Cisco seems to be pushing this upgrade, as per the below text in the release notes:

"Resolution of issue introduced in x.3.4 release. The driver for the x.3.6 release is to correct CSCsr47032, a defect introduced in x.3.4 that results in the database gradually filling up with unneeded audit logs. This defect can lead to a file system overflow when archiving is enabled or exporting is used for migration purposes.

Customers should upgrade to the x.3.6 release soon as possible to avoid consuming all hard drive space on CS-MARS 20/20R, 25/25R, 50, and 55 models. High-end models are not in danger of disk overflow but may experience a malfunction of archiving and export depending on the quantity of accumulated audit logs.

To determine whether an appliance is affected by this defect, click ADMIN > System Maintenance >View the Audit Trail, and look for messages like "8/18/08 3:50:11 PM PDT Unknown User (unknown) Database insert: DbReportResult: DbReportResult:215178". If thousands of such messages appear from the previous hour, an upgrade to x.3.6 is strongly recommended "

Regards

Farrukh

I just upgrade one of our test boxes.

Upgrade went well, still testing to see if there are any new "issues".

One thing to keep in mind is read the release notes. make sure that you follow the upgrade instructions. Specifically to do a raid status and diskusage to ensure there are no issues.

Another thing that I have to keep reminding myself, is that according to the csmars documentation, if you have not done a reboot in 90 days (I think) or there have been a certain number of reboots, it does a chkdisk on the entire file system. So if the appliance does not come back and respond to a ping right away after the reboot, there is a good possiblity that it is doing the chkdisk. This takes quite some time depending on the model you have.