TrueNAS (VM on Proxmox) Random Reboots & Segfault in libcrypto.so.3 – Possible Cause?

TrueNAS General
1

System Details:

  • Host OS: Proxmox VE 7.4-19 (pve-manager/7.4-19/f98bf8d4)
  • Kernel: Linux 5.15.158-2-pve
  • CPU: 6 x Intel(R) Core™ i5-9600K CPU @ 3.70GHz (1 Socket)
  • Motherboard: ASRock H370M-HDV
  • RAM: Corsair DDR4 Vengeance LPX 2x32GB 2666MHz (64GB total)
  • HBA: LSI SAS 9207-8i
  • HDD: 5 x WD DC HC520 12TB (RAIDZ2 in TrueNAS)
  • Power Supply: Corsair SF750
  • Network Adapter: Realtek (onboard)
  • Virtualization: TrueNAS ElectricEel-24.10.1 running as a VM on Proxmox

Issue:

  • Random reboots of the TrueNAS VM, but /var/log/messages does not show anything relevant.
  • Segfault error in libcrypto.so.3
python3[38163]: segfault at c18 ip 00007fc2114bff88 sp 00007ffff8b43cf8 error 4 in libcrypto.so.3[7fc211352000+27c000] likely on CPU 3 (core 3, socket 0)
Feb 13 20:52:06 truenas kernel: Code: 74 08 ff e0 66 0f 1f 44 00 00 e9 63 64 e9 ff 0f 1f 00 48 8b 05 59 e1 24 00 48 3b 05 7a cc 24 00 74 08 ff e0 66 0f 1f 44 00 00 <48> 85 ff 74 23 8b 05 25 e1 24 00 85 c0 74 0a c7 05 17 e1 24 00 00

Samba-related reboot: The VM rebooted once when I accessed a Samba share from my MacBook.

  1. Could the segfault in libcrypto.so.3 be triggering the VM reboots?
  2. Could the Realtek NIC on the Proxmox host be causing instability? Would switching to an Intel-based NIC help?
  3. What’s the best way to debug unexplained reboots in Truenas if there’s nothing logged in /var/log/messages? (reboots don’t happen if I stop the TrueNas VM)

Any troubleshooting tips or insights would be greatly appreciated!

2

I have performed MemTest86 and it did pass

3 
@TrueNAS-Bot start tutorial
4

That’s quite an outdated version of Proxmox VE. 8.3 is current.

Is the CPU type of the VM set to “host”?

5

Hi, thanks for your reply.

I have followed the default configuration and the default is kvm64. Should I change that to host?

6

these is the configuration
these is the configuration1796×844 105 KB

I have changed the CPU type to host now.

7

Also machine should be q35.

8

Thank you. I will update the settings and see if I have these random reboots and other issues.

Also, I plan on upgrading to the newest version of proxmox.

9

In the OPNsense community we had quite a number of unexplicable crashes in virtual instances that turned out to be caused by CPU bugs and could be fixed by applying the latest microcode updates.

I don’t know about your CPU. All of those were the quite recent N100 and N5105 types.

But I would check. For sure there is a microcode update package for debian. Another reason to update to a current Proxmox release. These packages must be installed and run on the host system.

1 Like
10

I have updated the microcode, and it is stable now (no reboots).

However, I see these errors in the truenas.

Feb 15 14:13:46 truenas kernel: traps: binaryurpWriter[265809] general protection fault ip:7fc657158a7a sp:7fc656b18c70 error:0 in libuno_cppu.so.3[7fc657151000+1f000]
Feb 15 14:57:25 truenas systemd-journald[569]: Data hash table of /var/log/journal/603158a2cbcf420280adf746ae026ec0/system.journal has a fill level at 75.0 (8535 of 11377 items, 6553600 file size, 767 bytes per hash table item), suggesting rotation.
Feb 15 14:57:25 truenas systemd-journald[569]: /var/log/journal/603158a2cbcf420280adf746ae026ec0/system.journal: Journal header limits reached or header out-of-date, rotating.
Feb 15 14:57:25 truenas kernel: binaryurpWriter[502324]: segfault at 0 ip 00007fbaab338a7a sp 00007fbaaacf8c70 error 6 in libuno_cppu.so.3[7fbaab331000+1f000] likely on CPU 0 (core 0, socket 0)
Feb 15 14:57:25 truenas kernel: Code: ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 41 55 41 54 55 53 48 83 ec 18 64 48 8b 04 25 28 00 00 00 48 89 44 24 08 31 c0 83 2f 01 74 20 48 8b 44 24 08 64 48 2b 04 25 28 00 00 00 0f 85

11

Best open a new thread about that. No idea, I am not a Linux guru.

1 Like