Netbeans JRuby debugging woes

[Zedisdead]

Hi

I have two diffrent machines one XP and one Vista and i have troubles running ruby and rails debugging on both. I've tried with both the boundled JRuby and i tried on netbeans 6.1, 6.5 beta, and the latest dayily alll with the same result.

Symptoms

XP: I get the 15 sec timeout problem. The XP machine works occasionally but i havn't been able to reproduce how. However uninstallling Java and Netbeans and then reinstalling them again seem to help a while. (Could it be a Gem ?)

Vista bombs right away from what i can see is that something i spits out an error message compling about unexpectd "\Netbeans " what i guess is a path

MRI debugging works on both.

 

Any ideas? Is there anyway of tracing whats going on ?

 

Regards

 

/Mikael

[Martin Krauskopf]

mikael wrote:
[...]

> Symptoms
>
> XP: I get the 15 sec timeout problem. The XP machine works occasionally
> but i havn't been able to reproduce how. However uninstallling Java and
> Netbeans and then reinstalling them again seem to help a while. (Could
> it be a Gem ?)
>
> Vista bombs right away from what i can see is that something i spits out
> an error message compling about unexpectd "\Netbeans " what i guess is a
> path

Hi Mark,

it would be strange then that it works once and not the other time.

> MRI debugging works on both.
>
> Any ideas? Is there anyway of tracing whats going on ?

There are not such bugs known at the moment. So if you would be able to
reproduce to get that timeout again with turned logging on, log as
described here:

http://wiki.netbeans.org/RubyDebugging#section-RubyDebugging-HowToFileABug

could help to reveal the culprit:

        m.


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Martin Krauskopf]

Martin Krauskopf wrote:

> mikael wrote:
> [...]
>> Symptoms
>>
>> XP: I get the 15 sec timeout problem. The XP machine works
>> occasionally but i havn't been able to reproduce how. However
>> uninstallling Java and Netbeans and then reinstalling them again seem
>> to help a while. (Could it be a Gem ?)
>>
>> Vista bombs right away from what i can see is that something i spits
>> out an error message compling about unexpectd "\Netbeans " what i
>> guess is a path
>
> Hi Mark,

Ehm. I meant "Hi Mikael". Race condition in my multitasking, I guess ;)
Sorry.

        m.


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Zedisdead]

Hi Martin,

I run the trace and i found somethings fishy in the logs it tries to debug
iconfileparser.rb which is correct but further down i see that i sends the
command to the debugger for a totaly diffrent file in a diffrent place
(main.rb)

Could it be cached somewhere ? Also another thing that is strange it says
that it kills the debugger process but the java instances still hangs
around when i look in the taskmanager.

