[PVE-User] pve-kernel-4.13.13-5-pve hang with high IO
elacunza at binovo.es
Wed Jan 24 09:35:22 CET 2018
We just installed Proxmox 5.1 on a i7-3770 CPU @ 3.40GHz machine with
32GB of RAM two days ago. It has 4 disks with mdraid6.
It has been working very well for two days; more than 1TB of VM disk
have been transferred to that Proxmox, and they have been moved from a
Directory based storage to a LVM-thin storage successfully.
Last night at 23:32 it hanged, and we had to hard-reboot it this
morning. On that moment it was receiving about 100MB/s of network
traffic to a VM backed on LVM-thin disk, and writing the data to disk
(several rsync transfers). Don't know if that activity is related to the
hang really, syslog just shows about 20 lines of garbage.
Anyone has seen something like this lately? I first thought about
Meltdown/Spectre patches and reboot issues, but don't know what's the
state in current Proxmox kernels.
I can see the following in kernel boot log, doesn't seem it is loading a
very recent microcode:
Jan 24 09:18:30 orion kernel: [ 0.000000] microcode: microcode
updated early to revision 0x1c, date = 2015-02-26
Also BIOS seems old, shouldn't be patched:
Jan 24 09:18:30 orion kernel: [ 0.000000] DMI: System manufacturer
System Product Name/P8H77-M PRO, BIOS 9002 05/30/2014
# pveversion -v
proxmox-ve: 5.1-36 (running kernel: 4.13.13-5-pve)
pve-manager: 5.1-43 (running version: 5.1-43/bdb08029)
Thanks a lot
Zuzendari Teknikoa / Director Técnico
Binovo IT Human Project, S.L.
Astigarraga bidea 2, 2º izq. oficina 11; 20180 Oiartzun (Gipuzkoa)
More information about the pve-user