E-Series EF system rebooting due to cache block size
Applies to
- EF300 and EF600 with NVME drives AND the following conditions:
- Data Assurance is disabled on one or more volumes in a volume group
- Controllers are configured for 8KB cache blocks
- NetApp SANtricity OS 11.70.3 and before
Issue
EF-Series controller(s) reboot randomly and may cause volume failures.
- Download a Support bundle
- Open the
stateCaptureDatafile - Verify if the below error is displayed:
Example:
Log Entry #44 (Core 0) Sep-16-2022 09:25:06 AM ----
PCI SERR Exception
SERR Summary: 0x00000020
SERR Return Status: 0x00000000
SERR Collection Count: 0x2
PCI-E Root to PLX9797-1 (Unit 1)
VID 0x8086 DID 0x2030 B100:D0:F0
Bridge Secondary PCI Status = 0x2000
PCI-E XP Uncorrectable Status = 0x00000040
SERR Summary Flags = 0x80000021
PLX9797-1 Port 0 to Drive11 (Unit 15)
VID 0x10b5 DID 0x9797 B102:D11:F0
PCI-E Device Status = 0x000D
PCI-E AER Uncorrectable Status = 0x00100000
Header Log 0 = 0x05008001
Header Log 1 = 0x64000003
Header Log 2 = 0x71000000
Header Log 3 = 0xf5b8304c
PCI-E AER Correctable Status = 0x00008000
PCIe Link down.
SERR Summary Flags = 0x80000063
NOTE: Recovered from NVME Link-Down Errors