Repeatable RAID 0 Superblock Corruption
I have a set of industrial hardened workstations running Linux 3.8 ES. Each workstation utilizes a RAID 0 configuration utilizing (2) 300GB SCSI drives. Both systems are designed to capture large amounts of process data and telemetry. The data is captured in an Oracle database and then processed with post processing software.
One of the workstations runs fine and has experienced no failures. The second system will run for an variable amount of time and then crash, resulting in a corrupted superblock on the RAID.
The problematic system has been rebuilt with Linux only and configured. I am running a stress test right now and it does not seem to cause the system to crash.
My next step will be to clear Oracle of suspicion. I seem to remember Oracle being a part of a realtime data collection problem years ago.
I will post further details. I would appreciate any advise on stress test utilities for the hardware, and a good way to beat Oracle to death.
I will collect the hardware configuration and post it soon. I have just been asked to help with this project. I am still getting up to speed.
|All times are GMT -5. The time now is 10:10 AM.|