Rather useful command that can be used for testing network bandwidth, which in my case was used for testing our 10Gbit network card on a direct loop could get a decent speed. First your need to copy over the binary to install, which as you will see you need to configure yourself -
# cd iperf-2.0.5
Run the configure command to set it up for your system:
In our case there where some missing filesets, details of which can be found in the config.log in the install directory. So install the additional packages:
# yum install gcc
# yum install gcc-c++
# yum install make
Then run the configure again, followed by a make:
# make install
Now that its installed, go and install the code on the other server that you need to test with. Once that is completed we can start iperf in server mode on one box:
# iperf -s
Client connecting to 10.10.0.11, TCP port 5001
TCP window size: 85.3 KByte (default)
[ 4] local 10.10.0.12 port 60000 connected with 10.10.0.11 port 5001
[ 3] local 10.10.0.12 port 59999 connected with 10.10.0.11 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 8.64 GBytes 7.42 Gbits/sec
[ 4] 0.0-10.0 sec 1.98 GBytes 1.70 Gbits/sec
[SUM] 0.0-10.0 sec 10.6 GBytes 9.11 Gbits/sec
# yum install libstdc++.ppc64
# yum install rpm-build
From this you should be able to simple follow the prompts from the install script, and as long as you have sufficient space all should install OK.
Check the installed version as follows:
# /usr/<install-path>/bin/java -version
To toggle the CPUs offline/online consistently across the cores.
# ppc64_cpu --smt=x
Linux numbers all of the CPUs sequentially across the cores.
# ppc64_cpu --smt=1
Will "turn off" cpu1,cpu2, cpu3 of the four CPUs associated with each core, leaving cpu0 running. 0,1,2,3 is just the example for the first core. That repeats up through the CPUs.
# ppc64_cpu --smt=2
Insures that cpu0, cpu1 are "on", and cpu2, cpu3 are "off" for each core.
# ppc64_cpu --smt=4
Insures that all cpus are "on"
With SMT=1 mode (aka SMT off), the Linux scheduler should shift to the appropriate scheduling mode.
Mounting from AIX to Redhat
I keep forgetting these commands even though I'm mounting filesystems all the time, so here is a summary of what you need to do to get the mounts working from a AIX NFS server to a fresh Redhat 6.4 install.
So the first install error:
# mount -t nfs 172.19.80.40:/nimfs /mnt
mount: wrong fs type, bad option, bad superblock on 172.19.80.40:/nimfs,
missing codepage or helper program, or other error
(for several filesystems (e.g. nfs, cifs) you might
need a /sbin/mount.<type> helper program)
In some cases useful info is found in syslog - try
dmesg | tail or so
Means the following package is missing:
# yum install nfs-utils-lib.ppc64
# mount -t nfs 172.19.80.40:/nimfs /mnt
mount.nfs: Remote I/O error
This relates to setting the NFS type:
# mount -t nfs 172.19.80.40:/nimfs /mnt -o nfsvers=3
mount.nfs: rpc.statd is not running but is required for remote locking.
mount.nfs: Either use '-o nolock' to keep locks local, or start statd.
mount.nfs: an incorrect mount option was specified
And last, as it says rpc.statd is not running, you can start this, or do as I did and:
# mount -t nfs 172.19.80.40:/nimfs /mnt -o nfsvers=3 -o nolock
# df -h
Filesystem Size Used Avail Use% Mounted on
45G 4.9G 38G 12% /
tmpfs 1002M 0 1002M 0% /dev/shm
/dev/sda2 485M 44M 416M 10% /boot
172.19.80.40:/nimfs 1.6T 1.5T 153G 91% /mnt
Happy days, now I have my mount.
Need to install a minimal desktop of your server environment then do the following:
# yum install gdm, gnome-panel, gnome-shell, nautilus
You may also want:
# yum install gnome-packagekit, NetworkManager-gnome, xdg-user-dirs-gtk
Then for a nice experience:
# yum install firstboot, ntp, plymouth-theme-charge, yum-plugin-fastestmirror, yum-presto
Don't forget video drivers:
# yum install xorg-x11-drivers
From here you can build it up as you like:
If you trying to get vnc working then ensure to add the following:
# yum install vnc-server twm xterm