Regards

/Mikael

FINARE [org.netbeans.api.ruby.platform.RubyPlatform]: Looking for
'rdebug-ide' executable; used intepreter: 'C:\Program Files\NetBeans 6.5
Beta\ruby2\jruby-1.1.3\bin\jruby.bat'
FINARE [org.netbeans.api.ruby.platform.RubyPlatform]: Found 'rdebug-ide':
'C:\Program Files\NetBeans 6.5 Beta\ruby2\jruby-1.1.3\bin\rdebug-ide'
FIN [org.rubyforge.debugcommons.Util]: Running [basedir:
C:\ws\betstonesites-main-mru\Code\IconFileParser\lib]: "C:\Program
Files\NetBeans 6.5 Beta\ruby2\jruby-1.1.3\bin\jruby.bat
-J-Djruby.reflection=true -J-Djruby.compile.mode=OFF
-IC:\ws\betstonesites-main-mru\Code\IconFileParser\lib
-IC:\ws\betstonesites-main-mru\Code\IconFileParser\test -r C:\Documents and
Settings\mru\Local Settings\Temp\io-synchronizer16294.rb C:\Program
Files\NetBeans 6.5 Beta\ruby2\jruby-1.1.3\bin\rdebug-ide _0.2.1_ -p 4703 -d
-- C:\ws\betstonesites-main-mru\Code\IconFileParser\lib\iconfileparser.rb"
FIN [org.rubyforge.debugcommons.Util]: Environment:
{USERPROFILE=C:\Documents and Settings\mru,
PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH,
USERDNSDOMAIN=COMBINATION.SE, JAVA_HOME=C:/Program Files/Java/jdk1.6.0_07,
COMMANDER_INI=C:\WINDOWS\WINCMD.INI, SystemDrive=C:,
TEMP=C:\DOCUME~1\mru\LOCALS~1\Temp, ProgramFiles=C:\Program Files,
Path=C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\Program
Files\Common Files\Roxio Shared\DLLShared\;C:\Program Files\Common
Files\Roxio Shared\9.0\DLLShared\;c:\Program Files\Microsoft SQL
Server\90\Tools\binn\;C:\Program Files\Perforce;C:\Program
Files\MySQL\MySQL Server 5.0\bin;C:\Program
Files\QuickTime\QTSystem\;C:/Program
Files/Java/jdk1.6.0_07\bin;C:/prog/jruby-1.1.3\bin,
JRUBY_HOME=C:/prog/jruby-1.1.3, HOMEDRIVE=C:, PROCESSOR_REVISION=1706,
CLIENTNAME=Console, =C:=C:\Program Files\NetBeans 6.5 Beta\bin,
USERDOMAIN=COMBINATION, QTJAVA=C:\Program
Files\Java\jre1.6.0_07\lib\ext\QTJava.zip, ALLUSERSPROFILE=C:\Documents and
Settings\All Users, VS90COMNTOOLS=C:\Program Files\Microsoft Visual Studio
9.0\Common7\Tools\, PROCESSOR_IDENTIFIER=x86 Family 6 Model 23 Stepping 6,
GenuineIntel, ULTRAMON_LANGDIR=C:\Program Files\UltraMon\Resources\en,
COMMANDER_PATH=C:\totalcmd, COMMANDER_DRIVE=C:, SESSIONNAME=Console,
TMP=C:\DOCUME~1\mru\LOCALS~1\Temp, PROCESSOR_ARCHITECTURE=x86,
LOGONSERVER=\\AQUILA, CommonProgramFiles=C:\Program Files\Common Files,
OS=Windows_NT, FP_NO_HOST_CHECK=NO, HOMEPATH=\Documents and Settings\mru,
PROMPT=$P$G, PROCESSOR_LEVEL=6, COMPUTERNAME=MRU-DESK1,
SystemRoot=C:\WINDOWS, windir=C:\WINDOWS, NUMBER_OF_PROCESSORS=2,
USERNAME=mru, ComSpec=C:\WINDOWS\system32\cmd.exe, APPDATA=C:\Documents and
Settings\mru\Application Data}
FIN [org.rubyforge.debugcommons.Util]: Sending command debugger: b
C:\Documents and Settings\mru\My
Documents\NetBeansProjects\test\lib\main.rb:6
FIN [org.rubyforge.debugcommons.Util]: Starting ReadersSupport readloop:
class org.rubyforge.debugcommons.ReadersSupport command loop
FIN [org.rubyforge.debugcommons.Util]: SocketException. Loop [class
org.rubyforge.debugcommons.ReadersSupport command loop]: Connection reset
FINARE [org.netbeans.modules.ruby.debugger.Util]: Performing "kill"
FINARE [org.netbeans.modules.ruby.debugger.Util]: Finishing session:
localhost:4703
FIN [org.rubyforge.debugcommons.Util]: Sending command debugger: exit



---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Martin Krauskopf]

mikael wrote:
> Hi Martin,
>
> I run the trace and i found somethings fishy in the logs it tries to debug
> iconfileparser.rb which is correct but further down i see that i sends the
> command to the debugger for a totaly diffrent file in a diffrent place
> (main.rb)

That's harmless. NetBeans sends all Ruby breakpoint listed in Breakpoint
view (A-S-F5 or Menu -> Windows -> Debugging -> Breakpoing).

