Hello,
We're on DSpace 5 with Postgres, but this question is about our apache 
proxy. 
I've noticed that in certain cases when httpd (we're on apache2.2) is 
running but tomcat is initializing or hung up (such as when it hits the jvm 
memory limit or runs out of system resources), that mod_jk still sends 
requests to tomcat and seems to wait indefinitely for a response. 
One potentially reproducible symptom of this happens when I restart the 
stack. If I leave httpd running and restart tomcat and postgres, I see an 
explosion of postgres connections when tomcat finishes initializing. If I 
stop httpd and wait until tomcat is finished initializing (when I see the 
message "Server Startup in #### seconds" in the tomcat log), and then 
restart httpd it starts with many fewer postgres connections. 

I see that mod_jk has a number of timeout options that can be set up on 
workers, for example worker.name.socket_timeout and 
worker.name.ping_timeout. I've looked at 
https://wiki.duraspace.org/display/DSPACE/ModJk, but it doesn't mention 
these timeout options. I'm wondering if one of these directives might be an 
appropriate way to solve the problem. 
Has anyone else experienced the problem described above?  Is anyone else 
using these mod_jk worker timeout directives with DSpace?  

Reference for the worker directives I mentioned is here: 
https://tomcat.apache.org/connectors-doc/reference/workers.html

Thanks,
Seth Robbins

-- 
You received this message because you are subscribed to the Google Groups 
"DSpace Technical Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/dspace-tech.
For more options, visit https://groups.google.com/d/optout.

Reply via email to