Openfire service is causing the cpu usage on the windows 2003 server to be at 100%. It show very hign and constant in the Task Manager. What could be causing this to happen? It is causing my server to run very slow. This is the only service running on this VMWare server. I recently upgraded Openfire to 3.6.3, the latest version.
Here is the warning log information for the time period the CPU was at 100 percent: I have nothing from the error log for that day and time period. it only gives current day.
2009.01.22 10:42:31 Going to buffer response body of large or unknown size. Using getResponseBodyAsStream instead is recommended.
2009.01.22 10:42:32 Going to buffer response body of large or unknown size. Using getResponseBodyAsStream instead is recommended.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:30 Sample task not run because less then a second has passed since last sample.
2009.01.22 13:53:53 Sample task not run because less then a second has passed since last sample.
2009.01.22 18:29:42 TLS was required by the server and connection was never secured. Closing connection : org.jivesoftware.openfire.nio.NIOConnection@a14e84 MINA Session: (SOCKET, R: /130.39.32.165:2011, L: /130.39.12.46:5222, S: 0.0.0.0/0.0.0.0:5222)
The usage started around 1130am and went until we restarted the VMserver at 2pm. Once we rebooted the server CPU usage dropped down to 11 percent which is fine and is running a lot smoother. I stopped and restarted the openfire service just to make sure nothing was hanging around to bump up the CPU usage again. once that was done all Client Spark connections were cut (of coarse), but the client did not reconnect its connection on its own like it was supposed to. Had to exit spark and then reopen it and log back into itsince auto login did not work and all users were disconnected also.
Well, there was no information in the Error Log for the time period that the CPU usage was 100%, says my co-worker. Is there anything that would make the openfire service run at a high percentage, the percentage it was running at was 65%?
The problem is fixed now because of a rebooted of the server, but just curious for future issues.
I cant say anything about high CPU. But the problem that Spark is not reconnecting is probably not related to this. I see this often too, but other clients like Exodus still manage to reconnect.
little memory causes every java process to run at 100% CPU, so you may want to check the free memory in the admin console. If this does not help you should identify the thread which is using high CPU. You may review the “Openfire Support - Documents”, if this does not help I may upload a plugin which lists all threads and the used CPU.