Last night i downgraded Unraid. It's been all good since, knock on wood.

A couple of things I've noticed today are:
 

1. network messages

Syslog gives me these messages over and over:

Dec 22 20:15:23 mothership kernel: e1000e: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
Dec 22 20:15:23 mothership kernel: br0: port 1(eth0) entered listening state
Dec 22 20:15:23 mothership kernel: br0: port 1(eth0) entered listening state
Dec 22 20:15:25 mothership kernel: e1000e 0000:00:1f.6 eth0: Detected Hardware Unit Hang:
Dec 22 20:15:25 mothership kernel:  TDH                  <0>
Dec 22 20:15:25 mothership kernel:  TDT                  <1>
Dec 22 20:15:25 mothership kernel:  next_to_use          <1>
Dec 22 20:15:25 mothership kernel:  next_to_clean        <0>
Dec 22 20:15:25 mothership kernel: buffer_info[next_to_clean]:
Dec 22 20:15:25 mothership kernel:  time_stamp           <104f363f0>
Dec 22 20:15:25 mothership kernel:  next_to_watch        <0>
Dec 22 20:15:25 mothership kernel:  jiffies              <104f367e9>
Dec 22 20:15:25 mothership kernel:  next_to_watch.status <0>
Dec 22 20:15:25 mothership kernel: MAC Status             <40080083>
Dec 22 20:15:25 mothership kernel: PHY Status             <796d>
Dec 22 20:15:25 mothership kernel: PHY 1000BASE-T Status  <3c00>
Dec 22 20:15:25 mothership kernel: PHY Extended Status    <3000>
Dec 22 20:15:25 mothership kernel: PCI Status             <10>

Probably related, there appears to be packets dropped on my NICs:

My network is set up like this:

 

2. DMAR

Syslog also gives these messages repeatedly:

Dec 22 20:15:42 mothership kernel: dmar: DMAR:[DMA Read] Request device [04:00.0] fault addr 100000000 
Dec 22 20:15:42 mothership kernel: DMAR:[fault reason 06] PTE Read access is not set
Dec 22 20:15:42 mothership kernel: dmar: DRHD: handling fault status reg 3
Dec 22 20:15:42 mothership kernel: dmar: DMAR:[DMA Read] Request device [04:00.0] fault addr ffff2000 
Dec 22 20:15:42 mothership kernel: DMAR:[fault reason 06] PTE Read access is not set
Dec 22 20:15:42 mothership kernel: dmar: DRHD: handling fault status reg 3
Dec 22 20:15:42 mothership kernel: dmar: DMAR:[DMA Read] Request device [04:00.0] fault addr fffec000 
Dec 22 20:15:42 mothership kernel: DMAR:[fault reason 06] PTE Read access is not set
Dec 22 20:15:42 mothership kernel: dmar: DRHD: handling fault status reg 3
Dec 22 20:15:42 mothership kernel: dmar: DMAR:[DMA Read] Request device [04:00.0] fault addr fffc8000 
Dec 22 20:15:42 mothership kernel: DMAR:[fault reason 06] PTE Read access is not set
Dec 22 20:15:42 mothership kernel: dmar: DRHD: handling fault status reg 3
Dec 22 20:15:42 mothership kernel: dmar: DMAR:[DMA Read] Request device [04:00.0] fault addr ffe8e000 
Dec 22 20:15:42 mothership kernel: DMAR:[fault reason 06] PTE Read access is not set

It puzzles me, as 'device [04:00.0]' does not appear at all under 'System devices':

 

I find that a SCSI device have a somewhat similar address, but this isn't the same as "04:00:0", is it?:

[4:0:0:0]    disk    ATA      WDC WD30EFRX-68E 0A82  /dev/sde 

3. PCI sound card suddenly in another IOMMU group than before

This is more of a curiousity and annoyance than a real problem. I'm about try fixing this by changing card slots, but I'd thought I'd mention it anyway:
Back when I first set up my Unraid server, I purchased this card to pass it through to my planned VM, as I figured passing the onboard sound probably would have undesireable outcome on other components on my motherboard:

05:00.0 Audio device: Creative Labs SB Recon3D (rev 01)

All was good from the get go. I used it on my original 6.1.9 build and tjen later onto my 6.3.5 build (in a new VM). Last night, after downgrading, my VM (both old and newly created) would not boot with this card passed through. After taking a closer look at the 'System devices', I noticed that the card now had been assigned into an IOMMU-group with lots of other devices:

/sys/kernel/iommu_groups/7/devices/0000:00:1b.2
/sys/kernel/iommu_groups/7/devices/0000:00:1b.3
/sys/kernel/iommu_groups/7/devices/0000:03:00.0
/sys/kernel/iommu_groups/7/devices/0000:04:01.0
/sys/kernel/iommu_groups/7/devices/0000:05:00.0

These other devices are:

00:1b.2 PCI bridge: Intel Corporation Sunrise Point-H PCI Root Port #19 (rev f1)

00:1b.3 PCI bridge: Intel Corporation Sunrise Point-H PCI Root Port #20 (rev f1)

03:00.0 PCI bridge: ASMedia Technology Inc. ASM1083/1085 PCIe to PCI Bridge (rev 04)

04:01.0 Ethernet controller: Intel Corporation 82541PI Gigabit Ethernet Controller (rev 05)

I suspect this change in IOMMU-grouping are a result of me updating the motherboard BIOS while I still was running Unraid 6.3.5, but then again; Why wouldn't this have prohibited me running my VM then, as it is now?

 

4. Wrong amount of memory reported

Noticed this now as I already had created this thread. Dashboard reports 65GB of memory. In reality I have 32GB. Have my downgrade magically duplicated my RAM, making my sweet box even more gnarly?:p Seriously; What is happening?

 

 

Sorry for the long post. Trying to be thorough and get everything running as nice as possible.

Suggestions on how to resolve one or more of my numbered problems are more than welcome.

Edit: For the record, I'm running my system with PCIe ACS Override setting enabled. I am under the impression that this is required for me to pass GPU and so forth through to my VM. If I'm wrong, please let me know.

Edited December 22, 20178 yr by blodfjert
another edit

1. i've now bonded the network cards in 'active-backup mode'. Still getting some packet drops, but not experiencing any problems in stability or slow networking.

 

2. DMAR messages are still the same.

 

3. Tried moving my PCIEx1 sound card to another slot (PCIEx4), and now the card is not even available as a passable choice in the VM edit.
Under System devices its recognized as:

07:00.0 Audio device: Creative Labs SB Recon3D (rev 01)

It's grouped with the following devices in IOMMU-group 7:

00:1c.0 PCI bridge: Intel Corporation Sunrise Point-H PCI Express Root Port #1 (rev f1)
00:1c.4 PCI bridge: Intel Corporation Sunrise Point-H PCI Express Root Port #5 (rev f1)

06:00.0 USB controller: ASMedia Technology Inc. Device 1242

I've currently disabled the PCI ACS Override. To my great joy my VM is still working with passthrough on the GPU and other needed peripherals. Think I'll keep this setup for now.

Solved the VM sound by hooking my old Hercules Muse USB-soundcard and passing it through.

 

4. For some weird reason "Installed memory" is still reported wrongly on the dashboard.

Well. I'll leave this hanging. Hopefully i'll get some pointers and tips on these issues as the holidays are over. For now; Merry xmas everyone.