Re: Assertion Results Bug Fix 3

Jonathan, I have tried your changes here (only in non-GUI mode) and found no problems. In fact, your changes seem to resolve a problem where I was getting a large number of exceptions such as the following in jmeter.log: 12/28/2002 11:19:41 PM ERROR - jmeter.engine: java.lang.ArrayIndexOutOfBound

Assertion Results Bug Fix 3

Here is a much cleaner (IMHO) fix for the Assertion Results threading problem. Could someone else try this out and see how it works for them? It works fine for me in both GUI and non-GUI mode. This fix also avoids the problem that Scott found with the previous fix when running in non-GUI mode

Re: Assertion Results bug? -- here's the fix 2 - Perhaps not!

At the moment, I'm rather baffled. As I started adding debug statements, it stopped happening (in non gui-mode; it never happens in gui mode as far as I can tell). I kept removing debug statements until there were none left and it still didn't happen. One thing I did discover before it stopped

Re: Assertion Results bug? -- here's the fix 2 - Perhaps not!

Thanks Scott. I didn't even realize there was a non-GUI mode, but I see it now in the documentation (since I am looking for it, I guess). I am able to duplicate it in non-gui mode, but not in gui mode. I will do some further looking into it. Thanks, Jonathan >>> [EMAIL PROTECTED] 12/22/02

Re: Assertion Results bug? -- here's the fix 2

Hi Scott, > Can you please clarify the motivation for your query? Are you promoting a > "no patches without test cases" philosophy or is there something special > about this particular change that concerns you? I'm aware of reproducing thread related errors. I also trust Jonathan and you, my po

Re: Assertion Results bug? -- here's the fix 2 - Perhaps not!

Jonathan, Your fix may function well under some circumstances, but is does not seem to cater for non-gui testing where I end up with a jmeter.log full of these (I don't think I checked the log when I was running in gui mode): 12/22/2002 9:08:27 PM ERROR - jmeter.engine: java.lang.ClassCastExcept

Re: Assertion Results bug? -- here's the fix 2

Thanks again, Scott, for preparing the patch. To answer your question on how pinned down the problem was, I was able to reproduce it on a run of a dozen or so requests that had 10 threads running. Which request and which thread that it occurred on was variable for each run, although it did ha

Re: Assertion Results bug? -- here's the fix 2

Hi Michal, Short answer: No. Longer answer: a. in my experience issues relating to thread safety can sometimes be a little tricky to reproduce, let alone to reproduce consistently (I do note your use of 'almost unfailing'). I get the impression that Jonathan may in face have such an example, but

Re: Assertion Results bug? -- here's the fix 2

Hi, Thanks very much for the patch. Do you have (Jonathan, Scott, ... ?) an 'almost unfailig' method for reproducing this error? (test scenario?) best regards Michal Kostrzewa -- To unsubscribe, e-mail: For additional commands, e-mail:

Re: Assertion Results bug? -- here's the fix 2

I am using cvs to prepare the patch and am running my test script that is loaded with assertions to confirm correct behaviour. Cheers, Scott -- Scott Eade Backstage Technologies Pty. Ltd. http://www.backstagetech.com.au .Mac Chat/AIM: seade at mac dot com On 20/12/2002 9:56 AM, "Jonathan Carlso

Re: Assertion Results bug? -- here's the fix 2

Don't forget that my changes were based on 1.8 because that's all I could get for source (The nightly builds aren't getting put out there and I can't get to CVS). It may be safe to assume those methods didn't change in CVS after 1.8 but you should probably check that just to be sure if you haven

Re: Assertion Results bug? -- here's the fix 2

Jonathan, 1. The compiler and matcher definitions were private. 2. I think you forgot to include: Public final static String TEST_PATTERNS = "Assertion.test_patterns"; I'll post the patch to bugzilla shortly. Cheers, Scott -- Scott Eade Backstage Technologies Pty. Ltd. http://www.backsta

