Hello,

I recently switched from a Ryzen 5 3600 to a Ryzen 9 3900x on my unraid server and have been having issues with USB and other device pass through errors. The issues occurs when I run a Virtual Machine with a USB device passed through but it may also occur with other devices passed through as well. I receive loads of "Detected Tx Unit Hang" errors on the ethernet ports and the web traffic pauses every so often. My nvme drives also keep disconnecting which has lead to corruption of a few of my virtual machines. I am unsure what to do as I have tried many settings in my motherboards BIOS and also two different installs of unraid. Unraid reports that HVM and IOMMU are both enabled but I still get these issues. It has worked so far while no devices are passed through to the VM but as soon as i start it (even with just a printer passed through) i get these issues again. I might have to switch back to the Ryzen 5 as this setup is not usable. I have also tried updating my bios to the latest version which made no improvement. Some help would really be appreciated? I can provide more info if required.

I am getting similar issues to the ones in this post:

I am running unraid version 6.9.2

I have more info on the errors I have been experiencing. Here's something from the logs:

Jun 25 17:36:08 Tower kernel: ------------[ cut here ]------------
Jun 25 17:36:08 Tower kernel: NETDEV WATCHDOG: eth0 (igb): transmit queue 0 timed out
Jun 25 17:36:08 Tower kernel: WARNING: CPU: 13 PID: 0 at net/sched/sch_generic.c:442 dev_watchdog+0xcf/0x12b
Jun 25 17:36:08 Tower kernel: Modules linked in: xt_CHECKSUM ipt_REJECT nf_reject_ipv4 ip6table_mangle ip6table_nat iptable_mangle nf_tables vhost_net tun vhost vhost_iotlb tap xt_nat xt_tcpudp veth xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 br_netfilter xfs md_mod ip6table_filter ip6_tables iptable_filter ip_tables x_tables bonding igb i2c_algo_bit ipmi_ssif amd64_edac_mod wmi_bmof edac_mce_amd kvm_amd kvm crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel crypto_simd cryptd glue_helper i2c_piix4 input_leds nvme rapl k10temp ahci i2c_core ch341 nvme_core ccp led_class usbserial usblp acpi_ipmi libahci wmi ipmi_si button acpi_cpufreq [last unloaded: i2c_algo_bit]
Jun 25 17:36:08 Tower kernel: CPU: 13 PID: 0 Comm: swapper/13 Not tainted 5.10.28-Unraid #1
Jun 25 17:36:08 Tower kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./X470D4U, BIOS L4.21 04/15/2021
Jun 25 17:36:08 Tower kernel: RIP: 0010:dev_watchdog+0xcf/0x12b
Jun 25 17:36:08 Tower kernel: Code: 79 b7 00 00 75 38 48 89 ef c6 05 63 79 b7 00 01 e8 79 dd fc ff 44 89 e1 48 89 ee 48 c7 c7 ef 7f de 81 48 89 c2 e8 50 16 10 00 <0f> 0b eb 10 41 ff c4 48 05 40 01 00 00 41 39 f4 75 9d eb 16 48 8b
Jun 25 17:36:08 Tower kernel: RSP: 0018:ffffc90000470ed8 EFLAGS: 00010286
Jun 25 17:36:08 Tower kernel: RAX: 0000000000000000 RBX: ffff888135b30438 RCX: 00000000ffffdfff
Jun 25 17:36:08 Tower kernel: RDX: 00000000000004bd RSI: 0000000000000001 RDI: 0000000000000247
Jun 25 17:36:08 Tower kernel: RBP: ffff888135b30000 R08: 0000000000000000 R09: 00000000ffffdfff
Jun 25 17:36:08 Tower kernel: R10: ffffc90000470d08 R11: ffffc90000470d00 R12: 0000000000000000
Jun 25 17:36:08 Tower kernel: R13: ffffc90000470f10 R14: ffffc90000470f10 R15: ffffffff820060c8
Jun 25 17:36:08 Tower kernel: FS: 0000000000000000(0000) GS:ffff8897deb40000(0000) knlGS:0000000000000000
