[SOLVED] Bad ZFS performance with SAS3416 HBA

TodorPetkov

Well-Known Member
Mar 21, 2020
60
9
48
Hello,

I am having issues with bad ZFS performance on 4 identical server with following hardware:
DELL PowerEdge R6525, CPU 2x AMD EPYC 7262 8-Core Processor, 256GB RAM, HBA Symbios Logic SAS3416, 2 x SSD KPM5XVUG480G, 2 x SAS AL15SEB24EQY.

Software version of PVE:
proxmox-ve: 6.3-1 (running kernel: 5.4.106-1-pve) pve-manager: 6.3-7 (running version: 6.3-7/85c4930a) pve-kernel-5.4: 6.3-8 pve-kernel-helper: 6.3-8 pve-kernel-5.4.106-1-pve: 5.4.106-1 pve-kernel-5.4.103-1-pve: 5.4.103-1 pve-kernel-5.4.73-1-pve: 5.4.73-1 ceph-fuse: 12.2.11+dfsg1-2.1+b1 corosync: 3.1.2-pve1 criu: 3.11-3 glusterfs-client: 5.5-3 ifupdown: residual config ifupdown2: 3.0.0-1+pve3 ksm-control-daemon: 1.3-1 libjs-extjs: 6.0.1-10 libknet1: 1.20-pve1 libproxmox-acme-perl: 1.0.8 libproxmox-backup-qemu0: 1.0.3-1 libpve-access-control: 6.1-3 libpve-apiclient-perl: 3.1-3 libpve-common-perl: 6.3-5 libpve-guest-common-perl: 3.1-5 libpve-http-server-perl: 3.1-1 libpve-storage-perl: 6.3-9 libqb0: 1.0.5-1 libspice-server1: 0.14.2-4~pve6+1 lvm2: 2.03.02-pve4 lxc-pve: 4.0.6-2 lxcfs: 4.0.6-pve1 novnc-pve: 1.1.0-1 openvswitch-switch: 2.12.3-1 proxmox-backup-client: 1.1.3-1 proxmox-mini-journalreader: 1.1-1 proxmox-widget-toolkit: 2.5-2 pve-cluster: 6.2-1 pve-container: 3.3-5 pve-docs: 6.3-1 pve-edk2-firmware: 2.20200531-1 pve-firewall: 4.1-3 pve-firmware: 3.2-2 pve-ha-manager: 3.1-1 pve-i18n: 2.3-1 pve-qemu-kvm: 5.2.0-6 pve-xtermjs: 4.7.0-3 qemu-server: 6.3-10 smartmontools: 7.2-pve2 spiceterm: 3.1-1 vncterm: 1.6-2 zfsutils-linux: 2.0.4-pve1

SSD drives are in ZFS RAID1, same for SAS drives.

I am testing with fio:
fio --rw=write --ioengine=sync --fdatasync=1 --directory=/XXX--size=500m --bs=2300 --name=mytest

On SSD mirror, the performance is good, but on SAS mirror fio tells me it will take like hour and half to complete.
I came accross https://zfsonlinux.topicbox.com/groups/zfs-discuss/T974149177bf5463c-M06b290f6709c4d2efc94ad52 and started suspecting the HBA controller as problematic. If anyone has such controller and can run tests, can you share results?

I am open to ideas how to fix this.

Thanks in advance.
 
Hi,

Turns out it was a wrong setup. Once I added cache and log devices to the SAS HDD pool, performance went through the roof.

Marking the thread as solved.
Thanks
 
  • Like
Reactions: guletz