Re: Change image timeout

2010-05-17 Thread Vallard Benincosa
xCAT puts that in the site table which is on default xCAT 1.3 installations
at /opt/xcat/etc/site


On Mon, May 17, 2010 at 10:35 AM, Roger Herling roger.herl...@marist.eduwrote:

 Where do I set that?


 Roger Herling
 Sr. Desktop Administrator
 Marist College
 PH: 845-575-3347



 From:
 Andy Kurth andy_ku...@ncsu.edu
 To:
 vcl-dev@incubator.apache.org
 Date:
 05/14/2010 10:03 AM
 Subject:
 Re: Change image timeout



 I'm assuming you're using xCAT 1.3.  Would it be possible to configure
 xCAT to
 store the images in /opt/image/x86?  The VCL xCAT code really needs the
 image
 path to contain a 3rd component because it determines where the images
 reside as
 follows:

 managementnode.installpath/OS.sourcepath/image.architecture

 If you can configure xCAT to use /opt/image/x86, then
 managementnode.installpath
 should be set to /opt and I believe the path should be constructed
 correctly.

 Hope this helps,
 Andy

 Roger Herling wrote:
  xCAT root path is /opt/xcat according to the vcld log
 
  but the images are actuially store in /opt/x86
 
  Roger Herling
  Sr. Desktop Administrator
  Marist College
  PH: 845-575-3347
 
 
 
  From:
  Andy Kurth andy_ku...@ncsu.edu
  To:
  vcl-dev@incubator.apache.org
  Date:
  05/11/2010 10:49 AM
  Subject:
  Re: Change image timeout
 
 
 
  I believe the problem is related to the Install Path setting for your
  management
  node:
  Management Nodes  Edit Management Node Information  Edit
 
  Based on the paths being used, I'm guessing yours is set to
 'storage/x86'.
 
  Where does the partimage .gz file reside?  If it actually resides in
  '/storage/image/x86' then change the Install Path setting to 'storage'.
 
  The .tmpl file residing in this directory may indicate another problem
  with the
  xCAT root path.  Where is xCAT installed and which version of xCAT are
 you
 
  using?  Search your log file for 'xCAT root path found:'.  What path
 does
  this
  line contain?
 
  Regards,
  Andy
 
 
  Roger Herling wrote:
  [r...@vcl ~]# du -c
  /storage/x86/image/x86/winvista-Windows7FacStaff49-v0*
  4   /storage/x86/image/x86/winvista-Windows7FacStaff49-v0.tmpl
  4   total
 
  [r...@vcl ~]# ls -l
  /storage/x86/image/x86/winvista-Windows7FacStaff49-v0*
  -rw-r--r-- 1 root root 352 May 10 12:09
  /storage/x86/image/x86/winvista-Windows7FacStaff49-v0.tmpl
 
 
  Roger Herling
  Sr. Desktop Administrator
  Marist College
  PH: 845-575-3347
 
 
 
  From:
  Andy Kurth andy_ku...@ncsu.edu
  To:
  vcl-dev@incubator.apache.org
  Date:
  05/11/2010 09:06 AM
  Subject:
  Re: Change image timeout
 
 
 
  The command that is being executed to check the size should be:
  du -c /storage/x86/image/x86/winvista-Windows7FacStaff49-v0*
 
  Try running this.  What output does it generate?  The last line should
  look like:
   total
 
  The code appears to be finding the 'total' line but is either not
  parsing
  it
  correctly, not calculating the size correctly, or the du command isn't
  correct.
 
