dimdim 4.5 just see cherrypy icon sometimes

giving the folllowing commands give me no results:

rpm -qa | grep dimdim

This forum topic is really about the issue described in the first few posts. If you can't install it, or can but are constantly getting the cherry icon, that's really a different issue.

These aren't my own forums so it's not my place to say to start a new post, but personally as I really want the issue originally described to be fixed, I would rather the problem wasn't watered down with other people having different issues installing. This isn't an installation issue at all, it's a bug once everything is up and running perfectly for hours on end, *then* there is a problem randomly after that.

When a topic is started, it continues to other issues.
and if these are not your forums, you better keep you mouth shut.

The dimdim support really sux. Many new posts are left without any reply. That is why people prefer to post in the other related posts.

Originally Posted by picseltech

Please see attached image showing the issue:



Thanks,

Originally Posted by picseltech

Hi,
All users are using FF3 here, but across Windows XP, Mac and Linux (Ubuntu).

It appears that dimdim 4.5 stays running happily for around about 1 day then just fails after then. I have to do a ./StopDimDim.sh and a ./StartDimDim.sh to get it working again, basically the problem is that if it 'stops working' you see a cherrypy icon up where the favicon area is in the browser, plus you also see the same icon in the browser window. That's it. You don't see anything else. (This is attempting to visit any dimdim page in a browser, for example http://mydimdimhost/dimdim/).

I went thru the logs to see if I could fathom a reason for this and I can't see anything glaringly obvious, but it seems to be most often when you end a meeting. It doesn't happen every time a meeting ends, but it seems to be the most common link I can come up with at this point.

Here is an excerpt of a log. You will note at around 12:18 i restarted the DimDim service and it sucessfully worked again. The meeting ended around 11:16:

INFO | jvm 5 | 2008/12/11 11:16:41 | Stream Control --- appHandle -x-1228994201517
INFO | jvm 5 | 2008/12/11 11:16:41 | Stream Control call complete in -1
INFO | jvm 5 | 2008/12/11 11:25:02 | ********* Sending out first (10 minute) warning.
STATUS | wrapper | 2008/12/11 12:18:46 | TERM trapped. Shutting down.
INFO | jvm 5 | 2008/12/11 12:18:46 | Received from the stream publisher:
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- cm -stop-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- conferenceKey -itmeeting-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- resourceId -r11e259d24affw4q6j
8kb1h7eh-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- mediaId -68cf689c5a0ba34c91119
6a6-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- presenterId -0-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- presenterPassKey --
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- appName -null-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- appHandle -x-1228994783274
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control call complete in -0
INFO | jvm 5 | 2008/12/11 12:18:46 | Wrapper Manager: The Wrapper code did not ping th
e JVM for 58 seconds. Quit and let the Wrapper resynch.
INFO | jvm 5 | 2008/12/11 12:18:46 | Dec 11, 2008 11:27:14 AM org.apache.coyote.http11
.Http11BaseProtocol pause
INFO | jvm 5 | 2008/12/11 12:18:46 | INFO: Pausing Coyote HTTP/1.1 on http-127.0.0.1-4
0000
INFO | jvm 5 | 2008/12/11 12:18:46 | Dec 11, 2008 11:27:15 AM org.apache.catalina.core
.StandardService stop
INFO | jvm 5 | 2008/12/11 12:18:46 | INFO: Stopping service Catalina
INFO | jvm 5 | 2008/12/11 11:16:41 | Stream Control --- appHandle -x-1228994201517
INFO | jvm 5 | 2008/12/11 11:16:41 | Stream Control call complete in -1
INFO | jvm 5 | 2008/12/11 11:25:02 | ********* Sending out first (10 minute) warning.
STATUS | wrapper | 2008/12/11 12:18:46 | TERM trapped. Shutting down.
INFO | jvm 5 | 2008/12/11 12:18:46 | Received from the stream publisher:
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- cm -stop-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- conferenceKey -itmeeting-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- resourceId -r11e259d24affw4q6j
8kb1h7eh-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- mediaId -68cf689c5a0ba34c91119
6a6-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- presenterId -0-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- presenterPassKey --
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- appName -null-
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control --- appHandle -x-1228994783274
INFO | jvm 5 | 2008/12/11 12:18:46 | Stream Control call complete in -0
INFO | jvm 5 | 2008/12/11 12:18:46 | Wrapper Manager: The Wrapper code did not ping th
e JVM for 58 seconds. Quit and let the Wrapper resynch.
INFO | jvm 5 | 2008/12/11 12:18:46 | Dec 11, 2008 11:27:14 AM org.apache.coyote.http11
.Http11BaseProtocol pause
INFO | jvm 5 | 2008/12/11 12:18:46 | INFO: Pausing Coyote HTTP/1.1 on http-127.0.0.1-4
0000
INFO | jvm 5 | 2008/12/11 12:18:46 | Dec 11, 2008 11:27:15 AM org.apache.catalina.core
.StandardService stop
INFO | jvm 5 | 2008/12/11 12:18:46 | INFO: Stopping service Catalina

Any assistance much appreciated.

