Installation of Globus Toolkit 4.0.8 on Scientific Linux 5.5 running virtually within VMware on openSUSE 11.1 ¶
In order to run Globus Toolkit in a virtual machine you need:
- A host (here: ursaminor, 129.206.110.201)
- A basic operating system (here: openSUSE 11.1)
- A virtualizer (here: VMware)
- An operation system on the virtual machine (here: Scientific Linux 5.5)
- A Globus Toolkit installation (here: Globus Toolkit 4.0.8)
- A personal User Certificate
Installation of openSUSE 11.1 ¶
- Download openSUSE 11.1 (decide and retain 32 or 64 bit)
- Install openSUSE 11.1
- Use default settings
- Choose KDE 3 as desktop
- Use Yast ("System" → "Date and Time") to synchronize with NTP Server (here: 129.206.110.100)
Hint: There is VMware support up to Linux Kernel 2.6.38. Hence openSUSE 11.1 (2.6.27), openSUSE 11.2 (2.6.31), openSUSE 11.3 (2.6.34) and openSUSE 11.4 (2.6.37), but not openSUSE 12.1 (3.x)!
Installation of VMware ¶
Download ¶
- Download VMware Server
- Free of charge
- You have to register
- You get a registration code
- Download VMware Player
- Download vmware-server-2.0.x-kernel-2.6.3x-install.sh
Install ¶
- Use the install script as described by Radu Cotescu to install VMware Server
- Use default settings
- You need the registration code for the VMware Server
- Install VMware Player
Installation of Scientific Linux 5.5 ¶
Download ¶
- Download Scientific Linux 5.5 Install DVD (32 or 64 bit)
Install ¶
- Run VMware Player
- Make settings according to Deliverable 5.1 p. 21 et seqq.
- Install Scientific Linux from image
- Make settings according to Deliverable 5.1 p. 28 et seqq.
- Improve VMware Player
- Update (automatic reminder while starting VMware Player)
- Install VMware tools (start from menu while virtual machine is running, see Deliverable 5.1 p. 46 et seqq.)
- Synchronize virtual machine time with host time
- vmware-toolbox
- Check "Time synchronization between the virtual machine and the host operating system"
- "Close"
- Make network connection to virtual machine (here: 129.206.112.253 respectively dgsi.zah.uni-heidelberg.de)
- Ask your system administrator (here: Dr. Peter Schwekendiek)
Virtual_Globus_Toolkit_A.tar.bz2¹
Setup of Globus Toolkit 4.0.8 ¶
Virtual_Globus_Toolkit_B.tar.bz2¹
Setting up Globus Toolkit 4.0.8 Security Infrastructure �
Setting up Globus Toolkit 4.0.8 Security Infrastructure
Virtual_Globus_Toolkit_C.tar.bz2¹
Setup of Information Provider MDS4 and GeoMaint
Setup of Information Provider MDS4 and GeoMaint
Virtual_Globus_Toolkit_D.tar.bz2¹
Check Installation ¶
- gsissh dgsi.zah.uni-heidelberg.de
Submit Tests ¶
Try globusrun-ws -submit ... dgsi.zah.uni-heidelberg.de ... with other hosts, too, for the following three tests:
Basic ¶
- globusrun-ws -submit -s -F dgsi.zah.uni-heidelberg.de -so out -se err -c /bin/date
Delegating user credentials...Done. Submitting job...Done. Job ID: uuid:cd291162-4b35-11e1-96d7-000c296d7bcb Termination time: 01/31/2012 11:30 GMT Current job state: CleanUp-Hold Current job state: CleanUp Current job state: Done Destroying job...Done. Cleaning up any delegated credentials...Done.
- more err
- more out
Mon Jan 30 12:30:25 CET 2012
- rm err
- rm out
Simple ¶
- vi simple.xml
- Edit (i)
<job> <executable>/bin/date</executable> </job>
- Save (ESC :wq)
- Edit (i)
- globusrun-ws -submit -s -F dgsi.zah.uni-heidelberg.de -job-description-file simple.xml
Delegating user credentials...Done. Submitting job...Done. Job ID: uuid:fdc54d9e-4b36-11e1-8057-000c296d7bcb Termination time: 01/31/2012 11:38 GMT Current job state: Active Current job state: CleanUp-Hold Mon Jan 30 12:38:56 CET 2012 Current job state: CleanUp Current job state: Done Destroying job...Done. Cleaning up any delegated credentials...Done.
