11-23-2016 05:05 AM
Hello fellow Alfrescians,
I have multiple data in my custom Data List (say 3,000 rows). When opening the list, it takes too long, and when it finishes it prompts 'No list items'. I checked the JVM-Memory in https://localhost:8443/solr4/#/, and it consumes 99.9%. Kindly give me some advice on how to optimize the settings in Alfresco to fix this.
Best Regards,
June
11-28-2016 04:15 AM
Hi June:
5Gb may be a good starting point for JVM, but the needed heap depends on many things such concurrent users, massive processes.... You need to profile JVM to see how much memory you need.
But if you are experiencing other type problems with SOLR such as broken indexes, you should check your installation and find the cause of the frequent corruption of indices in your solr.log (for example, a wrong hot backup process...) . By the way, Solr is a great consumer of cpu and memory..., and the needed memory depends on the number of nodes in the repository:
Calculate the memory needed for Solr nodes | Alfresco Documentation
Finally, and as I pointed out in my previous message, all of this does not ensure a good behaviour for the big query
Regards.
--C.
11-24-2016 01:53 PM
Hi June:
It may help to diagnose, to inspect javascript and solr queries in catalina.out (and solr.log). Check this to obtain more info about the query:
- How to get logs for Alfresco querys - zylk
On the other hand, better JVM resources (more heap) or faster disks for SOLR indices can help a little bit, although only this, is not a silver bullet.
Regards.
--C.
11-27-2016 08:43 PM
Hi Cesar,
What I did was, increase the JVM Memory (JAVA_OPT) to 5 GB. Am I doing it right?.
Also, do you know by any chance how to optimize solr4 to avoid doing reindexing? Because I encountered broken indexes most of the time.
Best Regards,
June
11-28-2016 04:15 AM
Hi June:
5Gb may be a good starting point for JVM, but the needed heap depends on many things such concurrent users, massive processes.... You need to profile JVM to see how much memory you need.
But if you are experiencing other type problems with SOLR such as broken indexes, you should check your installation and find the cause of the frequent corruption of indices in your solr.log (for example, a wrong hot backup process...) . By the way, Solr is a great consumer of cpu and memory..., and the needed memory depends on the number of nodes in the repository:
Calculate the memory needed for Solr nodes | Alfresco Documentation
Finally, and as I pointed out in my previous message, all of this does not ensure a good behaviour for the big query
Regards.
--C.
11-28-2016 04:25 AM
Thank you so much, this is good information to start looking into. I appreciate the help.
Best Regards,
June
Tags
Find what you came for
We want to make your experience in Hyland Connect as valuable as possible, so we put together some helpful links.