Zfs Drive config on FreeNas

raidztruenaszfs

Couple of related questions.

Background: I have a stock of drives lying about that I want to use in a HP Proliant Microserver which has a 4 slot SAS cage with FreeNas. I don't want to spend any more on parts.

1) With Zfs and RAIDZ, how bad would it be to mix 2 x 1TB Seagate Barracudas with 2 x 1TB WD Green drives. The latter Green drives are I believe only 5,400 RPM, but I can't find that info for sure.

2) Which would be better, the above mis-match in a RAIDZ or 2 x 3TB Seagate Barracuda in a mirror. Considering both performance and data security

3) I have a spare 250GB drive and a spare SATA slot (for optical drive, not sure on speed) as well, would that be a performance gain to use that for the ZIL and L2ARC (over just using the main drives).

Thanks in advance.

Best Answer

If your raid-z VDEV has mixed spindle speeds speeds performance will be like the slowest disks (IO will arrive sooner from the 7200rpm disks, but will wait for data from the 5400rpm before the block is reconstructed.) Your greens are somewhere in the 5400-5900rpm range.
(A) Data security: mirror wins. You'd survive a single disk failing and if you were lucky (33% chance) you could survive a second disk too. Rebuild times are faster with mirrors too. (B) Performance: mirror wins. Double the write performance, better read performance since simultaneous reads are serviced by both halves of the mirror. (C) Capacity (not asked): with mismatched disk sizes raid-z will be limited to the smallest disk (2x1TB + 2x3TB acts just like 4x1TB, so 3TB usable). With mirrors you get all that capacity (2x1TB + 2x3TB = 4TB usable).
A random extra hard disk as L2ARC or ZIL likely won't yield measurable performance gains for most workloads. Maybe as ZIL if the disk was faster than pool media (10k vs 7200rpm). If used as L2Arc maybe a few extra pool IOPS could be eeked under load from any L2Arc cache hits, but I can't imagine it's worth it using an extra old hard drive.

Don't forget to use ashift=12 at pool creation time if your disks have 4K sectors (WD 1TB Green 'EARS' Drives do).

Related Solutions

New Nexentastor installation – can’t run bonnie

Personally, I'd run the bonnie++ benchmark from the command line without the Nexenta wrapper.

First, you'll need to log in as the root user and get into the full command shell mode using:

option expert_mode = 1 followed by !bash

Answer y the prompt that says You are about to enter the Unix ("raw") shell and execute low-level Unix command(s).

From here, you can download and install the proper bonnie++ executable.

# Update the apt package database
root@nyx:/volumes# apt-get update

# Install bonnie++
root@nyx:/volumes# apt-get install bonnie++

# Change to the target directory
root@nyx:/volumes# cd /volumes/Data1/your.cifs.folder

At this point, you can run a bonnie benchmark with or without parameters.
I usually run bonnie++ -u root -n 64:100000:16:64, but you'll be fine running bonnie++ -u root inside of the directory you wish to work in.

Post your results. There are a few tweaks needed for Nexenta when used without SAS or SCSI disks. You should at least see 80-90MB/s sequential write speed to a RAIDZ array. Read speed should be better than that.

ZFS – Impact of L2ARC cache device failure (Nexenta)

ZFS does not do disk I/O, device drivers below ZFS do disk I/O. If the device does not respond in a timely manner, or as in this case, disrupts all other devices on the expander, then it is not visible as a failure to ZFS. All ZFS sees is a slow I/O.

There is a bug in Intel X-25M firmware that affects their behaviour during heavy loads and can cause reset storms. This problem affects all OSes and cannot be solved at the OS layer. Please contact your hardware supplier for fixes or remediation.

If a read is expected to be satisfied by the L2ARC, then the read will be attempted there. ZFS then relies on the lower-layer drivers to report an error. For this case, the drive continues to reset and retry for as many as 5 minutes before declaring the I/O as failed, depending on the driver, device, and default timeout settings. Only after the lower layer drivers declare the I/O as failed will ZFS retry on the pool.

NexentaStor's volume-check and disk-check runners look for additional error messages and alert you via email and fault logging. The disk-check runner has been improved in the 3.1 release to help alert you for specifically the conditions exhibited by broken firmware in SSDs.

Bottom line: your hardware is faulty and will need to be fixed or replaced.

Best Answer

Related Solutions

New Nexentastor installation – can’t run bonnie

ZFS – Impact of L2ARC cache device failure (Nexenta)

Related Topic