Recently updated my UNRAID Server MB to a Gigabyte Aorus Master Z590 so I could utilise the onboard 10Gbe NIC. This has been working flawlessly and without any obvious issues.

Sadly though, I now get random errors in my syslog, sometimes multiple in a 24hr period and others a day or 2 apart.

Oct  7 06:11:43 Wacko-UNRAID kernel: pcieport 0000:00:1c.4: AER: Multiple Correctable error message received from 0000:06:00.0

Oct  7 06:11:43 Wacko-UNRAID kernel: atlantic 0000:06:00.0: PCIe Bus Error: severity=Correctable, type=Physical Layer, (Receiver ID)

Oct  7 06:11:43 Wacko-UNRAID kernel: atlantic 0000:06:00.0:   device [1d6a:d107] error status/mask=00000041/0000a000

Oct  7 06:11:43 Wacko-UNRAID kernel: atlantic 0000:06:00.0:    [ 0] RxErr                  (First)

Oct  7 06:11:43 Wacko-UNRAID kernel: atlantic 0000:06:00.0:    [ 6] BadTLP 

 

From what I have researched and my limited understanding this seems to relate to the ASPM of the device (Power Management)

I have tried to enable the ASPM features in the BIOS: PEG ASPM, PCH ASPM and also enabled Native ASPM which allows for OS implementation of ASPM. I did not enable DMI ASPM as I heard this caused issue. Enabling these settings had zero effect on the error.

I have searched through the forums and have found suggestions to add “ pcie_aspm=off” to the sys config found here https://forums.unraid.net/topic/118286-nvme-drives-throwing-errors-filling-logs-instantly-how-to-resolve/#findComment-1165004

But I am somewhat reluctant to do so as I don’t understand the full implications of doing so.

I have attached screenshots to help identify the system device and of the log error found in Syslog.

As always any advise or knowledge on how to proceed etc would be greatly appreciated.   

9 hours ago, wacko37 said:

to add “ pcie_aspm=off”

This is typically the best solution, NIC may use a little more power with that option

6 hours ago, JorgeB said:

This is typically the best solution, NIC may use a little more power with that option

Thanks for the time and expert advice, much appreciated.

I worry if "pcie_aspm=off" will have a negative energy efficiency effect on the other devices in the server, as I have a gpu binded to the VFIO for my VM etc.

So my question here would be if "pcie_aspm=off" is applied will it prevent the other PCIe from entering a power saving mode and therefore effecting the overall power efficiency of the server?

And if the above is indeed correct, would leaving the server in its current state be ok? (Apart from the few error logs a day, i see not other negative effects that im aware of)

AFAIK it will apply to all PCIe devices.

Should be fine to leave like that apart from the log spam

53 minutes ago, JorgeB said:

AFAIK it will apply to all PCIe devices.

Should be fine to leave like that apart from the log spam

Thanks so much! Now i can put my mind to rest....... hopefully 😁