Hey everyone. My server had been solid up until recently, when I began experiencing new problems. Most importantly, the web UI stops working after a few days of uptime. Secondarily — and perhaps this is related — it's outputting messages I've never seen before, such as:

May 25 17:10:12 ANDRAS4 kernel: Call Trace:
May 25 17:10:12 ANDRAS4 kernel: lookup_fast+0x1d2/0x27a
May 25 17:10:12 ANDRAS4 kernel: path_openat+0x2b6/0xc07
May 25 17:10:12 ANDRAS4 kernel: ? filename_lookup.part.16+0xa5/0xcc
May 25 17:10:12 ANDRAS4 kernel: do_filp_open+0x4c/0xa9
May 25 17:10:12 ANDRAS4 kernel: ? _copy_to_user+0x22/0x28
May 25 17:10:12 ANDRAS4 kernel: do_sys_open+0x132/0x1ce
May 25 17:10:12 ANDRAS4 kernel: do_syscall_64+0x57/0xe6
May 25 17:10:12 ANDRAS4 kernel: entry_SYSCALL_64_after_hwframe+0x44/0xa9
May 25 17:10:12 ANDRAS4 kernel: RIP: 0033:0x145931420380
May 25 17:10:12 ANDRAS4 kernel: Code: 25 00 00 41 00 3d 00 00 41 00 74 36 48 8d 05 87 c3 0d 00 8b 00 85 c0 75 5a 89 f2 b8 01 01 00 00 48 89 fe bf 9c ff ff ff 0f 05 <48> 3d 00 f0 ff ff 0f 87 84 00 00 00 48 83 c4 68 5b 5d c3 0f 1f 44 
May 25 17:10:12 ANDRAS4 kernel: RSP: 002b:00007ffcdba6f960 EFLAGS: 00000246 ORIG_RAX: 0000000000000101
May 25 17:10:12 ANDRAS4 kernel: RAX: ffffffffffffffda RBX: 0000000000420f23 RCX: 0000145931420380
May 25 17:10:12 ANDRAS4 kernel: RDX: 0000000000000000 RSI: 0000000000664860 RDI: 00000000ffffff9c
May 25 17:10:12 ANDRAS4 kernel: RBP: 0000000000628270 R08: 0000000000000000 R09: 00000000ffffffff
May 25 17:10:12 ANDRAS4 kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000008
May 25 17:10:12 ANDRAS4 kernel: R13: 0000000000420f23 R14: 0000000000000000 R15: 00001459314fe4c0
May 25 17:10:12 ANDRAS4 kernel: Modules linked in: dm_mod dax xt_CHECKSUM iptable_mangle ipt_REJECT ebtable_filter ebtables ip6table_filter ip6_tables vhost_net tun vhost tap veth ipt_MASQUERADE iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 iptable_filter ip_tables nf_nat xfs md_mod it87 hwmon_vid igb i2c_algo_bit alx mdio edac_mce_amd kvm_amd kvm crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel pcbc aesni_intel aes_x86_64 crypto_simd cryptd k10temp ahci i2c_piix4 glue_helper i2c_core libahci wmi_bmof mxm_wmi wmi ccp button pcc_cpufreq acpi_cpufreq [last unloaded: i2c_algo_bit]
May 25 17:10:12 ANDRAS4 kernel: ---[ end trace 08e156b27250f0bc ]---
May 25 17:10:12 ANDRAS4 kernel: RIP: 0010:__d_lookup_rcu+0x5a/0x12f
May 25 17:10:12 ANDRAS4 kernel: Code: 89 ea 41 89 ee d3 ea 48 8d 04 d0 48 8b 18 48 89 e8 48 c1 e8 20 48 89 04 24 48 83 e3 fe 48 85 db 0f 84 c4 00 00 00 4c 8d 6b f8 <44> 8b 63 fc 4c 39 7b 10 0f 85 aa 00 00 00 48 83 7b 08 00 0f 84 9f 
May 25 17:10:12 ANDRAS4 kernel: RSP: 0018:ffffc90003f13bc0 EFLAGS: 00010282
May 25 17:10:12 ANDRAS4 kernel: RAX: 000000000000000c RBX: ffef88028cd56008 RCX: 000000000000000a
May 25 17:10:12 ANDRAS4 kernel: RDX: 000000000006f25d RSI: ffffc90003f13d50 RDI: ffff88028fcc3440
May 25 17:10:12 ANDRAS4 kernel: RBP: 0000000c1bc97651 R08: ffffc90003f13d50 R09: 8e04502e0637f6d7
May 25 17:10:12 ANDRAS4 kernel: R10: ffffc90003f13c2c R11: 8080808080808080 R12: ffffc90003f13c88
May 25 17:10:12 ANDRAS4 kernel: R13: ffef88028cd56000 R14: 000000001bc97651 R15: ffff88028fcc3440
May 25 17:10:12 ANDRAS4 kernel: FS:  00001459314fe540(0000) GS:ffff88082ec40000(0000) knlGS:0000000000000000
May 25 17:10:12 ANDRAS4 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
May 25 17:10:12 ANDRAS4 kernel: CR2: 00000000006ab000 CR3: 0000000282d2e000 CR4: 00000000003406e0
May 25 17:10:27 ANDRAS4 kernel: general protection fault: 0000 [#19] SMP NOPTI
May 25 17:10:27 ANDRAS4 kernel: CPU: 0 PID: 13062 Comm: lsof Tainted: G      D           4.18.20-unRAID #1
May 25 17:10:27 ANDRAS4 kernel: Hardware name: Gigabyte Technology Co., Ltd. AX370-Gaming K7/AX370-Gaming K7, BIOS F10 12/07/2017
May 25 17:10:27 ANDRAS4 kernel: RIP: 0010:__d_lookup+0x3e/0x12b
May 25 17:10:27 ANDRAS4 kernel: Code: 51 44 8b 26 8b 0d fd 58 d5 00 48 8b 05 ee 58 d5 00 44 89 e2 d3 ea 48 8d 04 d0 48 8b 18 48 83 e3 fe 48 85 db 0f 84 dc 00 00 00 <44> 39 63 18 0f 85 ca 00 00 00 48 8d 43 50 48 89 c7 48 89 04 24 e8 
May 25 17:10:27 ANDRAS4 kernel: RSP: 0018:ffffc90039777c80 EFLAGS: 00010206
May 25 17:10:27 ANDRAS4 kernel: RAX: ffffc9000047b378 RBX: 0010000000000000 RCX: 000000000000000a
May 25 17:10:27 ANDRAS4 kernel: RDX: 000000000006f46f RSI: ffffc90039777df0 RDI: ffff8807b58db980
May 25 17:10:27 ANDRAS4 kernel: RBP: ffff8808071f9800 R08: 61c8864680b583eb R09: ffff8807b58db980
May 25 17:10:27 ANDRAS4 kernel: R10: 0000000000000000 R11: 8080808080808080 R12: 000000001bd1bfbc
May 25 17:10:27 ANDRAS4 kernel: R13: ffffffffffffffff R14: ffffc90039777df0 R15: ffff8807b58db980
May 25 17:10:27 ANDRAS4 kernel: FS:  000014583c9ce540(0000) GS:ffff88082ec00000(0000) knlGS:0000000000000000
May 25 17:10:27 ANDRAS4 kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
May 25 17:10:27 ANDRAS4 kernel: CR2: 00000000006ab000 CR3: 00000007b8c96000 CR4: 00000000003406f0

