Problem compiling Omnet++ on Windows 7

206 views
Skip to first unread message

Petko Pedev

unread,
Jul 19, 2015, 12:08:53 PM7/19/15
to omn...@googlegroups.com
Greetings,

I've been trying to install Omnet++ on my computer for academical purposes. It is a Windows 7 Toshiba laptop with 2 GB of RAM and a 1.60  GHz 32 bit processor.

When I start compiling all the files with Mingw I get to the point on the screenshot, and then the compiling process stops and the LED of my hard drive becomes solid green. After that I have to restart the PC, so that I can fix the issue with the LED. This happens with version 5.0b1 of Omnet++. I get the same result with the latest stable version - 4.6 of Omnet++. What seems to be the problem? Any help will be greatly appreciated.

Best Regards,

Petko Pedev
error message.jpg

Michael Kirsche

unread,
Jul 20, 2015, 7:13:16 AM7/20/15
to omn...@googlegroups.com
Any virus scanners working in the background maybe?
Apart from that, the compilation of OMNeT does take some time, especially on slower machines like yours. Any signs of a real "hang up" of your machine in the Task Manager (like a constant increase in memory usage of the GCC process while the compilation progress in MinGW halts)?

Petko Pedev

unread,
Jul 20, 2015, 10:04:33 AM7/20/15
to omn...@googlegroups.com
Well, I am using AVG Antivirus Free Edition 2015 and whether I disable the protection or not, I still get to the same point. I will try to run it again with a disabled protection and AVG program turned off, and will see what will happen.

Michael Kirsche

unread,
Jul 20, 2015, 4:55:50 PM7/20/15
to omn...@googlegroups.com
Don't forget to run 'make cleanall' and then again './configure' and 'make' when you've disabled your virus scanner.

Petko Pedev

unread,
Jul 23, 2015, 12:14:58 PM7/23/15
to OMNeT++ Users, petko...@gmail.com
Hello,

After I disable my antivirus and write the commands "make cleanall" and etc. I still get the same result. I guess it is not the  antivirus. Waiting for a reply.

Regards,

Michael Kirsche

unread,
Jul 24, 2015, 3:42:47 AM7/24/15
to OMNeT++ Users, petko...@gmail.com
Very strange. Sometimes, the downloaded archive gets corrupted while downloading, that's when random errors happen during the compilation process. But I've never heard of the compilation hanging / running into a loop.
Two options:
(1) download the ZIP archive again, unpack it (which unpacker are you using), start the MinGW prompt and ./configure and make it again.
if it doesn't work, try it on some other machine or
(2) use a virtual machine with a Linux-based system and use the Linux version of OMNeT.

You could also try to compile the 4.6 version, if the problem is only with the newer beta release, maybe some of the MinGW tools are running into problems on your PC.

Petko Pedev

unread,
Jul 25, 2015, 3:53:47 PM7/25/15
to OMNeT++ Users, michael...@b-tu.de
Greetings Michael Kirsche,

So, by the time I read your last answer, the 5.01b was not on the site any more, and I had to download 5.02b. I was using Winrar as an unpacker. I removed it and tried without it. I got the same result. Then I tried with Winzip, and again, after compiling got the same result. The reason I try with these beta versions is, because it happened before with the latest stable one, which is 4.6. For option two, it works indeed on a virtual machine with Ubuntu OS on it. But the thing is that virtual machine is on my workstation where I work. And I want to use Omnet++ at home, and not staying after work everyday. I guess it is either my Windows missing something in order to complete the compiling process, or it is too slow and needs some more time... I tried with version 4.6 on another virtual Windows 8.1 machine and it worked fine, so I guess the only option is that I am missing some packages. Of course the processor of the physical machine running that Windows 8.1 virtual machine is more powerful than mine at home, but still... 
Thank you for helping me. :)

Rudolf Hornig

unread,
Jul 27, 2015, 5:12:27 AM7/27/15
to OMNeT++ Users, petko...@gmail.com, michael...@b-tu.de, petko...@gmail.com
My two cents: I suspect that the issue is with the low amount of RAM on the machine. The screenshot in the first message suggest that the whole compilation process stops at the linking phase which is the most memory intensive one. If the opsystem runs out of the physical it starts to swap out existing memory areas causing the effects you describe (your harddisk starts running all time and everything seems to stop). When paging kicks in, the performance of the whole process drops by several magnitudes rendering the whole thing essentially useless).