Hint: Output of /bin/date is "Mon Jan 30 12:38:56 CET 2012"!
Complex ¶
- vi program.sh
- Edit (i)
echo "This is the output of program.sh!"
- Save (ESC :wq)
- Edit (i)
- chmod +x program.sh
- vi complex.xml
- Edit (i)
<job> <executable>/bin/bash</executable> <argument>programme.sh</argument> <stdout>${GLOBUS_USER_HOME}/output</stdout> <stderr>${GLOBUS_USER_HOME}/errors</stderr> <fileStageIn> <transfer> <sourceUrl>gsiftp://dgsi.zah.uni-heidelberg.de/home/agrid/agrid107/program.sh</sourceUrl> <destinationUrl>file:///${GLOBUS_USER_HOME}/programme.sh</destinationUrl> </transfer> </fileStageIn> <fileStageOut> <transfer> <sourceUrl>file:///${GLOBUS_USER_HOME}/output</sourceUrl> <destinationUrl>gsiftp://dgsi.zah.uni-heidelberg.de/home/agrid/agrid107/out</destinationUrl> </transfer> <transfer> <sourceUrl>file:///${GLOBUS_USER_HOME}/errors</sourceUrl> <destinationUrl>gsiftp://dgsi.zah.uni-heidelberg.de/home/agrid/agrid107/err</destinationUrl> </transfer> </fileStageOut> <fileCleanUp> <deletion> <file>file:///${GLOBUS_USER_HOME}/output</file> </deletion> <deletion> <file>file:///${GLOBUS_USER_HOME}/errors</file> </deletion> <deletion> <file>file:///${GLOBUS_USER_HOME}/programme.sh</file> </deletion> </fileCleanUp> </job>
- Save (ESC :wq)
- Edit (i)
- globusrun-ws -submit -S -F dgsi.zah.uni-heidelberg.de -f complex.xml
Delegating user credentials...Done. Submitting job...Done. Job ID: uuid:1b5e3ac8-4b41-11e1-bc46-000c296d7bcb Termination time: 01/31/2012 12:51 GMT Current job state: StageIn Current job state: Active Current job state: StageOut Current job state: CleanUp Current job state: Done Destroying job...Done. Cleaning up any delegated credentials...Done.
- ls -l
total 712 -rw-r--r-- 1 agrid107 agrid 1140 Jan 30 13:48 complex.xml -rw-r--r-- 1 agrid107 agrid 0 Jan 30 13:51 err -rw-r--r-- 1 agrid107 agrid 34 Jan 30 13:51 out -rwxr-xr-x 1 agrid107 agrid 41 Jan 30 13:51 program.sh -rw-r--r-- 1 agrid107 agrid 50 Jan 30 12:38 simple.xml
- more err
- more out
This is the output of program.sh!
