SSD - Alignment + trim

I might be wrong, but I doubt the question you should be asking is whether OMV does it - the question should probably be whether the kernel does it. Debian is fairly old, but newer backport kernel might have a good enough support for it. You should probably get a confirmation from someone with more experience with that stuff though.

Here is a wiki page about SSD on debian:

https://wiki.debian.org/SSDOpt…#Partitions_and_Alignment

Zitat

Since wheezy all tools should automatically align filesystems and partitions to the 4096 byte page size.[/quote]

[quote]The default I/O scheduler queues data to minimize seeks on HDDs, which is not necessary for SSDs. Thus, use the deadline scheduler that just ensures bulk transactions won't slow down small transactions: Install sysfsutils and

echo "block/sdX/queue/scheduler = deadline" >> /etc/sysfs.conf

(adjust sdX to match your SSD) reboot or

echo deadline > /sys/block/sdX/queue/scheduler

Alles anzeigen

This is what i would do if i had a SSD on the NAS.

regards

Yah, just read the udev rules and its setting the noop scheduler for ssd.

Zitat

Set "discard" mount option in /etc/fstab for the ext4 filesystem, swap partition, Btrfs, etc. See mount(8).

on my install discard wasnt set on swap

i have just carried out a fresh install of OMV using the iso image at 1.2 and the updated it to 1.6 through the gui

my boot drive is a crucial SSD, this one 120GB CSSD-F120GBLS LS http://www.scan.co.uk/products…e-nand-read-535mb-s-write

having looked at fstab, from what i understand it doesnt look like it has been detected as a SSD during installation.
can someone confirm this is so, and what the correct options are to use under OMV 1.6 using the standard kernel?

root@OMV:~# cat /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'blkid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
# / was on /dev/sdb1 during installation
UUID=f9cbc3f8-d4d8-4cc9-b23a-465a79b92545 /               ext4    errors=remount-ro 0       1
# swap was on /dev/sdb5 during installation
UUID=87e2063d-2299-461d-9dd9-b2ec60e15532 none            swap    sw              0       0
/dev/sda1       /media/usb0     auto    rw,user,noauto  0       0
tmpfs           /tmp            tmpfs   defaults        0       0
# >>> [openmediavault]
# <<< [openmediavault]
root@OMV:~#

cheers guys, im looking forward to building a proper NAS out of OMV, my previous effort was lashed together from old disks but the new build uses a SSD boot and 6x 4TB drives in RAID6 as storage

andy

That looks as an usual fstab to me and I do not know if you can tell by the entries if a disk is a SSD or not. That doesn't make a difference for the fstab entries.
You can check with cat /sys/block/sda/queue/rotational if the system sees the disk as a spinning one or as a SSD. If you get "0" it is a SSD, "1" means it is a spinning disk.

for ssd domething like this:

UUID=blahblah / ext4 noatime,nodiratime,discard,errors=remount-ro 0 1

UUID=blahblah none swap sw,discard 0 0

Discard is the option you need for trim support

Aha. And if discard doesn't appear does that mean the SSD wasn't properly detected during reboot?

thanks for the info datadigger and tom-tav. i get a 0 return from the /rotational command so thats looks good.
i need to add the discard and noatime options to fstab

heres a debian page i found which explains what is needed to get the best out of SSD. i havent read it all yet as im just about to go to bed but it looks helpful. ill test tomorrow

https://wiki.debian.org/SSDOptimization

like said above, on my installation the root partition was setup as ssd, the swap wasnt. But fortunately its a 1 minute job to change fstab and reboot