FatMin

Top 10 Free Vmware Administration Tools

In an effort to more fully immerse myself into the world of Vmware and Virtualization, I have been spending a good bit of time, as of late, poking around the virtual blogosphere looking for blogs that are "must reads". There is no shortage of great information out there to soak in.

Stumbled across the post below today and figured I would post it here so that I can be sure to find it in the future, when i actually get the time to read the blogs that I am finding.

Top 10 Free Tools that must be in every VMware Administrators toolkit

Multiple Bonding Modes in RHEL5

In RHEL 4, if you had only one bond, you would define your bonding mode in the /etc/modprobe.conf. Note that this configuration only lets you create one bonding device. If you wanted two bonds with different bonding options you had to do something similar to what you see below.

Instead of this in your modprobe.conf...

alias bond0 bonding
options bond0 miimon=100 mode=1
alias bond1 bonding
options bond1 miimon=100 mode=1

You would have this...

install bond0 /sbin/modprobe bonding -o bond0 miimon=100 mode=1
install bond1 /sbin/modprobe bonding -o bond1 miimon=100 mode=0

If you didn't and you tried to define both bond configs in your modprobe, you will only have one module loaded called bonding and bond1 will get the options of bond0. Not exactly what you want.

Note that you may need to modify your modprobe.com and add the following line... max_bonds=2.

In RHEL 5 you actually define the bonding options in your bond config file (/etc/sysconfig/network-scripts/ifcfg-bondo for example), with the BONDING_OPTS variable: BONDING_OPTS="mode=1 miimon=100". You will need to do something similar in both ifcfg-bond* files.

Finding Large Files in Unix/Linux

Back in the day I used to have some really good notes on this topic, but alas they are long gone, so I have decided to put together a small list with some helpful examples. Note that a lot of folks will show you examples where they pipe their find out to a 'rm' command. This is bad mo-jo in my opinion and should be avoided. Mass deleting files based on size alone is not a good practice. There are exceptions, but as a rule I avoid it like hoof and mouth disease.

Use this simple command to find large directories. To find directories over 1GB
[root@localhost]# du -h / | grep ^[0-9\.]*G

G can be replaced with M if looking for MB instead.

To find directories over 10GB and sort the output with the largest directories on top
[root@localhost]# du -h / | grep ^[1-9][0-9][0-9\.]*G | sort -rn

To find files within a size limit.
find /etc -size +100k -size -150k

Find files under /, and do not cross filesystems. This will keep you from searching pesky nfs and san mounts.

find / -name -depth +100k

Show directory sizes using du
 du -sk /usr/*

Find the largest 10 files
find / -size +15M -printf "%s - %p\n" | sort -n | tail

Cool Linux Command: ifstat

Ifstat is a little tool to report interface activity, just like iostat/vmstat do for other system statistics.

Usage is simple...

>ifstat
       eth0
 KB/s in  KB/s out
    0.87      0.61
    0.07      0.13
    0.13      0.13
    0.07      0.13
    0.13      0.13
    0.07      0.13
    0.13      0.13
    0.07      0.13
    0.13      0.13
    0.07      0.13

Mrepo (mount: could not find any free loop device)

When attempting to mount my loopback isos in CentOS5 via mrepo I ran into the following error

mrepo --remount                                           
mount: could not find any free loop device                                     
mount: could not find any free loop device                                     
mount: could not find any free loop device

Apparently there is a default limit of 8 for the number of loopback devices that you can mount, however this can be changed by adding the following to /etc/modprobe.conf. Below I am changing the default of 8 to 64.

options loop max_loop=64

Then umount your currently mounted isos

mrepo --umount

Then unload and reload the loop module

rmmod loop && modprobe loop

Then mount your isos again

mrepo --remount

Adding A Swap File to Linux

SWAP File

The following commands illustrate how to add a swap file of 2GB to a RedHat system:

# dd if=/dev/zero of=/temp_swap bs=1024 count=2097152
# mkswap /temp_swap
# swapon /temp_swap

To enable it at boot time, edit /etc/fstab to include:

/temp_swap swap swap defaults 0 0

After adding the new swap file and enabling it, verify it is enabled by viewing the output of the command:

# free -mo

Remove a Lun From Redhat

Note that removing a SAN lun is a semi-tedious process that must be followed. Failure to do so will result in your server being left in a funky state. The process varries depending on how you are using your disk

Using LVM?

For Example: Is the disk part of a Logical Volume and under LVM control? Then shrink the volume group and then remove the disk from the volume group using pvremove.

vgreduce VolGroup <device name from `vgdisplay -v vgname` output>
pvremove <device name from `pvscan` output>

Kpartx & Multipath


Despite its name, its not a KDE-tool for partitioning disks.
Kpartx is a Partition device manager for device-mapper devices that is usually shipped as part of 
multipath. Its purpose is to read the partition table of a device or a
file and create device map entries for the individual partitions.

