Intel I350-T4 Errors

bigjme · June 26, 2025, 8:41am

Hi All,

First off, this is a follow up post to my post on the pfSense forum as i want to ensure i have not missed anything

So the elaborate, i am 99% sure that i have a faulty I350 but as this is my second unit after an initial RMA for the same issue, i want to ensure i have not missed anything

My system specs are as follows:
Cpu: AMD 9700X
Mobo: MAG B850 TOMAHAWK MAX WIFI
Memory: 128GB
OS: Scale 24.10.2.2

This is a virtualised pfSense router with all 4 ports of the I350 passed through. The system works flawlessly for around 6 to 9 hours then starts throwing trillions of errors on all 4 interfaces at the same time and all connections are lost but the VM is still up and responds on the console

The previous card had the same issue which makes me think it is a faulty unit but i want to check i have not missed anything.

(i have tried uploading an image to show it but that is not allowed)

I have swapped the card for a known working I340-T4 and that works flawlessly for weeks with no problems under the exact same config and settings

Is anyone aware of any known issues or something i may have missed with this that could be causing the issue?

Regards,
Jamie

SmallBarky · June 26, 2025, 7:25pm

If you need to do images, browse some other forum threads and do the Bot Tutorial, if you haven’t yet to get your trust level up.

TrueNAS-Bot
Type this in a new reply and send to bring up the tutorial, if you haven’t done it already.

@TrueNAS-Bot start tutorial

You might want to post a link, once you can, to the pfsense post.

bigjme · June 26, 2025, 8:25pm

Many thanks!

The pfsense post is here:

For consistency, the image is also as follows:

bigjme · June 30, 2025, 8:56am

I have pushed a few updates over on the netgate forum for anyone that may be following this, if there are any of the images i have posted that you feel may be relevant to include here for consistency, please let me know

bigjme · June 30, 2025, 9:24pm

Ok so as suggested on the netgate forum, i have removed the pci passthrough for now and have the vm configured with virtual nics passing through from the host directly with them simply unconfigured

One issue i have and this is likely more for over here, with this approach, CARP addresses do not work via the virtual nic so i can’t really test much besides leaving it to just run

I’m assuming this is due to the way carp works. I do have
Trust Guest Filters enabled on all the interfaces but that does not seem to help

Is anyone here aware of a way i can get this working to at least debug this properly?

Edit got it, needed to add each interface as a bridge to allow it to work. Now i can test it properly

bigjme · July 2, 2025, 6:44pm

Sorry to keep posting updates to my own topic!

Latest is that the nic does the same when on Truenas directly. I unloaded the igb drivers and reloaded them and saw this in dmesg