My diagnostics file is attached. Could someone take a look and let me know what's going on? Thanks!

andras4-diagnostics-20190525-1645.zip

Are you using the Ryzen workarounds?

https://forums.unraid.net/topic/80006-random-crashes-restarts/?do=findComment&comment=742911

 

13 hours ago, johnnie.black said:

Are you using the Ryzen workarounds?

https://forums.unraid.net/topic/80006-random-crashes-restarts/?do=findComment&comment=742911

 

 

I believe I've done all the workarounds. And, as I said, the system was stable with uptimes of several months until just recently. Here are the AMD optimizations I did. At least, all the ones I can remember. I should have made a note.

 

BIOS:

Global C-States set to DISABLED

Power Supply Idle set to TYPICAL

SVM set to ENABLE

 

UnRaid:

Set rcu_nocbs kernel parameter

Edited May 26, 20197 yr by cyberspectre

Try safe mode without any dockers/VMs for a few days, if it still stops responding it's likely hardware related.

15 hours ago, johnnie.black said:

Try safe mode without any dockers/VMs for a few days, if it still stops responding it's likely hardware related.

Any idea what hardware I need to replace if it does?

Sure enough, it happened again with no VMs or Dockers running. I ran memtest86+ and got no errors in the memory... Is it definitely hardware? What would cause this?

51 minutes ago, cyberspectre said:

Sure enough, it happened again with no VMs or Dockers running. I ran memtest86+ and got no errors in the memory... Is it definitely hardware? What would cause this?

Yikes, just kidding. I ran memtest86 for longer and did indeed get memory errors. Going to try to isolate which module now.