10:14 PM, Philippe Mouawad <
> > > > >> philippe.mouawad@gmail.com> wrote:
> > > > >>
> > > > >> > Hello,
> > > > >> > Would it be possible to upload your test plan to Bug 60818?
> > > > >> >
> > > > >> > We'd like to understand how this occured.
> > > > >> > It may be a bug , but it may also be due to your test plan.
> > > > >> >
> > > > >> > Regards
> > > > >> >
> > > > >> > On Thu, Mar 2, 2017 at 5:02 PM, Scott Dean <psdean@ncsu.edu>
> > wrote:
> > > > >> >
> > > > >> >> Found the problem.  It was a data error in the .csv
results
> file.
> > > > >One
> > > > >> of
> > > > >> >> the lines did not have a value for Group Name and this
caused
> the
> > > > >import
> > > > >> >> to
> > > > >> >> fail when there was no value to add to the append to
the label
> > > > >value.
> > > > >> No
> > > > >> >> clue why there was no thread group name for the record
or how
> to
> > > > >> reproduce
> > > > >> >> the bad result data.
> > > > >> >>
> > > > >> >> -- Scott Dean
> > > > >> >>    North Carolina State University
> > > > >> >>
> > > > >> >> On Wed, Mar 1, 2017 at 11:58 AM, Scott Dean <psdean@ncsu.edu>
> > > > >wrote:
> > > > >> >>
> > > > >> >> > I have run my tests in non-gui mode and generated
a results
> > > > >file.  I
> > > > >> am
> > > > >> >> > now trying to pull the results into a summary report
listener
> > > > >and
> > > > >> >> getting a
> > > > >> >> > error message in Result File Loader: "Error loading
result
> file
> > > > >-- see
> > > > >> >> log
> > > > >> >> > file".  When I look in the jmeter.log file, all
I see is
> "WARN
> > -
> > > > >> >> > jmeter.reporters.ResultCollector: Problem reading
JTL: <name
> > of
> > > > >the
> > > > >> >> > file>".  I am assuming that this is the log
file I am
> supposed
> > > > >to look
> > > > >> >> in.
> > > > >> >> > In addition, I get the error when "Include group
name in
> > label?"
> > > > >is
> > > > >> >> checked
> > > > >> >> > but I do not  when that setting is not checked.
> > > > >> >> >
> > > > >> >> >  Is there anything I can do to get more info about
the error,
> > > > >like a
> > > > >> >> > setting that will give more verbose log output?
 I am running
> > > > >jmeter
> > > > >> >> with a
> > > > >> >> > heap size of 2048m and the result file is 81.6
MB.
> > > > >> >> >
> > > > >> >> > Thanks
> > > > >> >> >
> > > > >> >> > -- Scott Dean
> > > > >> >> >    North Carolina State University
> > > > >> >> >    Raleigh NC.
> > > > >> >> >
> > > > >> >>
> > > > >> >
> > > > >> >
> > > > >> >
> > > > >> > --
> > > > >> > Cordialement.
> > > > >> > Philippe Mouawad.
> > > > >> >
> > > > >> >
> > > > >> >
> > > > >>
> > > > >>
> > > > >> --
> > > > >> Cordialement.
> > > > >> Philippe Mouawad.
> > > > >>
> > > >
> > > > ------------------------------------------------------------
> ---------
> > > > To unsubscribe, e-mail: user-unsubscribe@jmeter.apache.org
> > > > For additional commands, e-mail: user-help@jmeter.apache.org
> > > >
> > > >
> > >
> >
> >
> >
> > --
> > Cordialement.
> > Philippe Mouawad.
> >
>



-- 
Cordialement.
Philippe Mouawad.

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message