unRaid already have edac, but for some reason is missing edac-util

 

It would provide much easier way to troubleshoot RAM and PCIe issues.

I would also like this.  In its absence I have been referring to this post which suggests the same information is available without the tool.

+1 for this

+1 here, too.

+1

+1

On 10/16/2019 at 12:54 PM, flaggart said:

I would also like this.  In its absence I have been referring to this post which suggests the same information is available without the tool.

 

Quote

# grep "[0-9]" /sys/devices/system/edac/mc/mc*/csrow*/ch*_ce_count
grep: /sys/devices/system/edac/mc/mc*/csrow*/ch*_ce_count: No such file or directory

 

Not seeing anything about EDAC in dmsg too. Does Unraid still support ECC?

On 9/22/2022 at 8:30 PM, realies said:

Not seeing anything about EDAC in dmsg too. Does Unraid still support ECC?

EDAC should usually load automatically if you CPU/Motherboard/Memory Controller supports it.

 

However you can load it manually using:

modprobe amd64_edac

for AMD CPUs

 

or for Intel 10th, 11th, 12th Gen processors with:

modprobe igen6_edac

 

If you get an error like "modprobe: ERROR: could not insert 'MODULNAME': No such device" then it's most likely the wrong module since this means that the module doesn't find a compatible hardware device.

 

BTW you can get all available modules with this command:

ls -la /lib/modules/*-Unraid/kernel/drivers/edac/

 

The main issue with EDAC is that it is really noisy at times and also can give you a lot of false positives and can ultimately drive you crazy...

EDAC also reports PCIe errors from what I know and since not all PCIe devices follow the entire PCIe standard there can be many, many, many, maaaannnny issues at certain times and with certain hardware combinations.

Here is an example from my Server.

 

Mainboard: Supermicro X10DRI-F

CPU: Intel Xeon E5-2630 v4

RAM: 4x 8GB ECC Memory

UNRAID: 6.12.6

 

# dmesg | grep EDAC
[  129.472264] EDAC MC: Ver: 3.0.0
[  129.482175] EDAC sbridge: Seeking for: PCI ID 8086:6fa0
[  129.482194] EDAC sbridge: Seeking for: PCI ID 8086:6fa0
[  129.482213] EDAC sbridge: Seeking for: PCI ID 8086:6f60
[  129.482226] EDAC sbridge: Seeking for: PCI ID 8086:6fa8
[  129.482234] EDAC sbridge: Seeking for: PCI ID 8086:6fa8
[  129.482244] EDAC sbridge: Seeking for: PCI ID 8086:6f71
[  129.482251] EDAC sbridge: Seeking for: PCI ID 8086:6f71
[  129.482261] EDAC sbridge: Seeking for: PCI ID 8086:6faa
[  129.482269] EDAC sbridge: Seeking for: PCI ID 8086:6faa
[  129.482279] EDAC sbridge: Seeking for: PCI ID 8086:6fab
[  129.482286] EDAC sbridge: Seeking for: PCI ID 8086:6fab
[  129.482292] EDAC sbridge: Seeking for: PCI ID 8086:6fac
[  129.482296] EDAC sbridge: Seeking for: PCI ID 8086:6fac
[  129.482302] EDAC sbridge: Seeking for: PCI ID 8086:6fad
[  129.482307] EDAC sbridge: Seeking for: PCI ID 8086:6fad
[  129.482312] EDAC sbridge: Seeking for: PCI ID 8086:6f68
[  129.482317] EDAC sbridge: Seeking for: PCI ID 8086:6f79
[  129.482325] EDAC sbridge: Seeking for: PCI ID 8086:6f6a
[  129.482333] EDAC sbridge: Seeking for: PCI ID 8086:6f6b
[  129.482341] EDAC sbridge: Seeking for: PCI ID 8086:6f6c
[  129.482365] EDAC sbridge: Seeking for: PCI ID 8086:6f6d
[  129.482374] EDAC sbridge: Seeking for: PCI ID 8086:6ffc
[  129.482377] EDAC sbridge: Seeking for: PCI ID 8086:6ffc
[  129.482384] EDAC sbridge: Seeking for: PCI ID 8086:6ffd
[  129.482387] EDAC sbridge: Seeking for: PCI ID 8086:6ffd
[  129.482394] EDAC sbridge: Seeking for: PCI ID 8086:6faf
[  129.482399] EDAC sbridge: Seeking for: PCI ID 8086:6faf
[  129.482617] EDAC MC0: Giving out device to module sb_edac controller Broadwell SrcID#0_Ha#0: DEV 0000:ff:12.0 (INTERRUPT)
[  129.482623] EDAC sbridge:  Ver: 1.1.2 



# lsmod | grep edac
sb_edac                24576  0
edac_core              65536  1 sb_edac


# grep "[0-9]" /sys/devices/system/edac/mc/mc*/csrow*/ch*_ce_count
/sys/devices/system/edac/mc/mc0/csrow0/ch0_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch1_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch2_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch3_ce_count:0



# lshw -class memory | grep ecc
       capabilities: ecc    
       configuration: errordetection=multi-bit-ecc


# mcelog --client
Memory errors
SOCKET 0 CHANNEL 0 DIMM 0
DMI_NAME "P1-DIMMA1" DMI_LOCATION "P0_Node0_Channel0_Dimm0"
corrected memory errors:
        0 total
        0 in 24h
uncorrected memory errors:
        0 total
        0 in 24h

SOCKET 0 CHANNEL 1 DIMM 0
DMI_NAME "P1-DIMMB1" DMI_LOCATION "P0_Node0_Channel1_Dimm0"
corrected memory errors:
        0 total
        0 in 24h
uncorrected memory errors:
        0 total
        0 in 24h

SOCKET 0 CHANNEL 2 DIMM 0
DMI_NAME "P1-DIMMC1" DMI_LOCATION "P0_Node0_Channel2_Dimm0"
corrected memory errors:
        0 total
        0 in 24h
uncorrected memory errors:
        0 total
        0 in 24h

SOCKET 0 CHANNEL 3 DIMM 0
DMI_NAME "P1-DIMMD1" DMI_LOCATION "P0_Node0_Channel3_Dimm0"
corrected memory errors:
        0 total
        0 in 24h
uncorrected memory errors:
        0 total
        0 in 24h

 

One can get the Information without `edac-util`, but maybe it would be still nice to have.

 

EDAC Kernel Module does load automatically.

 

I don't know if UNRAID will trigger a warning via WebUI if ECC Errors are detected, but this may be OT.

 

+1

Edited January 9, 20242 yr by pixeldoc81