[168683.759683] igb 0000:05:00.3 enp5s0f3: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168685.039818] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168685.807653] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168687.087693] igb 0000:05:00.2 enp5s0f2: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168687.855687] igb 0000:05:00.3 enp5s0f3: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168688.879965] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168689.903630] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168690.927699] igb 0000:05:00.2 enp5s0f2: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168691.951708] igb 0000:05:00.3 enp5s0f3: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168692.975871] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168693.999640] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168695.023693] igb 0000:05:00.2 enp5s0f2: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168695.791663] igb 0000:05:00.3 enp5s0f3: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168696.815772] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168697.839653] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168698.863696] igb 0000:05:00.2 enp5s0f2: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168699.887665] igb 0000:05:00.3 enp5s0f3: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168700.437650] igb 0000:05:00.3: removed PHC on enp5s0f3
[168700.911656] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168701.935637] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168702.179403] igb 0000:05:00.3 enp5s0f3 (unregistering): left allmulticast mode
[168702.179406] igb 0000:05:00.3 enp5s0f3 (unregistering): left promiscuous mode
[168702.468717] igb 0000:05:00.2: removed PHC on enp5s0f2
[168702.959649] igb 0000:05:00.2 enp5s0f2: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168704.702443] igb 0000:05:00.2 enp5s0f2 (unregistering): left allmulticast mode
[168704.702447] igb 0000:05:00.2 enp5s0f2 (unregistering): left promiscuous mode
[168704.952707] igb 0000:05:00.1: removed PHC on enp5s0f1
[168705.007729] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168705.775759] igb 0000:05:00.1 enp5s0f1: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168707.776784] igb 0000:05:00.0: removed PHC on enp5s0f0
[168708.847740] igb 0000:05:00.0 enp5s0f0: malformed Tx packet detected and dropped, LVMMC:0xffffffff
[168710.590634] igb 0000:05:00.0 enp5s0f0 (unregistering): left allmulticast mode
[168710.590637] igb 0000:05:00.0 enp5s0f0 (unregistering): left promiscuous mode
[168712.799086] igb: Intel(R) Gigabit Ethernet Network Driver
[168712.799089] igb: Copyright (c) 2007-2014 Intel Corporation.
[168712.799121] igb 0000:05:00.0: enabling device (0000 -> 0002)
[168712.799290] igb 0000:05:00.0 0000:05:00.0 (uninitialized): PCIe link lost
[168712.799522] igb: Failed to read reg 0x18!
[168712.799610] WARNING: CPU: 2 PID: 1106121 at drivers/net/ethernet/intel/igb/igb_main.c:745 igb_rd32+0x7c/0x90 [igb]