- rm err
- rm out
RFT Test ¶
- cp $GLOBUS_LOCATION/share/globus_wsrf_rft_test/transfer.xfr /tmp
- vi /tmp/transfer.xfr
- Edit (i)
true 16000 16000 false 1 true 1 null null false 10 gsiftp://dgsi.zah.uni-heidelberg.de:2811/tmp/rftTest.tmp gsiftp://dgsi.zah.uni-heidelberg.de:2811/tmp/rftTest_Done.tmp
- Save (ESC :wq)
- Edit (i)
- touch /tmp/rftTest.tmp
- rft -h dgsi.zah.uni-heidelberg.de -f /tmp/transfer.xfr
Number of transfers in this request: 1 Subscribed for overall status Termination time to set: 60 minutes Overall status of transfer: Finished/Active/Failed/Retrying/Pending 1/0/0/0/0 Overall status of transfer: Finished/Active/Failed/Retrying/Pending 0/1/0/0/0 All Transfers are completed
Gatekeeper Test ¶
- globus-personal-gatekeeper -start
GRAM contact: dgsi.zah.uni-heidelberg.de:40276:/C=DE/O=GermanGrid/OU=ZAH/CN=Klaus Rieger
In the following line take the number you got before:
- globus-job-run "dgsi.zah.uni-heidelberg.de:40276:/C=DE/O=GermanGrid/OU=ZAH/CN=Klaus Rieger" /bin/hostname
dgsi.zah.uni-heidelberg.de
- globus-personal-gatekeeper -killall
killing gatekeeper: "dgsi.zah.uni-heidelberg.de:40276:/C=DE/O=GermanGrid/OU=ZAH/CN=Klaus Rieger"
Webservice Test ¶
- counter-client -s https://dgsi.zah.uni-heidelberg.de:8443/wsrf/services/CounterService
Got notification with value: 3 Counter has value: 3
Documentation ¶
The following guide is available:
Virtual Machine ¶
Add Shared Folder:
- "Virtual Machine" → "Virtual Machine Settings"
- "Options" → "Shared Folders"
- Check "Always enabled"
- "Add..."
- Name: testbed_share
- Host Path: /home/rieger/testbed_share
- Checked: "Enabled"
- Unchecked: "Read-only"
- "OK"
- "Save"
- "Options" → "Shared Folders"
Shutdown virtual machine:
- "File" → "Suspend and Quit"
Add script to shared folder:
- Get script generate_nodes.sh
- scp generate_nodes.sh rieger@129.206.110.201:/home/rieger/generate_nodes.sh
Zip ¶
Update Zip ¶
To pack files bigger than 2 GB you need at least zip version 3.0! Therefore an update is necessary in case of openSUSE 11.1.
- Get from download.opensuse.org/source/factory/repo/oss/suse/src/ the most recent version of:
Copy file to host ursaminor:
- scp zip-3.0-7.3.src.rpm root@129.206.110.201:/root/
Login to host ursaminor as root, build and install zip:
- ssh root@129.206.110.201
- rpmbuild --rebuild zip-3.0-7.3.src.rpm
- cd /usr/src/packages/RPMS/i586/
- ls
zip-3.0-7.3.i586.rpm
- rpm -Uvh zip-3.0-7.3.i586.rpm
- exit
Create Zip File ¶
Login to host ursaminor as rieger and create zip-file:
- ssh rieger@129.206.110.201
- cd vmware/
- ls
Virtual_Globus_Toolkit
- zip -r fullblown.zip Virtual_Globus_Toolkit/
fullblown.zip¹
Automated generation of Virtual Machines ¶
Shared Folder ¶
- cd ~
- mkdir testbed_share
- ls
Desktop Documents generate_nodes.sh testbed_share vmware
- mv generate_nodes.sh testbed_share/
- mv vmware/Virtual_Globus_Toolkit_C/ testbed_share/
- exit
Run Script on Virtual Machine ¶
Restart virtual machine, then proceed:
- ssh globus@dgsi.zah.uni-heidelberg.de
- cd /mnt/hgfs/testbed_share/
- ls
generate_nodes.sh Virtual_Globus_Toolkit_C
- vi generate_nodes.sh
- Search "VM_FULL=" (?VM_FULL=)
- Change (i)
... VM_FULL=/root/fullblown #your prepared reference "full-blown" VM is to be found here DEST_DIR=/root/dgsi #will be created. If exist, will be deleted first. ...
to... VM_FULL=/mnt/hgfs/testbed_share/Virtual_Globus_Toolkit_C #your prepared reference "full-blown" VM is to be found here DEST_DIR=/mnt/hgfs/testbed_share/dgsi #will be created. If exist, will be deleted first. ...
- Save (ESC :wq)
- Change (i)
- Search "VM_FULL=" (?VM_FULL=)
- su
- chmod +x generate_nodes.sh
- ./generate_nodes.sh
Creating nodes for location pc, subnet 192.168.190.0 Creating node pc10 ... Starting virtual machine pc10 ...
1) available at medigrid-srv.gwdg.de/data/DGSI/publicshare/VMware_Image_GT408/