> Could it be cached somewhere ? Also another thing that is strange it says
> that it kills the debugger process but the java instances still hangs
> around when i look in the taskmanager.

The main problem is that the Java client cannot connect to the Ruby
server. Is there anything written in the output window? Seems that
rdebug-ide did not start for some reason.

Might be JRuby has the problem with spaces in path. MRI works?
Does installing NetBeans in path without spaces helps?

I might try this later today when I have access to Windows.

        m.


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Zedisdead]

> Might be JRuby has the problem with spaces in path. MRI works?
> Does installing NetBeans in path without spaces helps?
No i didn't make any difference is there any way of connecting to the
hanging debug sessions ?

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Martin Krauskopf]

Mikael Rudberg wrote:
>> Might be JRuby has the problem with spaces in path. MRI works?
>> Does installing NetBeans in path without spaces helps?
> No i didn't make any difference is there any way of connecting to the
> hanging debug sessions ?

During those 15s you might try to telnet to the given port:

   telnet localhost <port_number>

To have more time, you might increase timeout to e.g. 5 minutes:

http://wiki.netbeans.org/RubyDebugging#section-RubyDebugging-TimeoutProblem

Also when you are running with detailed logging:

http://wiki.netbeans.org/FaqRubyNBLogging

You will see in the log how it is run. It is line containing rdebug-ide.
Actually could you try to run from command line:

C:/path/to/your/ruby /path/to/rdebug-ide -p 12345 --
/path/to/some/ruby_file.rb

for sanity check. Whether the process fails or not.

        m.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Zedisdead]

>
>    telnet localhost <port_number>

Yep this work, if i type anything in the telnet prompt i get the following
output
<error>There is no thread suspended at the time and therefore no context to
execute 'b 10'</error>
But it seems to run

> You will see in the log how it is run. It is line containing rdebug-ide.
> Actually could you try to run from command line:
>
> C:/path/to/your/ruby /path/to/rdebug-ide -p 12345

Yes this works as well i copied the actual line from the log

"C:\Program Files\NetBeans\ruby2\jruby-1.1.3\bin\jruby.bat"
-J-Djruby.reflection=true -J-Djruby.compile.mode=OFF -I"C:\Documents and
Settings\mru\My Documents\NetBeansProjects\test\lib" -I"C:\Documents and
Settings\mru\My Documents\NetBeansProjects\test\test" -I"C:\Documents and
Settings\mru\My Documents\NetBeansProjects\test\spec" -r "C:\Documents and
Settings\mru\Local Settings\Temp\io-synchronizer60142.rb" "C:\Program
Files\NetBeans\ruby2\jruby-1.1.3\bin\rdebug-ide" _0.2.1_ -p 3727 -d --
"C:\Documents and Settings\mru\My
Documents\NetBeansProjects\test\lib\main.rb"

I'm stumped considering these line in the log

FIN [org.rubyforge.debugcommons.Util]: Starting ReadersSupport readloop:
class org.rubyforge.debugcommons.ReadersSupport command loop
FIN [org.rubyforge.debugcommons.Util]: SocketException. Loop [class
org.rubyforge.debugcommons.ReadersSupport command loop]: Connection reset

It's like the IDE tries uses the wrong ip/port

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...

[Martin Krauskopf]

mikael wrote:
[...]
> I'm stumped considering these line in the log
>
> FIN [org.rubyforge.debugcommons.Util]: Starting ReadersSupport readloop:
> class org.rubyforge.debugcommons.ReadersSupport command loop
> FIN [org.rubyforge.debugcommons.Util]: SocketException. Loop [class
> org.rubyforge.debugcommons.ReadersSupport command loop]: Connection reset
>
> It's like the IDE tries uses the wrong ip/port

I've checked the code-flow and the IDE seems to successfully connect to
the backend. But then the connection is either terminated or data cannot
be transported for some reason.
Might also some strict firewall, if you have anything like this, might
be worth to check it.

I've sent you binary, containing more logging, in private email which
might tell us more.

        m.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@...
For additional commands, e-mail: users-help@...