[168712.799706] Modules linked in: igb(E+) vhost_net(E) tun(E) vhost(E) vhost_iotlb(E) macvtap(E) macvlan(E) tap(E) xt_nat(E) xt_tcpudp(E) veth(E) xt_conntrack(E) nft_chain_nat(E) xt_MASQUERADE(E) nf_nat(E) nf_conntrack_netlink(E) nf_conntrack(E) nf_defrag_ipv6(E) nf_defrag_ipv4(E) xfrm_user(E) xt_addrtype(E) nft_compat(E) nf_tables(E) libcrc32c(E) crc32c_generic(E) nfnetlink(E) br_netfilter(E) bridge(E) stp(E) llc(E) nvme_fabrics(E) overlay(E) sunrpc(E) binfmt_misc(E) ntb_netdev(E) ntb_transport(E) ntb_split(E) ntb(E) ioatdma(E) ib_core(E) edac_mce_amd(E) kvm_amd(E) kvm(E) irqbypass(E) ghash_clmulni_intel(E) sha512_ssse3(E) sha256_ssse3(E) sha1_ssse3(E) amdgpu(E) aesni_intel(E) crypto_simd(E) cryptd(E) snd_hda_codec_hdmi(E) rapl(E) drm_exec(E) amdxcp(E) drm_buddy(E) gpu_sched(E) drm_suballoc_helper(E) drm_display_helper(E) snd_hda_intel(E) cec(E) snd_usb_audio(E) snd_intel_dspcfg(E) snd_usbmidi_lib(E) rc_core(E) snd_hda_codec(E) snd_rawmidi(E) snd_seq_device(E) snd_hda_core(E) mc(E) snd_hwdep(E) drm_ttm_helper(E)
[168712.799722]  snd_pcm(E) wmi_bmof(E) ttm(E) snd_timer(E) sp5100_tco(E) pcspkr(E) drm_kms_helper(E) snd(E) k10temp(E) watchdog(E) soundcore(E) ccp(E) cdc_mbim(E) joydev(E) cdc_wdm(E) sg(E) button(E) evdev(E) loop(E) drm(E) efi_pstore(E) configfs(E) ip_tables(E) x_tables(E) autofs4(E) zfs(POE) spl(OE) efivarfs(E) sr_mod(E) cdrom(E) hid_generic(E) cdc_ncm(E) cdc_ether(E) usbnet(E) usbhid(E) uas(E) hid(E) usb_storage(E) mii(E) sd_mod(E) ahci(E) nvme(E) ahciem(E) xhci_pci(E) libahci(E) xhci_hcd(E) ixgbe(E) nvme_core(E) libata(E) xfrm_algo(E) t10_pi(E) crc32_pclmul(E) mdio_devres(E) usbcore(E) scsi_mod(E) crc32c_intel(E) crc64_rocksoft(E) i2c_piix4(E) usb_common(E) scsi_common(E) crc64(E) crc_t10dif(E) i2c_algo_bit(E) libphy(E) dca(E) crct10dif_generic(E) mdio(E) crct10dif_pclmul(E) crct10dif_common(E) video(E) wmi(E) gpio_amdpt(E) gpio_generic(E) [last unloaded: igb(E)]
[168712.800964] RIP: 0010:igb_rd32+0x7c/0x90 [igb]
[168712.803497]  ? igb_rd32+0x7c/0x90 [igb]
[168712.803880]  ? igb_rd32+0x7c/0x90 [igb]
[168712.805033]  ? igb_rd32+0x7c/0x90 [igb]
[168712.805228]  ? igb_rd32+0x7c/0x90 [igb]
[168712.805422]  igb_get_invariants_82575+0xa6/0xf00 [igb]
[168712.805620]  igb_probe+0x3be/0x1520 [igb]
[168712.807796]  ? __pfx_igb_init_module+0x10/0x10 [igb]
[168713.136674] igb 0000:05:00.0: PHY reset is blocked due to SOL/IDER session.
[168714.866546] igb 0000:05:00.0: The NVM Checksum Is Not Valid
[168715.129000] igb: probe of 0000:05:00.0 failed with error -5
[168715.129479] igb 0000:05:00.1: enabling device (0000 -> 0002)
[168715.129619] igb 0000:05:00.1 0000:05:00.1 (uninitialized): PCIe link lost
[168715.130133] igb: Failed to read reg 0x18!
[168715.130381] WARNING: CPU: 10 PID: 1106121 at drivers/net/ethernet/intel/igb/igb_main.c:745 igb_rd32+0x7c/0x90 [igb]
[168715.130642] Modules linked in: igb(E+)
[168715.130922]  snd_timer(E) sp5100_tco(E) pcspkr(E) drm_kms_helper(E) snd(E) k10temp(E) watchdog(E) soundcore(E) ccp(E) cdc_mbim(E) joydev(E) cdc_wdm(E) sg(E) button(E) evdev(E) loop(E) drm(E) efi_pstore(E) configfs(E) ip_tables(E) x_tables(E) autofs4(E) zfs(POE) spl(OE) efivarfs(E) sr_mod(E) cdrom(E) hid_generic(E) cdc_ncm(E) cdc_ether(E) usbnet(E) usbhid(E) uas(E) hid(E) usb_storage(E) mii(E) sd_mod(E) ahci(E) nvme(E) ahciem(E) xhci_pci(E) libahci(E) xhci_hcd(E) ixgbe(E) nvme_core(E) libata(E) xfrm_algo(E) t10_pi(E) crc32_pclmul(E) mdio_devres(E) usbcore(E) scsi_mod(E) crc32c_intel(E) crc64_rocksoft(E) i2c_piix4(E) usb_common(E) scsi_common(E) crc64(E) crc_t10dif(E) i2c_algo_bit(E) libphy(E) dca(E) crct10dif_generic(E) mdio(E) crct10dif_pclmul(E) crct10dif_common(E) video(E) wmi(E) gpio_amdpt(E) gpio_generic(E) [last unloaded: igb(E)]
