Potential EXT4 kernel issue
We are using ext4 on a 4TB SAN with iSCSI and have experienced a few issues where the load average has hit 580, yes you read that right, and after quite a bit of research it would appear we may be hitting:
Bug 14830 - When other IO is running sync times go to 10 to 20 minutes
If you are running ext4 on a iSCSI SAN I would be very interested to here which scheduler you are running. My initial thoughts are to switch too deadline instead of noop even though we have a pretty well tuned SAN.