Search This Blog

Monday, October 10, 2011

XenServer v6 installation problem: UNSUPPORTED_INSTALL_METHOD - other-config:install-repository was not set to an appropriate value, and this is required for the selected distribution type

It has been a long time since I played with the XenServer for the last time.

I managed to install the XenServer without any problems. The XenConsole installation went flawlessly as well. But when I tried to create a new VS using the provided templates i run into the desribing problem:

What I did in XenCentre:
  1. XenCentre
  2. New VM
  3. Ubuntu Lucid Lynx 10.04(32 bit) 
  4. Then use the default settings and once you see the screen with the:
"Select the installation method for the operating system software you want to install on the new VM"

You seen on the screen a page where you should select the desired installation source. In my example the option for "Install from ISO library or DVD drive" was grey but you can still select one of the ISO images from your storage pool (if you defined it previously).

Problem:
If you attempt to run the VS to start the installation you will get an error like this:

UNSUPPORTED_INSTALL_METHOD - other-config:install-repository was not set to an appropriate value, and this is required for the selected distribution type

There is not much on the Citrix support/technical knowledge page for this. The solution is here although.

Solution:
  • To create VM and start the installation from a ISO file saved on the attached storage (NFS or CIFS) you have to use the correct profile: Other install media
  • To use the Ubuntu Lucid Lynx 10.04(32 bit) profile you have to specify an URL for the data to be retrived (for example like http://mirror.clarkson.edu/ubuntu/ )

References:
http://lists.xensource.com/archives/html/xen-users/2011-06/msg00394.html

Thursday, May 19, 2011

Screen trick how to simultaneously monitor network traffic and disk activity and processes in Linux

As a system administrator sooner than later you will discover the 'screen' command. It is very powerful tool that allows you practically do anything with your (black) terminal screen or run commands in background for you.

The tools we will use for monitoring:
- htop is a top like process monitoring tool for text consoles
- nethogs is a network monitoring tool (needs root privileges!)
- iotop is a disk I/O monitoring tool

To start a single screen session, divided in 3 vertical smaller screens, each running its own monitoring program try this recipe:

# cat <<END >/tmp/myscreen.cfg
startup_message off
screen -t top htop
split
focus
screen -t nethogs  nethogs wlan0
split
focus
screen -t iotop iotop
END

# screen -c /tmp/myscreen.cfg

References:
Top 10 Awesome Linux Screen Tricks

Wednesday, May 4, 2011

By using Ubuntu how to create a bootable USB thumb drive with Fedora distro

Although this it has already been possible to use USB thumb drive and boot from it for a longer time, there is still no one unified method to create this USB thumb drive.

Ubuntu has its own method:

"2) Burn your CD or create a USB drive"
http://www.ubuntu.com/download/server/download

Fedora has its own as well:

"3.2.1. Making Fedora USB Media on a Windows Operating System"
http://docs.fedoraproject.org/en-US/Fedora/14/html/Installation_Guide/Making_USB_Media.html

Fortunately for us a separate project unetbootin [1] comes in handy to do this job regardless what OS you are currently using.

To simply install and run follow this instructions bellow.

# aptitude install unetbootin 
# dpkg -L  unetbootin | grep bin
/usr/bin/unetbootin
# unetbootin


References:

[1]
UNetbootin - Homepage and Downloads

Using Fedora's liveusb-creator on Ubuntu Lucid Lynx

Sunday, May 1, 2011

Gdisk error message: Caution: invalid main GPT header, but valid backup; regenerating main header from backup!

At some point when experimenting with partitions on your disk you may get the following error message.

% gdisk /dev/sda
GPT fdisk (gdisk) version 0.7.1

Caution: invalid main GPT header, but valid backup; regenerating main header
from backup!

Caution! After loading partitions, the CRC doesn't check out!
Warning! Main partition table CRC mismatch! Loaded backup partition table
instead of main partition table!

Warning! One or more CRCs don't match. You should repair the disk!

Partition table scan:
  MBR: MBR only
  BSD: not present
  APM: not present
  GPT: damaged

Found valid MBR and corrupt GPT. Which do you want to use? (Using the
GPT MAY permit recovery of GPT data.)
 1 - MBR
 2 - GPT
 3 - Create blank GPT

As always you have to be very careful depending what you did and what do you want to do as next.

The three options for gdisk gives you the choice what partition tables you will examine. Fortunately for us gpart is not going to write on disk anything until we say so by executing the w (write) command (example bellow). The three options only influences how the data on the disk are going to be interpreted and presented for our review.