[168715.133023] RIP: 0010:igb_rd32+0x7c/0x90 [igb]
[168715.137255]  ? igb_rd32+0x7c/0x90 [igb]
[168715.137508]  ? igb_rd32+0x7c/0x90 [igb]
[168715.138247]  ? igb_rd32+0x7c/0x90 [igb]
[168715.138250]  ? igb_rd32+0x7c/0x90 [igb]
[168715.139694]  igb_get_invariants_82575+0xa6/0xf00 [igb]
[168715.139700]  igb_probe+0x3be/0x1520 [igb]
[168715.142610]  ? __pfx_igb_init_module+0x10/0x10 [igb]
[168715.472667] igb 0000:05:00.1: PHY reset is blocked due to SOL/IDER session.
[168717.202621] igb 0000:05:00.1: The NVM Checksum Is Not Valid
[168717.400883] igb: probe of 0000:05:00.1 failed with error -5
[168717.401370] igb 0000:05:00.2: enabling device (0000 -> 0002)
[168717.401509] igb 0000:05:00.2 0000:05:00.2 (uninitialized): PCIe link lost
[168717.402106] igb: Failed to read reg 0x18!
[168717.402357] WARNING: CPU: 2 PID: 1106121 at drivers/net/ethernet/intel/igb/igb_main.c:745 igb_rd32+0x7c/0x90 [igb]
[168717.402619] Modules linked in: igb(E+) vhost_net(E) tun(E) vhost(E) vhost_iotlb(E) macvtap(E) macvlan(E) tap(E) xt_nat(E) xt_tcpudp(E) veth(E) xt_conntrack(E) nft_chain_nat(E) xt_MASQUERADE(E) nf_nat(E) nf_conntrack_netlink(E) nf_conntrack(E) nf_defrag_ipv6(E) nf_defrag_ipv4(E) xfrm_user(E) xt_addrtype(E) nft_compat(E) nf_tables(E) libcrc32c(E) crc32c_generic(E) nfnetlink(E) br_netfilter(E) bridge(E) stp(E) llc(E) nvme_fabrics(E) overlay(E) sunrpc(E) binfmt_misc(E) ntb_netdev(E) ntb_transport(E) ntb_split(E) ntb(E) ioatdma(E) ib_core(E) edac_mce_amd(E) kvm_amd(E) kvm(E) irqbypass(E) ghash_clmulni_intel(E) sha512_ssse3(E) sha256_ssse3(E) sha1_ssse3(E) amdgpu(E) aesni_intel(E) crypto_simd(E) cryptd(E) snd_hda_codec_hdmi(E) rapl(E) drm_exec(E) amdxcp(E) drm_buddy(E) gpu_sched(E) drm_suballoc_helper(E) drm_display_helper(E) snd_hda_intel(E) cec(E) snd_usb_audio(E) snd_intel_dspcfg(E) snd_usbmidi_lib(E) rc_core(E) snd_hda_codec(E) snd_rawmidi(E) snd_seq_device(E) snd_hda_core(E) mc(E) snd_hwdep(E) drm_ttm_helper(E)
[168717.402644]  snd_pcm(E) wmi_bmof(E) ttm(E) snd_timer(E) sp5100_tco(E) pcspkr(E) drm_kms_helper(E) snd(E) k10temp(E) watchdog(E) soundcore(E) ccp(E) cdc_mbim(E) joydev(E) cdc_wdm(E) sg(E) button(E) evdev(E) loop(E) drm(E) efi_pstore(E) configfs(E) ip_tables(E) x_tables(E) autofs4(E) zfs(POE) spl(OE) efivarfs(E) sr_mod(E) cdrom(E) hid_generic(E) cdc_ncm(E) cdc_ether(E) usbnet(E) usbhid(E) uas(E) hid(E) usb_storage(E) mii(E) sd_mod(E) ahci(E) nvme(E) ahciem(E) xhci_pci(E) libahci(E) xhci_hcd(E) ixgbe(E) nvme_core(E) libata(E) xfrm_algo(E) t10_pi(E) crc32_pclmul(E) mdio_devres(E) usbcore(E) scsi_mod(E) crc32c_intel(E) crc64_rocksoft(E) i2c_piix4(E) usb_common(E) scsi_common(E) crc64(E) crc_t10dif(E) i2c_algo_bit(E) libphy(E) dca(E) crct10dif_generic(E) mdio(E) crct10dif_pclmul(E) crct10dif_common(E) video(E) wmi(E) gpio_amdpt(E) gpio_generic(E) [last unloaded: igb(E)]
[168717.405253] RIP: 0010:igb_rd32+0x7c/0x90 [igb]
[168717.408900]  ? igb_rd32+0x7c/0x90 [igb]
[168717.409393]  ? igb_rd32+0x7c/0x90 [igb]
[168717.410845]  ? igb_rd32+0x7c/0x90 [igb]