When removing a disk from a system, you will need to remove the partition map of the multipathed device.

kpartx -d /dev/mapper/mpath5 <to remove the /dev/mapper/mpath5p1 device>
multipath -f mpath5 <where mpath5 is the alias for the multipathed device <br>

see example:


mpath5 (360060e801004e780052fab2800000064) dm-7 HITACHI,DF600F
[size=1.0G][features=1 queue_if_no_path][hwhandler=0]
\_ round-robin 0 [prio=1][active]
 \_ 3:0:0:3 sdj 8:144 [active][ready]
 \_ 4:0:0:3 sdh 8:112 [active][ready]

Removing the disk from Linux Control.

This is the most important step, as not doing this part is what will screw up your box.

cd to /sys/class/scsi_device
then cd to the number of the SCSI device, for example 4:0:0:3 is the device id for /dev/sdh
then type

echo 1 > device/delete

Linux SAN Disk Managment via DM-Multipath

A little background...
Most of the time, I have used the RDAC driver in Linux to manage SAN disks in Linux. The RDAC driver is used to hide the complexity of multiple paths and to present redundant paths as a single path which can be used as you would a standard SCSI / IDE / SAS / SATA drive. Seeing only one device makes managing your disks much easier.

However where I work we only use RDAC with our IBM FastT, Sun 6140 and STK Flexline storage arrays. RDAC is not for LSI based storage such as Hitachi, Clarion, and EMC. For these servers we manage SAN disk with DM-Multipath.

Setup...
Setting up DM-Multipath is not hard, first you need to make sure that you install the package, device-mapper-mulitpath, and you will need to configure your multipath.conf and drop it into /etc. Below is some info on how to do so.

http://kbase.redhat.com/faq/docs/DOC-3691

You will also need to make sure that you enable the multipathd daemon. This daemon is in charge of checking for failed paths.

Multipath Command...
For those use to using RDAC, DM-Multipath takes some getting used to, especially when you see the output from fdisk -ll.

In one particular instance I was given the disk name of /dev/sdm as the name of the new disk on this box. The output from the fdisk -l command is not exactly helpful, as there are a ton of psuedo devices showing up in my output. This is where the multipath command comes in handy.

Linux + Cron + Pam Access

Ran into the following issue when setting up a user to have cron access on a RHEL 4 box

Oct  7 10:34:01 'hostname' pam_access[20206]: access denied for user `user' from `cron'
Oct  7 10:34:01 'hostname' crond[20206]: Permission denied

First and formost you should know that any user that you want to be able to user cron should be added to /etc/cron.allow. Almost anyone who is familiar with Cron should already know this.

Poking around a bit on google indicated that I needed to add the following to /etc/security/access.conf

+ALL:cron crond

Make sure that this line come before the line below

-:ALL:ALL

Below is the /etc/pam.d/crond file that I used

#
# The PAM configuration file for the cron daemon
#
#
auth       sufficient pam_rootok.so
auth       required   pam_stack.so service=system-auth
auth       required   pam_env.so
account    sufficient /lib/security/$ISA/pam_localuser.so
account    required   pam_stack.so service=system-auth
account    required   pam_access.so
session    required   pam_limits.so
session    required   pam_loginuid.so

Linux Library Madness

Been having a very hard time working with a DBA to install Oracle 10g on a RHEL 4 box. He was getting this message.

#java.lang.UnsatisfiedLinkError: /tmp/OraInstall2008-09-23_10-47-22AM/jre/1.4.2/lib/i386/libawt.so: libXtst.so.6: wrong ELF class: ELFCLASS64

First thing that I did was figure out what rpm provides this libXtss.so.6 file

#rpm -qf /usr/lib64/libXtst.so.6

this command responds with: libXtst-1.0.1-3.1

Now from what I have read Oracle 10g does not like 64 bit libraries, so I needed to figure out what architecture that I have installed.

# rpm -qa --qf="%{n}-%{v}-%{r}.%{arch}.rpm\n" "libXtst"
libXtst-1.0.1-3.1.x86_64.rpm

So know I know that I have the 64 bit version of the rpm installed. I need to install the 32 bit version.

#libXtst.i386                             1.0.1-3.1              base
Matched from:
libXtst
X.Org X11 libXtst runtime library
X.Org X11 libXtst runtime library

Ok, there is the package. So lets install it.
yum install libXtst.i386

Issue resolved.

Additional One Liners:

rpm -ivh packages(s).rpm
install rpm file(s)

rpm -Uvh packages(s).rpm  
upgrade system with rpms

rpm -e package
remove package

rpm -q package  
show version of package installed

rpm -q -i package  
show all package metadata

rpm -q -f /path/file  
what package does file belong

rpm -ql package
lists files in a rpm package