Grubs Batch Feature

In grub.conf you have default= yeah,  where you specify which kernel is booted. leave this set to a kernel that you know is good and working. then to test a new kernel, you can tell grub to try that kernel only once with this command before you reboot

echo "savedefault --default=0 --once" | grub --batch


 if the server hangs, get the customer to reboot the server and it will then go back to the good kernel from grub.conf

Reduce the disk space/shrink the LVM in an Linux HVM using kpartx

kpartx -av /dev/vg_67/vm102_img

It will then say something about mapping the partition, and will probably create this:
/dev/vg_67/vm102_img1

This relates to that one partition.

now you can do
e2fsck -f -y /dev/vg_67/vm102_img1
resize2fs /dev/vg_67/vm102_img1 24G

then disable kpartx on the lv

kpartx -d /dev/vg_67/vm102_img

now lvreduce to 25G

lvrdeuce -L 25G /dev/vg_67/vm102_img

then you can use kpartx once more to increase the partition to fill the LV, just resize2fs /path you dont need to specify a size.

resize2fs /dev/vg_67/vm102_img

Permanent Change of the Time Zone on a Node/Server

First, make a backup of the existing localtime file.

mv /etc/localtime /etc/localtime.bak
Next, create the link:

ln -s /usr/share/zoneinfo/Europe/London /etc/localtime

Now you just need to test your change. Run “date” from the command line, and ensure that the appropriate time, date, and timezone are reported.

The settings will be reverted, if we haven't made the below change.

You need to edit the file /etc/sysconfig/clock and add the entry ZONE="Europe/London" and save the file.

Xen: Create bridge manually with the bond0

ifcfg-eth0

DEVICE=eth0
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
ISALIAS=no

ifcfg-eth1

DEVICE=eth1
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
ISALIAS=no

ifcfg-eth1

DEVICE=bond0
BOOTPROTO=none
BONDING_OPTS="mode=1 miimon=100"
ONBOOT=yes
BRIDGE=xenbr0

ifcfg-xenbr0

DEVICE=xenbr0
BOOTPROTO=static
IPADDR=x.x.x.x
NETMASK=x.x.x.x
ONBOOT=yes
TYPE=bridge

vi /etc/xen/xend-config.sxp
Comment out ALL (network-script network-bridge)

service xend restart
service network restart

HDD Health Check Alert

Physical Disk with Smartctl
---------------------------------------

for i in `lshw -class disk -class tape | grep "logical name" | cut -d: -f2`; do echo Disk $i ; echo "=========================================" ;smartctl --all $i | grep 'Status\|Drive' ; echo " "; done > /tmp/output.txt
mail -s "HDD Health Check Alert: `hostname`" user@domain.com  < /tmp/output.txt
rm -f /tmp/output.txt

With MegaRaid Controller

---------------------------------------

for i in `/opt/MegaRAID/MegaCli/MegaCli64 -PDlist -a0|grep Device.Id | awk '{print $3}'` ; do echo Device $i ; smartctl -a -d sat+megaraid,$i /dev/sda|grep 'PASSED\|Reallocated_Sector_Ct\|Temperature_Celsius'; echo " "; done > /tmp/output.txt

mail -s "HDD Health Check Alert: `hostname`" user@domain.com < /tmp/output.txt

rm -f /tmp/output.txt