Re: Assertion Results bug? -- error message fix

One small change that got missed before: There is an area in the evaluateResponse method that looks like this: if (!pass) { result.setFailure(true); result.setFailureMessage( "Test Failed, expec

Re: Assertion Results bug? -- here's the fix 2

Here are all the changes again, but with thread protection added for Perl5Compiler and pre-compilation of all regex patterns. I included the full methods since it seemed easier to do and because I added javadoc where there was none. Any news on getting the nightly builds going again? 1) The c

Re: Assertion Results bug? -- here's the fix

Hi Scott, Can you wait? I have some other issues that I'll send you. The Perl5Compiler isn't thread-safe either. I'll send everything all together. >>> [EMAIL PROTECTED] 12/19/02 12:03AM >>> Is anyone making this into a patch? I can do it and submit it to bugzilla if you like. Let me know.

Re: Assertion Results bug? -- here's the fix

Is anyone making this into a patch? I can do it and submit it to bugzilla if you like. Let me know. Scott -- Scott Eade Backstage Technologies Pty. Ltd. http://www.backstagetech.com.au .Mac Chat/AIM: seade at mac dot com On 19/12/2002 5:15 AM, "Jonathan Carlson" <[EMAIL PROTECTED]> wrote: >

Re: Assertion Results bug? -- here's the fix

Success! The ORO Perl5Matcher is not thread-safe. I checked the mailing list and found some comments about Perl5Matcher needing to be accessed via a ThreadLocal instance in mult-threaded situations. This is what I did to ResponseAssertion to fix my problem: 1) I changed the "matcher" static f

Re: Assertion Results bug?

I take back question #1. The matcher and compiler are static fields, so they have to be shared with all the threads. Are we sure that the ORO Perl5Matcher and Perl5Compiler are thread-safe? >>> [EMAIL PROTECTED] 12/18/02 10:25AM >>> Here's what I have so far... I put a "breakpoint" into the Re

Re: Assertion Results bug?

Here's what I have so far... I put a "breakpoint" into the ResponseAssertion#evaluateResponse(...) method. It only breaks if the contains test does not pass. I then used BeanShell to create a new matcher and a new compiler to test the response string. The new matcher and pattern compiler find

Re: Assertion Results bug?

Done. If you have any suggestions for debugging this, I'd appreciate it. I can recreate the problem at will and can rebuild from the source if printlns are needed. Jonathan >>> [EMAIL PROTECTED] 12/17/02 04:48AM >>> Dnia pon 16. grudzien 2002 22:17, Jonathan Carlson napisal: > Thanks, I hope a

Re: Assertion Results bug?

Dnia pon 16. grudzień 2002 22:17, Jonathan Carlson napisał: > Thanks, I hope a file of 32K is ok. It is attached. Note that this is > just an example because every time I run it it fails on a different couple > of responses that have the pattern text but just doesn't find it. Thanks, but I didn'

Re: Assertion Results bug?

I have found what may be an important clue, but it doesn't make sense to me. Because I can not do "finds" in the ViewResultsTree I have to copy it into a Notepad file (yes, I have to use Windows) to do searches on it. When I tried saving it I got a message saying there was at least one Unicod

Re: Assertion Results bug?

Thanks, I hope a file of 32K is ok. It is attached. Note that this is just an example because every time I run it it fails on a different couple of responses that have the pattern text but just doesn't find it. The string that should have been found in the attached example was "Further narrow

Re: Assertion Results bug?

> I've been noticing that in certain circumstances the Assertion Results > doesn't find a string that is there. After looking into it a little deeper > it appears that the line where the string occurs is very very long. Is it > possible that it has a weakness in this case? Could you please send

Assertion Results bug?

I've been noticing that in certain circumstances the Assertion Results doesn't find a string that is there. After looking into it a little deeper it appears that the line where the string occurs is very very long. Is it possible that it has a weakness in this case? Jonathan Minneapolis, Minne