# gdisk /dev/sda
Command (? for help): h
b    back up GPT data to a file
c    change a partition's name
d    delete a partition
i    show detailed information on a partition
l    list known partition types
n    add a new partition
o    create a new empty GUID partition table (GPT)
p    print the partition table
q    quit without saving changes
r    recovery and transformation options (experts only)
s    sort partitions
t    change a partition's type code
v    verify disk
w    write table to disk and exit
x    extra functionality (experts only)
?    print this menu

At this stage it is save to experiment and have a look how the partition tables look like. Depending on your choise different partitions may be printed ('p') .

REMEMBER to always quit the session with the 'q' (quit) and never with 'w' otherwise your experiments will be permanently saved on the disk.

In my case this error was misleading and confusing only. I could verify that my current partition schema was not GPT at all but the good old MSDOS one [1]. At this stage I knew that 'gdisk' is not the tool I wanted to use and I finished creating new partitions with 'parted'. I didn't have to use gdisk at all.

The reason why I had some parts of GPT data on my disk is unknown. I can only suspect that it was created when I played with the Windows tool 'EasyBCD'.

I could try to delete the GPT data but base on [2] never had to do it.

References
[1] How to find what type of disk partition schema do I use (msdos, gpt)

[2] Wiping Out Old GPT Data

How to find what type of disk partition schema do I use (msdos, gpt)

There are many disk partition schemes that can be used. Very likely on most x86 base computer the choice is often limited by the operating system we use although.

The list of supported labels (partition schemes) in GNU parted tool can be found here [1]:


label-type must be one of these supported disk labels:

* bsd
* loop (raw disk access)
* gpt
* mac
* msdos
* pc98
* sun


To find out what type of partition schema we currently have on your system please run the bellow command and check the value for the 'Partition Table'.

% parted /dev/sda unit mb print
Model: ATA ST2000DL003-9VT1 (scsi)
Disk /dev/sda: 2000399MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start     End       Size      Type     File system  Flags
 1      0.03MB    104856MB  104856MB  primary  ntfs         boot
 2      104857MB  209714MB  104858MB  primary  ntfs
 3      209714MB  312115MB  102401MB  primary
                                      
root@sysresccd /tmp % parted /dev/sda unit s print
Model: ATA ST2000DL003-9VT1 (scsi)
Disk /dev/sda: 3907029168s
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start       End         Size        Type     File system  Flags
 1      63s         204796619s  204796557s  primary  ntfs         boot
 2      204797952s  409597951s  204800000s  primary  ntfs
 3      409597952s  609599487s  200001536s  primary


References:

[1]
(parted) mklabel msdos

Others:

Parted User's Manual

GUID Partition Table

Make the most of large drives with GPT and Linux

Fun with GPT partitioning

Linux Creating a Partition Size Larger than 2TB

How to export/backup partition schema for a disk using old style disk partitions format like 'msdos' written into MBR sector

To backup the partition schema of the disk you can execute:

# sfdisk -d /dev/sda > part.txt

# example output can look like
# cat part.txt
# partition table of /dev/sda
unit: sectors

/dev/sda1 : start=       63, size=204796557, Id= 7, bootable
/dev/sda2 : start=204797952, size=204800000, Id= 7
/dev/sda3 : start=409597952, size=200001536, Id=83
/dev/sda4 : start=        0, size=        0, Id= 0

To restore the part.txt you can use:

# sfdisk /dev/hda < part.txt 
 
References:
man sfdisk 
Wiping Out Old GPT Data 

Howto capture and record the console screen output to a file on disk

The basic of capturing a data to a file require simple redirection or using of 'tee' program for example.

Examples:

$ ls -la > /var/tmp/output.ls.txt
$ ls -la | tee /var/tmp/output.ls.txt

But sometimes programs can be interactive or we simply want to capture all our session without to worry to redirect the stdout to a file.

The solution is to use the 'screen' tool.

Example #1: Screen basic usage

# to start the session with a name 'example' run
$screen -h
$screen -S example

# to leave the screen session type: CONTROL-a d
# you are placed back in the original shell

# to reattach to the created session
$ screen -ls
$ screen -r example

Example #1: Enable screen logging

The options '-L' instruct screen to create a log file that will capture all the commands output in the screen session.

$ screen -S logexample -L

# inside the screen session
$ echo 'some output'

# to leave the screen session type: CONTROL-a d

# this is the default log file for screen
$ ls -la screenlog.0

$ cat screenlog.0