Quick Summary:
Have multiple Dell PE320 servers (both tower and rack) deployed with Server 2012 or Server 2012 R2.
Servers experience a severe performance issue at random intervals (between a few weeks and a month). CPU usage reports 100% in Windows, although total of running processes is near 20%. Deferred Procedure Calls show between 3% and 5%. Services in Windows begin to crash, due to startup timeouts.
Typically, an update of various firmwares (idrac, lifecycle, perc) solve the issue - only to have it repeat again.
This past Friday, it occured the second time on a brand new installation. Updating of the firmware failed and bricked the iDRAC. New motherboard on the way Monday.
Last night, while trying to finish up a new deployment; a new R320 begin exhibiting the same issues. Recently I realized that all of these servers in question are using USB drives for backup. Figuring the USB driver may be the issue, I disabled and performed a reboot.
I'm posting this while waiting for it to get to a point to respond to RDP.
These are a mix of configurations:
a: T320, 6 core 2.0Ghz cpu, 24GB RAM (8x3G). PERC H310/RAID 1. SATA drives.
b: T320, 6 core v2 2.50Ghz, 24GB RAM (8x3G). PERC H710p/RAID5. N-SAS drives.
c: T320, 6 core v2 2.50Ghz, 24GB RAM (8x3G). PERC H710p/RAID5. SAS drives with 160GB Cachecade.
I have an ongoing ticket, for "B" (the one with a new motherboard on the way) - but this seems way beyond coincidental. Am I missing something on these deployments?