  Also, it would be helpful if you include the output from:
  ls -l /storage/x86/image/x86/winvista-Windows7FacStaff49-v0*
 
  -Andy
 
 
  Waldron, Michael H wrote:
  So you never answered whether you watched the capture occurring from
  the
  node console. Can you watch the console and see the capture running
  while
  monitoring the log that shows the capture file is size 0?
  What about any other error messages for the rest of the log capture
 for
 
  this imaging request?
  Mike Waldron
  Systems Specialist
  ITS Research Computing
  University of North Carolina at Chapel Hill
  CB #3420, ITS Manning, Rm 2509
  919-962-9778
 
 
  -Original Message-
  From: Roger Herling [mailto:roger.herl...@marist.edu]
  Sent: Monday, May 10, 2010 4:18 PM
  To: vcl-dev@incubator.apache.org
  Subject: RE: Change image timeout
 
  yup
  Roger Herling
  Sr. Desktop Administrator
  Marist College
  PH: 845-575-3347
 
 
 
  From:
  Waldron, Michael H mwald...@email.unc.edu
  To:
  'vcl-dev@incubator.apache.org' vcl-dev@incubator.apache.org
  Date:
  05/10/2010 04:17 PM
  Subject:
  RE: Change image timeout
 
 
 
  So you can do an ls -l
  /storage/x86/image/x86/winvista-Windows7FacStaff49-v0 and it shows
  some
  size?
 
 
 
  Mike Waldron
  Systems Specialist
  ITS Research Computing
  University of North Carolina at Chapel Hill
  CB #3420, ITS Manning, Rm 2509
  919-962-9778
 
 
  -Original Message-
  From: Roger Herling [mailto:roger.herl...@marist.edu]
  Sent: Monday, May 10, 2010 4:03 PM
  To: vcl-dev@incubator.apache.org
  Subject: RE: Change image timeout
 
  The image is definately being captured and I can force the image to be

  used after it fails by manually undeleting it from the mysql database
  Roger Herling
  Sr. Desktop Administrator
  Marist College
  PH: 845-575-3347
 
 
 
  From:
  Waldron, Michael H mwald...@email.unc.edu
  To:
  'vcl-dev@incubator.apache.org' vcl-dev@incubator.apache.org
  Date:
  05/10/2010 04:01 PM
  Subject:
  RE: Change image timeout
 
 
 
 

Re: Linux base image creation

2010-05-17 Thread Andy Kurth

Hello,
I realize I worded the last message poorly.  I meant to run ps to list the sshd 
processes so you could see the command that was used to start them.  This would 
tell you whether or not the -f options was being used when ext_ssh was started. 
 The problems you are seeing are mainly with the ext_ssh file.  The 
instructions were written for Redhat and need to be reworked for Ubuntu because 
its service startup scripts have a different structure.


I took a look at an Ubuntu installation.  There is no OPTIONS variable in 
/etc/init.d/ssh so the OPTIONS= line you added has no effect.  Instead, it 
uses a SSHD_OPTS variable.  Find the init-functions line and add the following 
line after it:


SSHD_OPTS=-f /etc/ssh/external_sshd_config

Change all sshd.pid strings to ext_sshd.pid.  The following sed command 
should work:

sed -i -r -e s/(ext_)?sshd\.pid/ext_sshd.pid/g /etc/init.d/ext_ssh

I have attached a script I used to configure sshd on my Ubuntu test image.  It's 
pretty raw but it works for me.  Please reply if you have any problems with it. 
 I'll update the documentation with this script if it's working properly.


Hope this helps,
Andy

Kiran N wrote:

Thanks Andy for the response!

After stopping all the SSH services, I restarted the external ssh by the
command
/etc/init.d/ext_ssh start
and as you said, ext_ssh is listening on the private IP address.
I am attaching the ssh, ext_ssh and external_ssh_config files.
Also the output for command used to start the external sshd process:
ps -ef | grep sshd
is not as you said.