Hi,

Sorry for the delayed response. Could you please provide me the following information from the machine where Dimdim server is running:

1. df -h
2. du -h /usr/local/dimdim
3. free


Also could you please send me the wrapper.log available under /usr/local/dimdim/ConferenceServer/apache-tomcat-5.5.17/logs to rajesh@dimdim.com.

Thanks,
-Rajesh
rajesh@dimdim.com

Originally Posted by picseltech

I really do hope that someone can help with this as I've searched high and low for an answer to the problem and I really don't think I'm doing anything wrong.

I thought it could be the RAM (we allocated 1GB to the VM) so I upped to 2GB but it made no difference. The load on the box is fine and the service works rock-solid when in use, but when left alone for a while the cherrypy icon comes back again and I have to manually restart the service before it can work.

I noticed I can do a restart without a ./StopDimDim.sh, I just need to do a ./StartDimDim.sh and that's enough to get it working again. Also, I don't know if it was a fluke but I started just the presentation manager via the start_pmgr.sh script. It reported a 'socket already in use' error but when I refreshed the dimdim web page the cherrypy icon had disappeared and dimdim was working as normal again. This isn't permanent though, as the cherrypy icon comes back again later, but potentially it may have something to do with the presentation manager. Again, I'm no dimdim employee so I'm stabbing in the dark here.

Please *please* can one of you dimdim guys help us here. I can't deploy dimdim in our corporate environment for testing until we can have general reliability of the product. It's a wonderful product otherwise and it seems like a tiny little bug is crippling it.

Thanks.

Hi,

We have tried in our test setup and we could not reproduce the issue. Can you please provide us the following:

1. wrapper.log available under /usr/local/dimdim/ConferenceServer/apache-tomcat-5.5.17/logs

2. cherrypy logs namely cherrypy_access.log and cherrypy_error.log available under /usr/local/dimdim/Mediaserver/www/logs

This would help us in debugging the issue.

Thanks,
-Rajesh
rajesh@dimdim.com

Originally Posted by ezaban

hi guys,

finally after days of working and after hundereds of tricked, I successfully installed dimdim on RPM package.

But I have dump question. How do you guys run that ./Startdimdim.sh command?
I did not find it anywhere.

Please give me some clue.

Hi,

The start script is available under /usr/local/dimdim. You can download the readme for rpm pack from sf.net which explains about the directory structure of Dimdim and details about how to start and stop Dimdim components.

Thanks,
-Rajesh
rajesh@dimdim.com

Originally Posted by ezaban

I tried to reinstall it again but it gives me this error:

# rpm -ivh Dimdim-4.5_SF_i386.rpm
Preparing... ########################################### [100%]
Checking preinstall checks
CentOS release 5.2 (Final)
Found previous version of dimdim, Please remove any previous version of Dimdim
Exit Status....
error: %pre(dimdim-4.5-1.i386) scriptlet failed, exit status 1
error: install: %pre scriptlet failed (2), skipping dimdim-4.5-1




I searched again carefully, dimdim is NOT installed under usr/local folder
so where is it hiding?

Hi,

Were you able to succeed with the installation?

Regards,
-Rajesh

Rajesh - thanks in advance for your help, an e-mail with logs, command outputs etc is on its way to you.

ezaban - It's not worth my time responding to your post. Best of luck with the installation.

Originally Posted by dhara01.dimdim

Hi,

Sorry for the delayed response. Could you please provide me the following information from the machine where Dimdim server is running:

1. df -h
2. du -h /usr/local/dimdim
3. free


Also could you please send me the wrapper.log available under /usr/local/dimdim/ConferenceServer/apache-tomcat-5.5.17/logs to rajesh@dimdim.com.

Thanks,
-Rajesh
rajesh@dimdim.com

Hi,

I have received the log files and looks like there is an issue with CPU usage of the machine.

In the wrapper.log file you have provided, i could see "The Wrapper code did not ping the JVM for 58 seconds. Quit and let the Wrapper resynch."


Are you using the VM image or the rpm pack of Dimdim 4.5? If you are running vm image, could you please let me know the output for "free", "top" and df -h from the base machine where vmware server is running when you are observing the non-responsive state?

The problem is because tomcat went to non-responsive state and Dimdim is a web application that is deployed in tomcat.

You can refer to http://wrapper.tanukisoftware.org/do...u-timeout.html where you can increase the timeout based on your machine specifications.

You have to configure and fine tune this property in wrapper.conf "wrapper.cpu.timeout=10"

Try changing the configuration and let me know if you still see the issue. Please keep us posted about the updates.


Thanks,
-Rajesh
rajesh@dimdim.com

Hi,

Thanks for the response.

We are using the virtual machine version of dimdim.

The CPU and memory are ok both on the virtual machine and the host when this happens, so I'm not sure that it's a resources issue.

I have sent you an e-mail of the things you asked for. Thanks again.