Hi all,
I don't know if I am right here but after using the DMS for a while I noticed that I have 100% load from the JAVA process from logicalDoc.
This is my System:
total used free shared buffers cached
Mem: 516300 236788 279512 0 12796 186336
-/+ buffers/cache: 37656 478644
Swap: 765212 0 765212
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda1 3889892 2173272 1519024 59% /
tmpfs 258148 0 258148 0% /lib/init/rw
udev 10240 572 9668 6% /dev
tmpfs 258148 0 258148 0% /dev/shm
top - 12:10:27 up 1:18, 2 users, load average: 0.00, 0.08, 0.45
Tasks: 46 total, 1 running, 45 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3%us, 1.0%sy, 0.0%ni, 98.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 516300k total, 236808k used, 279492k free, 12868k buffers
Swap: 765212k total, 0k used, 765212k free, 186340k cached
Hosted on a 3 GHz VMWare Server.
Does anyone have an idea ?
Kind regards
Christian
100 % Load after Startup
Moderator: car031
Post
Re: 100 % Load after Startup
Which version of the software are you using?
I suggest you set a scheduled task to restart the tomcat that runs logicaldoc.
For example you should execute the restart every night at about 2:00AM
I suggest you set a scheduled task to restart the tomcat that runs logicaldoc.
For example you should execute the restart every night at about 2:00AM
Post
Re: 100 % Load after Startup
I had mine on a virtual machine too (windows server 2008 OS) and it ran really slow. I put it on a different non VM machine (also server 08) and its runs a lot better. For some reason it just doesn't like VM's.
Post
Re: 100 % Load after Startup
LogicalDOC 5.x version requires a lot of CPU and memory, so you must have a big server and you must assign a lot of resources to the VM if you want to run it inside a VM.
With LD 6.x we put a lot of work in order to reduce the resources needed so you should easily run it inside a VM.
The only problems you can have during first indexing when there are a lot of files to parse and add to index.
In LD 6.1 we introduced the possibility to fine tune the indexing process and you can decide if you want to use the OCR or also the size of the batch of documents to be indexed.
With LD 6.x we put a lot of work in order to reduce the resources needed so you should easily run it inside a VM.
The only problems you can have during first indexing when there are a lot of files to parse and add to index.
In LD 6.1 we introduced the possibility to fine tune the indexing process and you can decide if you want to use the OCR or also the size of the batch of documents to be indexed.
Post
Re: 100 % Load after Startup
Is there anything else we can do to troubleshoot this issue? We are running on a CentOS XenServer VM with 2 cores and 8gb RAM, latest LogicalDOC 6.6 and the following Java versions:
Java(TM) SE Runtime Environment (build 1.6.0_38-b05)
Java HotSpot(TM) 64-Bit Server VM (build 20.13-b02, mixed mode)
When our app stalls and we see 100-150% Java CPU Utilization, our dms.log file fills up with Java memory errors:
08.01.2013 13:43:33,420 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.Group#-1 failed:
java.lang.OutOfMemoryError: PermGen space
08.01.2013 13:43:34,320 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.Group#4 failed:
java.lang.OutOfMemoryError: PermGen space
08.01.2013 13:43:35,498 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.UserHistory#3571722 failed:
java.lang.OutOfMemoryError: PermGen space
We updated our logicaldoc.sh file to the following:
JAVA_OPTS="-Xmx4000m -XX:MaxPermSize=256m -Djava.net.preferIPv4Stack=true -Djava.awt.headless=true"
However 4 hours after restarting we were back to 100% CPU util with the same errors outputting in our dms.log. Anything else we can try?
We have 1 document loaded, a few dozen (empty) folders, and 5 users....there's nothing loaded just a fresh install with some initial configurations.
Java(TM) SE Runtime Environment (build 1.6.0_38-b05)
Java HotSpot(TM) 64-Bit Server VM (build 20.13-b02, mixed mode)
When our app stalls and we see 100-150% Java CPU Utilization, our dms.log file fills up with Java memory errors:
08.01.2013 13:43:33,420 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.Group#-1 failed:
java.lang.OutOfMemoryError: PermGen space
08.01.2013 13:43:34,320 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.Group#4 failed:
java.lang.OutOfMemoryError: PermGen space
08.01.2013 13:43:35,498 ERROR (net.sf.ehcache.store.disk.DiskStorageFactory: ?) -- Disk Write of com.logicaldoc.core.security.UserHistory#3571722 failed:
java.lang.OutOfMemoryError: PermGen space
We updated our logicaldoc.sh file to the following:
JAVA_OPTS="-Xmx4000m -XX:MaxPermSize=256m -Djava.net.preferIPv4Stack=true -Djava.awt.headless=true"
However 4 hours after restarting we were back to 100% CPU util with the same errors outputting in our dms.log. Anything else we can try?
We have 1 document loaded, a few dozen (empty) folders, and 5 users....there's nothing loaded just a fresh install with some initial configurations.
Post
Re: 100 % Load after Startup
Every installation needs fine tunig when going in production, but in your case is evident that something strange happens in your environment.
You should have to ask for professional support to investigate the issue.
In any case try use these options in bin/logicaldoc.sh:
JAVA_OPTS="-Xmx3000m -XX:MaxPermSize=256m -Djava.net.preferIPv4Stack=true -Djava.awt.headless=true -XX:+UseConcMarkSweepGC -XX:+CMSClassUnloadingEnabled -XX:+CMSPermGenSweepingEnabled"
You should have to ask for professional support to investigate the issue.
In any case try use these options in bin/logicaldoc.sh:
JAVA_OPTS="-Xmx3000m -XX:MaxPermSize=256m -Djava.net.preferIPv4Stack=true -Djava.awt.headless=true -XX:+UseConcMarkSweepGC -XX:+CMSClassUnloadingEnabled -XX:+CMSPermGenSweepingEnabled"
Who is online
Users browsing this forum: No registered users and 28 guests