Hope this helps to figure out the problem!
#!/bin/bash

function set_config {
if [ $# -ne 3 ]
then
echo usage: set_config [config_file] [keyword] [value]
exit 1
fi

config_file=$1
keyword=$2
value=$3

if [ $value == '#' ]
then
#echo Commenting $keyword lines in $config_file
sed -i -r -e s/^[ #]*($keyword .*)/#\1/ $config_file
else
if [ `grep -i -r -c ^[ #]*$keyword  $config_file` == '0' ]
then
#echo Adding $keyword value to $config_file
echo $keyword $value  $config_file
else
escaped_value=$(echo $value | sed -e 's/\//\\\//g')
#echo Setting $keyword to $value in $config_file
 sed -i -r -e s/^[ #]*($keyword).*/\1 $escaped_value/ $config_file
fi
fi

#grep -i -r ^[ #]*$keyword $config_file

return 1;
}

clear
cp /etc/ssh/sshd_config /etc/ssh/sshd_config.ORIG

set_config '/etc/ssh/sshd_config' 'StrictModes' 'no'
set_config '/etc/ssh/sshd_config' 'X11Forwarding' 'yes'
set_config '/etc/ssh/sshd_config' 'KeyRegenerationInterval' '0'
set_config '/etc/ssh/sshd_config' 'MaxStartups' '#'

cp /etc/ssh/sshd_config /etc/ssh/external_sshd_config

set_config '/etc/ssh/external_sshd_config' 'PidFile' '/var/run/ext_sshd.pid'

sed -i -r -e s/^[ #]*AllowUsers.*//g /etc/ssh/sshd_config
sed -i -r -e s/^[ #]*AllowUsers.*//g /etc/ssh/external_sshd_config

sed -i -r -e s/^[ #]*ListenAddress.*//g /etc/ssh/sshd_config
sed -i -r -e s/^[ #]*ListenAddress.*//g /etc/ssh/external_sshd_config

IP0=$(ifconfig eth0 | grep 'inet addr' | awk '{print $2}' | awk -F: '{print 
$2}')
IP1=$(ifconfig eth1 | grep 'inet addr' | awk '{print $2}' | awk -F: '{print 
$2}')
echo IP eth0: $IP0
echo IP eth1: $IP1

echo AllowUsers root  /etc/ssh/sshd_config

echo ListenAddress $IP0  /etc/ssh/sshd_config
echo ListenAddress $IP1  /etc/ssh/external_sshd_config

cp /etc/init.d/ssh /etc/init.d/ext_ssh
sed -i -r -e s/(ext_)?sshd\.pid/ext_sshd.pid/g /etc/init.d/ext_ssh
sed -i -r -e s/\sshd\/\ext_sshd\/g /etc/init.d/ext_ssh
sed -i -r -e s/(.*init-functions)/\1\n\nSSHD_OPTS=\-f 
\/etc\/ssh\/external_sshd_config\/ /etc/init.d/ext_ssh

echo
echo Stopping sshd services...
service ssh stop
sleep 2
service ext_ssh stop
sleep 2

rm -f /var/run/*sshd*pid

echo
echo Starting sshd services...
service ssh start
sleep 2
service ext_ssh start

echo
echo sshd processes:
pgrep -fl sbin.sshd
echo
echo sshd.pid: `cat /var/run/sshd.pid`
echo ext_sshd.pid: `cat /var/run/ext_sshd.pid`


Re: VCL 2.2 with VMware Server 2.x or ESXi 4.x

2010-05-17 Thread Andy Kurth
This is just an update.  I have not committed the updated VMware code yet.  I 
have been trying to find out if there is any way for VCL to manage the latest 
free version of ESXi.  I don't think this will be possible.  Once a free license 
code is entered and the host leaves evaluation mode, most functions result in a 
RestrictedVersion fault.  The new code should work on ESX, Server 2.x, and 
ESXi if it is licensed or in evaluation mode.  I need to tidy some things up 
before committing, hopefully this week.


-Andy

Jose Higino wrote:

Thanks Andy,

I was trying to get a try this week next week will be out of schedule 
for me.. but nice to know about.. i will ramp up with that 3 weeks from 
now.


Regards