Hi all - just got a kernel mod crash with the new threadripper server (Taichi X399 + 1950X), and it is on the latest 6.4.1-rc1 build. Copying some files remotely to the unraid server via NFS mount point, and suddenly the copy operation hung on the remote file source end, and see following kernel message in the syslog:

 

Jan 31 09:57:39 Ryzen kernel: BUG: unable to handle kernel paging request at 00000000b0000000
Jan 31 09:57:39 Ryzen kernel: IP: handle_stripe+0x606/0x12a4 [md_mod]
Jan 31 09:57:39 Ryzen kernel: PGD fcf3f2067 P4D fcf3f2067 PUD 0
Jan 31 09:57:39 Ryzen kernel: Oops: 0000 [#1] PREEMPT SMP NOPTI
Jan 31 09:57:39 Ryzen kernel: Modules linked in: xt_nat veth xt_CHECKSUM iptable_mangle ipt_REJECT nf_reject_ipv4 ebtable_filter ebtables ip6table_filter ip6_tables vhost_net tun vhost tap ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 iptable_filter ip_tables nf_nat xfs nfsd lockd grace sunrpc md_mod nct6775 hwmon_vid igb ptp pps_core i2c_algo_bit edac_mce_amd kvm_amd kvm crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel pcbc aesni_intel aes_x86_64 crypto_simd glue_helper cryptd i2c_piix4 i2c_core wmi_bmof mxm_wmi ahci libahci ccp wmi button acpi_cpufreq [last unloaded: pps_core]
Jan 31 09:57:39 Ryzen kernel: CPU: 0 PID: 5850 Comm: unraidd Not tainted 4.14.15-unRAID #3
Jan 31 09:57:39 Ryzen kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./X399 Taichi, BIOS P2.00 11/21/2017
Jan 31 09:57:39 Ryzen kernel: task: ffff880ff4c1e200 task.stack: ffffc900072cc000
Jan 31 09:57:39 Ryzen kernel: RIP: 0010:handle_stripe+0x606/0x12a4 [md_mod]
Jan 31 09:57:39 Ryzen kernel: RSP: 0018:ffffc900072cfdc0 EFLAGS: 00010246
Jan 31 09:57:39 Ryzen kernel: RAX: 0000000000000094 RBX: ffff880fc5484a60 RCX: 0000000000000258
Jan 31 09:57:39 Ryzen kernel: RDX: 0000000000000100 RSI: 0000000000000003 RDI: ffff880ff2659268
Jan 31 09:57:39 Ryzen kernel: RBP: 0000000000000000 R08: 0000000000000000 R09: ffffc900072cfdc0
Jan 31 09:57:39 Ryzen kernel: R10: ffffc900072cfcf8 R11: ffffffff810633fc R12: ffff880fc5484b90
Jan 31 09:57:39 Ryzen kernel: R13: 0000000000000007 R14: 0000000000000003 R15: 00000000b0000000
Jan 31 09:57:39 Ryzen kernel: FS:  0000000000000000(0000) GS:ffff880ffc800000(0000) knlGS:0000000000000000
Jan 31 09:57:39 Ryzen kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jan 31 09:57:39 Ryzen kernel: CR2: 00000000b0000000 CR3: 0000000ff36d2000 CR4: 00000000003406f0
Jan 31 09:57:39 Ryzen kernel: Call Trace:
Jan 31 09:57:39 Ryzen kernel: unraidd+0xaf/0xff [md_mod]
Jan 31 09:57:39 Ryzen kernel: ? md_open+0x2c/0x2c [md_mod]
Jan 31 09:57:39 Ryzen kernel: ? md_thread+0xb7/0xc7 [md_mod]
Jan 31 09:57:39 Ryzen kernel: md_thread+0xb7/0xc7 [md_mod]
Jan 31 09:57:39 Ryzen kernel: ? wait_woken+0x68/0x68
Jan 31 09:57:39 Ryzen kernel: kthread+0x111/0x119
Jan 31 09:57:39 Ryzen kernel: ? kthread_create_on_node+0x3a/0x3a
Jan 31 09:57:39 Ryzen kernel: ? do_group_exit+0x95/0x95
Jan 31 09:57:39 Ryzen kernel: ret_from_fork+0x22/0x40
Jan 31 09:57:39 Ryzen kernel: Code: 00 4d 85 ff 0f 84 88 00 00 00 48 63 d6 48 69 d2 c8 00 00 00 48 8b 94 13 30 01 00 00 81 e2 00 03 00 00 48 81 fa 00 01 00 00 75 67 <49> 8b 17 48 85 d2 48 89 94 0b 48 01 00 00 74 02 0f 0b 48 8b 7c
Jan 31 09:57:39 Ryzen kernel: RIP: handle_stripe+0x606/0x12a4 [md_mod] RSP: ffffc900072cfdc0
Jan 31 09:57:39 Ryzen kernel: CR2: 00000000b0000000
Jan 31 09:57:39 Ryzen kernel: ---[ end trace 8303258ddfd0097a ]---
Jan 31 09:57:39 Ryzen kernel: note: unraidd[5850] exited with preempt_count 1

 

Any suggestions?

 

Thanks,

Jerry

1 hour ago, je2ryw said:

Any suggestions?

 

Did you happen to capture diagnostics.zip?  If not and server is still in above state, please do so and post.

Can you make it happen again?

Have you run memtest on this server recently?

The server has been hard reset, and the "reboot" cmd won't work. I will run the memtest once the the parity check completes. And I have the tail cmd running for the syslog file over a screen/ssh session, and next time if it happens again, I will do the diagnostics capture.

 

Thanks,

Jerry

14 minutes ago, je2ryw said:

The server has been hard reset, and the "reboot" cmd won't work. I will run the memtest once the the parity check completes. And I have the tail cmd running for the syslog file over a screen/ssh session, and next time if it happens again, I will do the diagnostics capture.

 

Thanks,

Jerry

 

Perfect.  Thank you!