[168717.411088]  ? igb_rd32+0x7c/0x90 [igb]
[168717.411330]  igb_get_invariants_82575+0xa6/0xf00 [igb]
[168717.411577]  igb_probe+0x3be/0x1520 [igb]
[168717.414240]  ? __pfx_igb_init_module+0x10/0x10 [igb]
[168717.744687] igb 0000:05:00.2: PHY reset is blocked due to SOL/IDER session.
[168719.474678] igb 0000:05:00.2: The NVM Checksum Is Not Valid
[168719.664847] igb: probe of 0000:05:00.2 failed with error -5
[168719.665336] igb 0000:05:00.3: enabling device (0000 -> 0002)
[168719.665475] igb 0000:05:00.3 0000:05:00.3 (uninitialized): PCIe link lost
[168719.666111] igb: Failed to read reg 0x18!
[168719.666361] WARNING: CPU: 10 PID: 1106121 at drivers/net/ethernet/intel/igb/igb_main.c:745 igb_rd32+0x7c/0x90 [igb]
[168719.666623] Modules linked in: igb(E+) vhost_net(E) tun(E) vhost(E) vhost_iotlb(E) macvtap(E) macvlan(E) tap(E) xt_nat(E) xt_tcpudp(E) veth(E) xt_conntrack(E) nft_chain_nat(E) xt_MASQUERADE(E) nf_nat(E) nf_conntrack_netlink(E) nf_conntrack(E) nf_defrag_ipv6(E) nf_defrag_ipv4(E) xfrm_user(E) xt_addrtype(E) nft_compat(E) nf_tables(E) libcrc32c(E) crc32c_generic(E) nfnetlink(E) br_netfilter(E) bridge(E) stp(E) llc(E) nvme_fabrics(E) overlay(E) sunrpc(E) binfmt_misc(E) ntb_netdev(E) ntb_transport(E) ntb_split(E) ntb(E) ioatdma(E) ib_core(E) edac_mce_amd(E) kvm_amd(E) kvm(E) irqbypass(E) ghash_clmulni_intel(E) sha512_ssse3(E) sha256_ssse3(E) sha1_ssse3(E) amdgpu(E) aesni_intel(E) crypto_simd(E) cryptd(E) snd_hda_codec_hdmi(E) rapl(E) drm_exec(E) amdxcp(E) drm_buddy(E) gpu_sched(E) drm_suballoc_helper(E) drm_display_helper(E) snd_hda_intel(E) cec(E) snd_usb_audio(E) snd_intel_dspcfg(E) snd_usbmidi_lib(E) rc_core(E) snd_hda_codec(E) snd_rawmidi(E) snd_seq_device(E) snd_hda_core(E) mc(E) snd_hwdep(E) drm_ttm_helper(E)
[168719.666648]  snd_pcm(E) wmi_bmof(E) ttm(E) snd_timer(E) sp5100_tco(E) pcspkr(E) drm_kms_helper(E) snd(E) k10temp(E) watchdog(E) soundcore(E) ccp(E) cdc_mbim(E) joydev(E) cdc_wdm(E) sg(E) button(E) evdev(E) loop(E) drm(E) efi_pstore(E) configfs(E) ip_tables(E) x_tables(E) autofs4(E) zfs(POE) spl(OE) efivarfs(E) sr_mod(E) cdrom(E) hid_generic(E) cdc_ncm(E) cdc_ether(E) usbnet(E) usbhid(E) uas(E) hid(E) usb_storage(E) mii(E) sd_mod(E) ahci(E) nvme(E) ahciem(E) xhci_pci(E) libahci(E) xhci_hcd(E) ixgbe(E) nvme_core(E) libata(E) xfrm_algo(E) t10_pi(E) crc32_pclmul(E) mdio_devres(E) usbcore(E) scsi_mod(E) crc32c_intel(E) crc64_rocksoft(E) i2c_piix4(E) usb_common(E) scsi_common(E) crc64(E) crc_t10dif(E) i2c_algo_bit(E) libphy(E) dca(E) crct10dif_generic(E) mdio(E) crct10dif_pclmul(E) crct10dif_common(E) video(E) wmi(E) gpio_amdpt(E) gpio_generic(E) [last unloaded: igb(E)]
[168719.669264] RIP: 0010:igb_rd32+0x7c/0x90 [igb]
[168719.672935]  ? igb_rd32+0x7c/0x90 [igb]
[168719.673429]  ? igb_rd32+0x7c/0x90 [igb]
[168719.674884]  ? igb_rd32+0x7c/0x90 [igb]
[168719.675127]  ? igb_rd32+0x7c/0x90 [igb]
[168719.675369]  igb_get_invariants_82575+0xa6/0xf00 [igb]
[168719.675616]  igb_probe+0x3be/0x1520 [igb]
[168719.678284]  ? __pfx_igb_init_module+0x10/0x10 [igb]
[168720.008929] igb 0000:05:00.3: PHY reset is blocked due to SOL/IDER session.
[168721.738903] igb 0000:05:00.3: The NVM Checksum Is Not Valid
[168721.916849] igb: probe of 0000:05:00.3 failed with error -5

