From the linked discussion provided, on our proof of concept server (very small footprint), we encountered issues with history delete not working as well. Considering that this was a test server, we need to understand what is required to resolve this issue on a larger scale when we install in production. I looked through the Zenoss ticket queue and read that the fix originally slated for 2.5.1 is now slated for 2.5.2.Can someone confirm when the fix will be made available? Does the fix resolve ALL of the known issues with history clean up?

We are finally getting ready to roll out a production version of Zenoss. Our DBA's are concerned about the database growing to the point where performance is impacted and clean up would be very time consuming depending on the database size (i.e. a long outage would be required). I would like to install the latest stable version of Zenoss so I do not have to worry about upgrading after the install or implementing adhoc routines to maintain the database.

The problem we run into with MySQL is that once the data grows (via innodb), it is difficult to reclaim space once history delete, optimize, etc. are run. We were informed that an export and import of the database is required to reclaim disk space. Has anyone run into database growth and maintenance issues before? Is there a best practice in place to manage MySQL for Zenoss?

We will be using 32GB RAM, 4 CPU, 3-500GB drives in a Raid 5 configuration which equates to about 600GB of data we can store. With 1500 monitors and a data retention rate of 1 year. It would be great to calculate approximately how large the database could get and how much storage is needed for RRD files. I suspect that there is no formula we can use.My management requires that we retain 1 year of data. Are there any performance issues we need to consider if we retain up to 1 year of history with the configuration mentioned?

Finally, is a procedure available to copy MySQL Events and ZenRRD files (say a 6 month snapshot of data) to another Zenoss server, disable active monitoring for all devices so no new events or RRD files are update and use that server as a reporting server for archived data? My thought process is that if we need to retain less than a year data, we could archive data every 6 months to another parallel system to read historical trending information.

I would appreciate any insight that can be provided regarding these questions.

- Ken