NFS Optimization - Revision history

Drew at 08:33, 25 January 2018

2018-01-25T08:33:07Z

← Older revision		Revision as of 02:33, 25 January 2018
Line 1:		Line 1:
	===Server Side===		===Server Side===
	====/etc/exports====		====/etc/exports====
	<~~pre~~>/mnt/raid5 192.168.15.142/32(rw,async,no_root_squash)</~~pre~~>		<syntaxhighlight lang=bash>
			/mnt/raid5 192.168.15.142/32(rw,async,no_root_squash)
			</syntaxhighlight>
	async - dramatic throughput increase, but dangerous if a client does not unmount cleanly..		async - dramatic throughput increase, but dangerous if a client does not unmount cleanly..

	====Tuning /etc/sysctl.conf====		====Tuning /etc/sysctl.conf====
	<~~pre~~>		<syntaxhighlight lang=bash>
	net.core.rmem_default = 262144		net.core.rmem_default = 262144
	net.core.rmem_max = 262144		net.core.rmem_max = 262144
Line 12:		Line 14:
	net.ipv4.ipfrag_high_thresh = 524288		net.ipv4.ipfrag_high_thresh = 524288
	net.ipv4.ipfrag_low_thresh = 393216		net.ipv4.ipfrag_low_thresh = 393216
	</~~pre~~>		</syntaxhighlight>

	<~~pre~~>		<syntaxhighlight lang=bash>
	echo 0 > /proc/sys/net/ipv4/tcp_sack		echo 0 > /proc/sys/net/ipv4/tcp_sack
	echo 0 > /proc/sys/net/ipv4/tcp_timestamps		echo 0 > /proc/sys/net/ipv4/tcp_timestamps
	</~~pre~~>		</syntaxhighlight>

	====TCP Segmentation offload====		====TCP Segmentation offload====
	This will take off some of the tcp overhead if your card supports it..		This will take off some of the tcp overhead if your card supports it..
	<~~pre~~># ethtool -K ethN tso on</~~pre~~>		<syntaxhighlight lang=bash>
			# ethtool -K ethN tso on
			</syntaxhighlight>

	===Client Side===		===Client Side===
	=====/etc/fstab=====		=====/etc/fstab=====
	NFSv4 Client:		NFSv4 Client:
	<~~pre~~>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs defaults 0 0</~~pre~~>		<syntaxhighlight lang=bash>
			192.168.15.20:/mnt/raid5 /mnt/raid5 nfs defaults 0 0
			</syntaxhighlight>

	NFSv3 Client:		NFSv3 Client:
	<~~pre~~>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs rsize=32768,wsize=32768,intr,hard 0 0</~~pre~~>		<syntaxhighlight lang=bash>
			192.168.15.20:/mnt/raid5 /mnt/raid5 nfs rsize=32768,wsize=32768,intr,hard 0 0
			</syntaxhighlight>
	[rw]size=32768 - NFSv3 maximum read write size<br/>		[rw]size=32768 - NFSv3 maximum read write size<br/>
	intr - if the mount drops, you'll still be able to ^C out of whatever operation your running<br/>		intr - if the mount drops, you'll still be able to ^C out of whatever operation your running<br/>
	hard - ~~hm?~~<br/>		hard - flush file locks before being able to unmount<br/>

Drew: /* /etc/fstab */

2011-09-20T04:24:24Z

/etc/fstab

← Older revision		Revision as of 22:24, 19 September 2011
Line 25:		Line 25:
	===Client Side===		===Client Side===
	=====/etc/fstab=====		=====/etc/fstab=====
			NFSv4 Client:
			<pre>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs defaults 0 0</pre>

			NFSv3 Client:
	<pre>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs rsize=32768,wsize=32768,intr,hard 0 0</pre>		<pre>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs rsize=32768,wsize=32768,intr,hard 0 0</pre>
	[rw]size=32768 - NFSv3 maximum read write size		[rw]size=32768 - NFSv3 maximum read write size<br/>
	intr - if the mount drops, you'll still be able to ^C out of whatever operation your running		intr - if the mount drops, you'll still be able to ^C out of whatever operation your running<br/>
	hard - hm?		hard - hm?<br/>

Drew at 03:21, 6 April 2009

2009-04-06T03:21:12Z

← Older revision		Revision as of 21:21, 5 April 2009
Line 14:		Line 14:
	</pre>		</pre>

			<pre>
	echo 0 > /proc/sys/net/ipv4/tcp_sack		echo 0 > /proc/sys/net/ipv4/tcp_sack
	echo 0 > /proc/sys/net/ipv4/tcp_timestamps		echo 0 > /proc/sys/net/ipv4/tcp_timestamps
			</pre>

	====TCP Segmentation offload====		====TCP Segmentation offload====

Drew at 03:18, 6 April 2009

2009-04-06T03:18:38Z

← Older revision		Revision as of 21:18, 5 April 2009
Line 17:		Line 17:
	echo 0 > /proc/sys/net/ipv4/tcp_timestamps		echo 0 > /proc/sys/net/ipv4/tcp_timestamps

	=====TCP Segmentation offload====		====TCP Segmentation offload====
	This will take off some of the tcp overhead if your card supports it..		This will take off some of the tcp overhead if your card supports it..
	<pre># ethtool -K ethN tso on</pre>		<pre># ethtool -K ethN tso on</pre>

Drew: New page: ===Server Side=== ====/etc/exports====

/mnt/raid5 192.168.15.142/32(rw,async,no_root_squash)

async - dramatic throughput increase, but dangerous if a client does not unmount cle...

2009-04-06T03:17:01Z

New page: ===Server Side=== ====/etc/exports==== <pre>/mnt/raid5 192.168.15.142/32(rw,async,no_root_squash)</pre> async - dramatic throughput increase, but dangerous if a client does not unmount cle...

New page

===Server Side===
====/etc/exports====
<pre>/mnt/raid5 192.168.15.142/32(rw,async,no_root_squash)</pre>
async - dramatic throughput increase, but dangerous if a client does not unmount cleanly..

====Tuning /etc/sysctl.conf====
<pre>
net.core.rmem_default = 262144
net.core.rmem_max = 262144
#
# Increase the fragmented packet queue length
net.ipv4.ipfrag_high_thresh = 524288
net.ipv4.ipfrag_low_thresh = 393216
</pre>

echo 0 > /proc/sys/net/ipv4/tcp_sack
echo 0 > /proc/sys/net/ipv4/tcp_timestamps

=====TCP Segmentation offload====
This will take off some of the tcp overhead if your card supports it..
<pre># ethtool -K ethN tso on</pre>

===Client Side===
=====/etc/fstab=====
<pre>192.168.15.20:/mnt/raid5 /mnt/raid5 nfs rsize=32768,wsize=32768,intr,hard 0 0</pre>
[rw]size=32768 - NFSv3 maximum read write size
intr - if the mount drops, you'll still be able to ^C out of whatever operation your running
hard - hm?