I have moved the card from a pci 3.0 1x slot to a gen 5 16x slot just in case its a pci slot issue but based on the above, i’m thinking its a second faulty nic

Regards,
Jamie

SmallBarky · July 2, 2025, 6:54pm

Replying so you’re not just talking to yourself.
Are you getting these cards from a good source. Two in a row sounds like a knockoff type problem. It’s just

bigjme · July 2, 2025, 7:14pm

haha, very kind of you!

Both cards were from the same source with the second being a replacement after my initial RMA, so i’m thinking the company likely got them from the same location

I have checked the cards physically and they seem legit but you just never know with these cards

I would post the company but they have been really good and understanding with this so i don’t think it would be fair on them but at the same time, i don’t want others going through the same!

The company themselves are refurbishers/resellers and have a pretty good reputation as a whole but you never know where they got the original stock from

shoulders · July 2, 2025, 10:21pm

They might not know the cards are knock offs.

There are tell tale signs. I will see if I have a link to an article telling you what to look for.

Might not be exactly the same card, but close enough.

shoulders · July 3, 2025, 5:24pm

This is an in-depth article on telling the difference between real and fake.

bigjme · July 3, 2025, 6:14pm

Many thanks

Looking through and comparing it to one of the cards i have:

Heatsink is larger
Delta chips look right but the lower number is 143101B?
Crystals are TXC and look 100% correct
The small inductors are both 3R3 430 VB1 - Possibly fake?
Large inductor is a different model (EE 831-03362F 1425 CN2) - Possibly fake but close?

Besides those, it seems correct but i’m unsure if their just a parts change on a revision or an actual issue

shoulders · July 3, 2025, 6:25pm

the big one i would guess, is delta engraved or written on.

bigjme · July 3, 2025, 6:34pm

I can confirm that Delta is engraved and i can feel it with my nail

truenas-fan · July 6, 2025, 6:22pm

I don’t think this is a hardware issue (put a fan blowing on the card, to ensure is not heat related).

But do love non-electronic people giving their illogical theories!
The cheap cards use the same main components and simply save/skip on a few non-critical elements.

I have used the cheap 4 port Chinese cards and only one went bad, possibly due to the inrush current flaw, but it failed fully so I just exchanged it for another card, that never failed for the next 5-7 years!

Ensure hardware offloading is turned off.
Do high-speed transfers to see if you can make it fail on-command.

bigjme · July 6, 2025, 6:39pm

Its currently at 4 days in the x16 slot so it may genuinely be an incompatibility with the i350 and 1x, vs i340 which are known to work in a 1x slot (i found this out today)

The unit is in a server case with 5 80mm fans blowing at it in an air conditioned 19 degree room, the unit itself is cool and never gets warm

I did also test a mixture of completely idles besides a few pings for 12 hours or more, constant speedtests and iperfs at gigabit speed on at least 2 ports at a time pushing hundreds of gigabytes through, nothing was able to trigger it directly, it was random in itself

At the moment it is looking like it may be related to the way SRI-OV and the additional features on this card function with the slot and it may simply be that it can not negotiate down to 1x and still do everything it needs so the firmware on the card goes haywire

I did not see anything when i was looking before ordering the card that this could be an issue and to my knowledge, it should fairly gracefully work with a 1x all be it bandwidth limited but this does not seem to be the case for this model mixed the an AM5 board

bigjme · July 18, 2025, 11:40am

For anyone interested in the exciting conclusions… it worked fine in the 16x slot for 2 weeks and is still in there now
I put an I340-T4 in the 1x slot at the same time and left that running and that has been perfectly fine as well

It seems to be an incompatibility between the 1x slot and the I350 specifically but i’m not sure why. In either case, the issue seems to be resolved

It may be something specific to AM5 and the I350 in the 1x, or just the I350 and the 1x alone but if anyone else for some reason tries the same, at least you know what symptoms manifest and what the cause was

Thanks again for those that helped and commented