Jun 25 17:36:08 Tower kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jun 25 17:36:08 Tower kernel: CR2: 00007fba2c19f000 CR3: 00000001672c0000 CR4: 0000000000350ee0
Jun 25 17:36:08 Tower kernel: Call Trace:
Jun 25 17:36:08 Tower kernel: <IRQ>
Jun 25 17:36:08 Tower kernel: call_timer_fn.isra.0+0x12/0x6f
Jun 25 17:36:08 Tower kernel: ? netif_tx_lock+0x7a/0x7a
Jun 25 17:36:08 Tower kernel: __run_timers.part.0+0x144/0x185
Jun 25 17:36:08 Tower kernel: ? update_process_times+0x68/0x6e
Jun 25 17:36:08 Tower kernel: ? tick_sched_timer+0x36/0x64
Jun 25 17:36:08 Tower kernel: run_timer_softirq+0x21/0x43
Jun 25 17:36:08 Tower kernel: __do_softirq+0xc4/0x1c2
Jun 25 17:36:08 Tower kernel: asm_call_irq_on_stack+0x12/0x20
Jun 25 17:36:08 Tower kernel: </IRQ>
Jun 25 17:36:08 Tower kernel: do_softirq_own_stack+0x2c/0x39
Jun 25 17:36:08 Tower kernel: __irq_exit_rcu+0x45/0x80
Jun 25 17:36:08 Tower kernel: sysvec_apic_timer_interrupt+0x87/0x95
Jun 25 17:36:08 Tower kernel: asm_sysvec_apic_timer_interrupt+0x12/0x20
Jun 25 17:36:08 Tower kernel: RIP: 0010:arch_local_irq_enable+0x7/0x8
Jun 25 17:36:08 Tower kernel: Code: 00 48 83 c4 28 4c 89 e0 5b 5d 41 5c 41 5d 41 5e 41 5f c3 9c 58 0f 1f 44 00 00 c3 fa 66 0f 1f 44 00 00 c3 fb 66 0f 1f 44 00 00 <c3> 55 8b af 28 04 00 00 b8 01 00 00 00 45 31 c9 53 45 31 d2 39 c5
Jun 25 17:36:08 Tower kernel: RSP: 0018:ffffc90000197ea0 EFLAGS: 00000246
Jun 25 17:36:08 Tower kernel: RAX: ffff8897deb62380 RBX: 0000000000000002 RCX: 000000000000001f
Jun 25 17:36:08 Tower kernel: RDX: 0000000000000000 RSI: 0000000021af2900 RDI: 0000000000000000
Jun 25 17:36:08 Tower kernel: RBP: ffff888105e0a400 R08: 00000045b37784d7 R09: 0000000000000000
Jun 25 17:36:08 Tower kernel: R10: 0000000000002a7d R11: 071c71c71c71c71c R12: 00000045b37784d7
Jun 25 17:36:08 Tower kernel: R13: ffffffff820c8c40 R14: 0000000000000002 R15: 0000000000000000
Jun 25 17:36:08 Tower kernel: cpuidle_enter_state+0x101/0x1c4
Jun 25 17:36:08 Tower kernel: cpuidle_enter+0x25/0x31
Jun 25 17:36:08 Tower kernel: do_idle+0x1a6/0x214
Jun 25 17:36:08 Tower kernel: cpu_startup_entry+0x18/0x1a
Jun 25 17:36:08 Tower kernel: secondary_startup_64_no_verify+0xb0/0xbb
Jun 25 17:36:08 Tower kernel: ---[ end trace e33252e23b1dde1d ]---

I am also seeing these errors for my network ports and nvme drives:

igb 0000:24:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xbfaeef80 flags=0x0000]

Starting to think that maybe my CPU or RAM is broken

My RAM appears fine. Passed a memtest86 test with no errors. I ran an OCCT cpu test on my processor and it came back with no errors after running for a while but I doubt that's conclusive seeing as I'm still having issues. I'm going to test this with a new motherboard and if that doesn't work, possibly look at contacting AMD or look for an alternative to unraid.

Tried this on a completely different motherboard, less issues at first, but after starting all of my VMs I started getting loads of "Detected Hardware Unit Hang" errors and my nvme drives once again stopped working with the error: "nvme nvme0: controller is down; will reset: CSTS=0x3, PCI_STATUS=0x10" on both nvme drives. Considering the memory test passed fine, and I have tried two different motherboards, I am more certain it's an issue with the CPU. Other people seem to have had no issues with running the 3900X with unraid so I doubt unraids the cause, so I'll contact AMD to try and resolve this.