Long story short, I think you will need a machine with more memory.
Rudolf

Alfonso Ariza Quintana

unread,
Jul 27, 2015, 6:01:21 AM7/27/15
to omn...@googlegroups.com

Java is acting up “An internal error occurred during: "Update Monitor".

Java heap space”

Error.log

Exception in thread "qtp11352311-92" Exception in thread "qtp11352311-87" Exception in thread "qtp11352311-85" java.lang.OutOfMemoryError: Java heap space

java.lang.OutOfMemoryError: Java heap space

Job found still running after platform shutdown.  Jobs should be canceled by the plugin that scheduled them during shutdown: org.eclipse.cdt.internal.core.pdom.PDOMIndexerJob

Job found still running after platform shutdown.  Jobs should be canceled by the plugin that scheduled them during shutdown: org.eclipse.cdt.internal.core.pdom.PDOMIndexerJob$ProgressUpdateJob

 

Exception: java.lang.OutOfMemoryError thrown from the UncaughtExceptionHandler in thread "qtp11352311-92"

 

Exception: java.lang.OutOfMemoryError thrown from the UncaughtExceptionHandler in thread "[Timer] - Main Queue Handler"

 

 

I have included the other log files in the email

--
You received this message because you are subscribed to the Google Groups "OMNeT++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to omnetpp+u...@googlegroups.com.
Visit this group at http://groups.google.com/group/omnetpp.
For more options, visit https://groups.google.com/d/optout.

log.7z

Petko Pedev

unread,
Jul 30, 2015, 3:11:07 PM7/30/15
to OMNeT++ Users, michael...@b-tu.de, rudolf...@gmail.com
Hello Rudolf Hornig,

Here is what I did so far. I tried on another laptop, which is running with 4 BG of RAM, and a faster Pentium processor. It managed to compile itself and I succeeded in running the Omnet++. After that I decided I should give it a try on a clean Windows 7 machine. I made a virtual machine on another physical workstation, using VMWare Virtual Machine software, simulating a single core processor and running 2 GB of RAM, and surprisingly the Omnet compiled and I could run it with no problems. So I guess the problem is either my processor, or some software package missing. I am still investigating it. Are there any log files which I can search for and post them here, so that everybody can take a look at them and give an opinion?

Also is it a problem that I am compiling Omnet++ directly in the Desktop folder? Although it was said in the manual that no matter where the directory should not contain empty spaces in its name, or any other strange characters, and the path to the desktop complies with these rules, but still...

Regards,

Petko Pedev

Michael Kirsche

unread,
Jul 31, 2015, 5:12:28 AM7/31/15
to OMNeT++ Users, petko...@gmail.com
Hello Petko,

software packages can't be missing, since the Windows installation comes in an all-inclusive download, everything you need is included in there.
Like Rudolf wrote, if you have a small amount of memory, the hard drive is used for caching. Even if you managed to compile it on a fresh Win7 VM with 2GB of RAM, that doesn't mean that it must work flawlessly on your machine, as it might have less than 2GB of free RAM available due to other programs running.
Did you run the VM on your laptop?

Best,
Michael

Rudolf Hornig

unread,
Jul 31, 2015, 5:23:09 AM7/31/15
to OMNeT++ Users, petko...@gmail.com, michael...@b-tu.de
Also difference can come from different settings. i.e. this behavior is largely dependent on the way how big is the paging area configured for windows (if at all). 

Logs etc will not help. The machine is seemingly goes into extreme swapping activity during the linking process. The linker is an external tool outside of our control. As for compiling in the desktop folder, that should nor be a problem.

Long story short, you need more memory (or you may experiment switching off the virtual memory paging. That would at least give you an abrupt crash instead of going into an almost infinite paging)

Fatemah Amani

unread,
Aug 23, 2015, 9:23:01 PM8/23/15
to OMNeT++ Users
hi, can you send me download link omnet++4.6 on window?
plllz help me
thanx
Reply all
Reply to author
Forward
0 new messages