100% CPU Usage after startup

[trent]

Hi guys,

I've got 100% CPU usage on one of my CPUs after startup of tomcat 5.

I installed using the WAR, and have mysql running.

I edited the database configuration, and the data folder configuration, and put the entire contents of "extensions" in common/shared/classes/alfresco.

I'm not getting any more errors in any tomcat logs, but the one CPU has been running at 100% since startup about 10 minutes ago.

I am unable to access any of my tomcat contexts, so obviously it must be stuck in a static initializer somewhere.  It's almost like there must be an infinite loop or something.

FYI: At first I was getting an out of memory error, but resolved that with the following…
-server -Xmx512m -Xms512m -XX:MaxPermSize=64m -XXermSize=64m -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSParallelRemarkEnabled

Any help would be appreciated.  I'd like to try out Alfresco to see if it's worth using for my web site content.

[trent]

Oh, by the way, it did create the files in the data directory just fine, and it did create the tables in the mysql database.  So file permissions, and mysql connections seem just fine.

[kevinr]

The first time you start the Alfresco server it will create the initial database schema automatically - this normally takes a couple of minutes. Sounds like yours is taking a lot longer - what is the machine spec? Also do you have a lot (maybe DEBUG) logging turned on? As by default some of the open-source libraries used by Alfresco generate a lot of DEBUG level logging which could slow down the server…

Thanks,

Kevin

[trent]

The first time you start the Alfresco server it will create the initial database schema automatically - this normally takes a couple of minutes. Sounds like yours is taking a lot longer - what is the machine spec? Also do you have a lot (maybe DEBUG) logging turned on? As by default some of the open-source libraries used by Alfresco generate a lot of DEBUG level logging which could slow down the server…

Thanks,

Kevin

Hi Kevin,

There is no output to any logs in the tomcat log directory, except to say that the context is being initialized.  I waited approximately 45 minutes before killing the process.

I have a Duo Core Athlon 64 CPU.  The system is pretty fast, and has 2G of memory, although I only allocated 1G to the Xen VM, once I realized I was running out of memory.  I am no longer getting out of memory errors.

processor       : 1
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 75
model name      : AMD Athlon™ 64 X2 Dual Core Processor 3800+
stepping        : 2
cpu MHz         : 2004.202


However, I just thought of something.  Perhaps I need to wipe the database again.  Maybe it ran out of memory in the middle of the initialization, and completely messed things up.  What do you think?

[kevinr]

The machine is more than enough my Core2 laptop can bootstrap the server in about 60 seconds. Yes i would try creating the db schema instance and removing the alf_data dir contents.

Thanks,

Kevin

[sbzoom]

This happened to me as well.  I had two installations, one RedHat, the other Windows.  The RedHat install runs great.  The Windows one I gave up on a while ago.  Everything seemed fine until someone visited the homepage.  Once someone accesses the application, the CPU spiked to 100% and stayed that for… for DAYS.  I'd stop all processes and try again, same effect.  So I just gave up on windows - things run great on linux.


Charlie

[trent]

This happened to me as well.  I had two installations, one RedHat, the other Windows.  The RedHat install runs great.  The Windows one I gave up on a while ago.  Everything seemed fine until someone visited the homepage.  Once someone accesses the application, the CPU spiked to 100% and stayed that for… for DAYS.  I'd stop all processes and try again, same effect.  So I just gave up on windows - things run great on linux.


Charlie

I'm having this issue on Linux.  But, I'm not running RedHat, I'm running gentoo.

[devg]

Was this issue ever resolved? This is happening to us on Linux and we already have a database/repository with lots of file data, so I don't want to start from scratch.

When I did full index recovery, it took almost 7 hours to finish, but right after saying 100% recovery complete, it went into this inifinite loop. Now it happens every time I restart - I don't want to try another index recovery becuase it takes 7 hours.

I recently upgraded to 2.1 from 2.0 - it worked fine for about a day after that.

Any help is appreciated.

This happened to me as well.  I had two installations, one RedHat, the other Windows.  The RedHat install runs great.  The Windows one I gave up on a while ago.  Everything seemed fine until someone visited the homepage.  Once someone accesses the application, the CPU spiked to 100% and stayed that for… for DAYS.  I'd stop all processes and try again, same effect.  So I just gave up on windows - things run great on linux.


Charlie

I'm having this issue on Linux.  But, I'm not running RedHat, I'm running gentoo.

[trent]

No, I never resolved the issue.  I kind of just gave up.  I was hoping for an easy to use, easy to install, content management system.  I just wanted to use it to maintain my own site.  But, I think Alfresco is over kill for my needs.

[kevinr]

This happened to me as well.  I had two installations, one RedHat, the other Windows.  The RedHat install runs great.  The Windows one I gave up on a while ago.  Everything seemed fine until someone visited the homepage.  Once someone accesses the application, the CPU spiked to 100% and stayed that for… for DAYS.  I'd stop all processes and try again, same effect.  So I just gave up on windows - things run great on linux.

We haven't had any customers report this. About half of our developers use Windows laptops also and haven't seen this. We'd really appreciate if when you see this happening, you press CTRL+BREAK to output a thread dump report into your console window then either email us or post the report here.

Thanks,

Kevin