RE: [coLinux-users] NFS performance and colinux

[Claude L. (QB/EMC) <cla...@er...>]

Hi George,

Thanks for your reply. I will give a try to the new release as soon as =
possible.

However, I have made some NFS troubleshooting and I have finally been =
able to go a bit further with the jumpstart and I have been able to =
narrow down the problem. The NFS performance problem seems to be =
produced by the IP fragmentation.

The default NFS block size on Linux is set to 4096 bytes (4K). When I =
set the rootopts value to 1024, the jumpstart process goes a lot =
better. This parameter set the rsize (read block size) option of the =
NFS mount request of the root filesystem coming from the jumpstart =
client. Unfortunately, all the other NFS mounted filesystems are not =
using this parameter and suffer from the fragmentation problem. So, the =
jumpstart fails later.

Basically, there is no fragmentation at 1024 but, at 4096, the =
fragmentation is very important. To verify that the fragmentation is =
responsible of the bad NFS performance, I have tested ping of 4K to a =
remote system from colinux:

PING 142.133.81.2 (142.133.81.2) 4096(4124) bytes of data.
4104 bytes from 142.133.81.2: icmp_seq=3D9 ttl=3D255 time=3D82.4 ms
4104 bytes from 142.133.81.2: icmp_seq=3D14 ttl=3D255 time=3D4.17 ms
4104 bytes from 142.133.81.2: icmp_seq=3D22 ttl=3D255 time=3D4.04 ms
4104 bytes from 142.133.81.2: icmp_seq=3D24 ttl=3D255 time=3D5.05 ms
4104 bytes from 142.133.81.2: icmp_seq=3D25 ttl=3D255 time=3D4.15 ms
4104 bytes from 142.133.81.2: icmp_seq=3D32 ttl=3D255 time=3D21.0 ms
4104 bytes from 142.133.81.2: icmp_seq=3D34 ttl=3D255 time=3D4.16 ms
4104 bytes from 142.133.81.2: icmp_seq=3D78 ttl=3D255 time=3D4.17 ms

--- 142.133.81.2 ping statistics ---
121 packets transmitted, 8 received, 93% packet loss, time 120805ms
rtt min/avg/max/mdev =3D 4.048/16.165/82.463/25.656 ms, pipe 2

The ping test shows the 93% of packet drops... It looks like the =
colinux network adapter suffers from an IP fragmentation/reassembly =
performance problem.

Please, can you try the same ping on a later colinux version (ping -s =
4096 <remote_ip_address>) if you have time and if you can ? So, we =
could already know if this problem is addressed with the new releases.

Thanks,

Claude LeFran=E7ois=20
cla...@er...


-----Original Message-----
From: col...@li...
[mailto:col...@li...]On Behalf Of gboutwel
Sent: Saturday, August 14, 2004 7:57 PM
To: col...@li...
Subject: Re: [coLinux-users] NFS performance and colinux


> I am using colinux 0.6.1 over kernel 2.4.26. The original =
distribution
=3D



Only advice I can give is to try the more recent snapshots.=20
Especially the 20040710 snapshot.  It will involve and upgrade
to an 2.6 kernel and all that that may entail for the distro
of your your choice.  It probalby take rebulding the kernel with
NFS settings, but it might not.  But network performance has
been something that has been worked and has gotten comments about
it betting better in recent snapshots.



> If somebody has an idea why the NFS service is so slow, please,
do not =3D

> hesitate to contact me.



If after trying the recent snapshots, you still have problems
we can perhaps look at them to see fi there is more we can do.
 Additionally, If you find that we could enable something more
in the kernels for recent snapshots to make NFS support better,
let us know.



George