11-06-2009
Not sure if this answers your question but NFS is a protocol which can be tuned in certain circumstances e.g. if you have dedicated HBAs in such way you can optimise I/O transfer by for instance changing the size of packets... and try to use fully the bandwidth you have only to serve that purpose (NAS). So you would need extra hardware for a start...
---------- Post updated at 10:46 ---------- Previous update was at 10:28 ----------
Do a search on NFS and jumbo frames...
10 More Discussions You Might Find Interesting
1. SuSE
Hello, I am running a Suse Linux server and I want to set up a NFS Server for a few Unix machines. For the root account at the unix client it works fine but it doesn't work for other users who have no root rights.
I've used the no_root_squash and the rw option in the etc/export file. My folder... (7 Replies)
Discussion started by: Ald-Edv
7 Replies
2. HP-UX
is it possible? i will only config on the windows side. is there a third party software to help me accomplish this? tnx (3 Replies)
Discussion started by: Amaru
3 Replies
3. Solaris
Hi all, i Followed these steps & still not able to get nfs service online ???:(
bash-3.2# svcs -a | grep nfs
disabled 17:48:02 svc:/network/nfs/cbd:default
disabled 17:48:02 svc:/network/nfs/client:default
disabled 20:56:06 svc:/network/nfs/status:default
offline ... (4 Replies)
Discussion started by: saurabh84g
4 Replies
4. UNIX for Dummies Questions & Answers
Hi,
I would like to know if It's possible configure NFS on MP-RAS Unix so mountd uses a static port instead of a dynamic one.
Thanks. (1 Reply)
Discussion started by: cuatrodos
1 Replies
5. Red Hat
Hello everyone,
I am running a RHEL4 server. We have 4 nics, which are paired to create bond0 and bond1. BOND0 is on a vlan which can see our storage system. BOND1 cannot see the storage system. I would like to lock down the system, so that NFS share can only mounted via BOND0. Has anyone done... (2 Replies)
Discussion started by: vada010
2 Replies
6. Solaris
Hi all
in my server all nfs dependices are online but nfs client is offline
root@BIWAPP1 # svcs -a|grep nfs
disabled Sep_05 svc:/network/nfs/server:default
online Sep_05 svc:/network/nfs/rquota:default
online Sep_05 svc:/network/nfs/mapid:default
online ... (9 Replies)
Discussion started by: spandhan
9 Replies
7. Solaris
Can any one please tell me the command for NFS client service restart.This is to resolve, NFS mount slowness issue.Mount is very slow for both read and write operations.The below commands are not helping out in this situation.
---------- Post updated at 08:59 AM ---------- Previous update was... (1 Reply)
Discussion started by: ksvaisakh
1 Replies
8. HP-UX
I have an HP Unix server with Oracle DB and want to write Datapumo export files across the network to IMB/Linux NFS. Will that work? (3 Replies)
Discussion started by: Duane McDonough
3 Replies
9. Red Hat
Hello,
I have a newly kickstarted RHEL 6.4 server that I'm trying to set-up as a kickstart server. I have done this before on other machines, but I am encountering some strange behaviour in this one.
# service nfs status
rpc.svcgssd is stopped
rpc.mountd is stopped
nfsd is stopped... (0 Replies)
Discussion started by: Chopstyx
0 Replies
10. Homework & Coursework Questions
Hi All,
Need your help to resolve below error in cluter.
Sep 9 05:37:30 node2 rgmanager: Starting disabled service service:Prod_Service
Sep 9 05:37:30 node2 rgmanager: HA LVM: Improper setup detected
Sep 9 05:37:30 node2 rgmanager: * initrd image needs to be newer than lvm.conf... (3 Replies)
Discussion started by: nitinredekar
3 Replies
LEARN ABOUT HPUX
nfs3_max_threads
nfs3_max_threads(5) File Formats Manual nfs3_max_threads(5)
NAME
nfs3_max_threads - control the number of kernel threads that perform asynchronous I/O for the NFS version 3 client
VALUES
Failsafe
Default
Allowed values
Recommended values
A warning will be issued at runtime if the tunable is set to a value greater than 256 threads since this is beyond the tested limit. This
is not a serious warning but just an information message for the administrator.
DESCRIPTION
controls the number of kernel threads that perform asynchronous I/O for the NFS version 3 client. Since NFS is based on RPC and RPC is
inherently synchronous, separate execution contexts are required to perform NFS operations that are asynchronous from the calling thread.
The operations which can be executed asynchronously are read for read-ahead, readdir for readdir read-ahead, and write for putpage and
pageio requests.
Who Is Expected to Change This Tunable?
The distributed file system administrator should examine this value depending on the network bandwidth available.
Restrictions on Changing
The tunable is dynamic; however, the number of threads for a file system is set when the file system is mounted. To affect a particular
file system, unmount and mount the file system after changing this parameter.
When Should the Value of This Tunable Be Raised?
If the network is very high bandwidth and the client and server have sufficient resources, increase this value to more effectively utilize
the available network bandwidth, the client resources, and the server resources.
What Are the Side Effects of Raising the Value?
An increase in the system resources and the possibility that (see nkthread(5)) may need to be increased. The total number of asynchronous
threads for NFS version 2 and NFS version 3 mount points cannot exceed 20% of the available threads defined by NFS mounts will fail if the
NFS mount command cannot guarantee the ability to create the maximum number of threads for that mount point.
When Should the Value of This Tunable Be Lowered?
In a very low bandwidth network, decrease this value so that the NFS client does not overload the network.
What Are the Side Effects of Lowering the Value?
Decreasing this value may impact NFS performance.
WARNINGS
All HP-UX kernel tunable parameters are release specific. This parameter may be removed or have its meaning changed in future releases of
HP-UX.
Installation of optional kernel software, from HP or other vendors, may cause changes to tunable parameter values. After installation,
some tunable parameters may no longer be at the default or recommended values. For information about the effects of installation on tun-
able values, consult the documentation for the kernel software being installed. For information about optional kernel software that was
factory installed on your system, see at
AUTHOR
was developed by Sun Microsystems, Inc.
SEE ALSO
kctune(1M), sam(1M), gettune(2), settune(2), nfs2_max_threads(5), nfs4_max_threads(5), nkthread(5).
Tunable Kernel Parameters nfs3_max_threads(5)