Bccd-ng - July-8-2012

From BCCD 3.0

(Difference between revisions)
Jump to: navigation, search
(Created page with "== BCCD ISO TESTING REPORT == {| {{prettytable}} |Author |Mobeen Ludin |- |Date Generated |July 7 / 2012 |- |Date Tested |July 8 / 2012 |- |Release Revision # |r3742 |- |} == S...")
 
(One intermediate revision not shown)
Line 78: Line 78:
====64-bit ====  
====64-bit ====  
-
Most of the Modules such as: GalaxSee, Life, Param_space, and CUDA programs worked fine.  
+
Most of the Modules such as: GalaxSee, Life, Param_space etc.. worked fine. CUDA had path issue. Error = ld: cannot find -lcudart.
 +
 
 +
It was solved by setting up the right LDFLAGS paths to    -L/bccd/software/cuda/3.2.16/Linux/BCCD/x86_64/lib64. This should be fixed for the Module also.  
'''Network Managment'''
'''Network Managment'''
Line 86: Line 88:
       $ sudo /bin/bccd-reset-network  
       $ sudo /bin/bccd-reset-network  
-
worked fine: and generated no errors.   
+
worked fine: and generated no errors.  although the issue of listing for dhcp queries on the external nic still existed. Which we found a fix for it
 +
 
 +
The Default for dhcp initscript interfaces was set to NULL. by changing it to eth0, now it only listens to queries on its internal nic.
-
 
'''Liberating BCCD'''
'''Liberating BCCD'''
$ sudo perl /root/liberate.pl --libdev /dev/sda
$ sudo perl /root/liberate.pl --libdev /dev/sda
    
    
-
Liberation pased 78 out of 78 tests.
+
Liberation pased 77 out of 78 tests. Test # 8 failed.
 +
 
 +
not ok 8 - Set LVM PV partition type.
===LIBERATED -- From Hard Disk===  
===LIBERATED -- From Hard Disk===  
Line 105: Line 110:
$ sudo /bin/bccd-reset-network
$ sudo /bin/bccd-reset-network
        
        
-
Network setup workd smoothly. No Errors.  
+
Network setup workd smoothly. No Errors. Same fix as above.
'''PXE-BOOTING'''
'''PXE-BOOTING'''
-
- PXE-Booting is working. However, with a some errors. The first computing node comes up fine. When you run:
+
- PXE-Booting is working.
-
    $ bccd-snarfhosts
+
-
    $ cat machines
+
-
you will see the computing nodes, which you can ping and ssh to it. However when you turn on the second computing node, it will come up, but the first computing node loses NFS. I think for some resons bccd-allowall is sending the node to sleeping. SEE THE PICTURES FOR ERRORS...
+
-
 
+
      
      
-
- All the Modules both compiles and run fine with openmpi, but with mpich2 Param_Space and Life does not compile, See pictures for ERRORS.
+
- All the Modules both compiles and run fine with openmpi and mpich2.
-
 
+
-
      http://cluster.earlham.edu/bccd-ng/testing/mobeen/ERROR-PICS/mpich2-module-errors.jpg
+
-
      http://cluster.earlham.edu/bccd-ng/testing/mobeen/ERROR-PICS/life-mpich2-errors.jpg
+
==== 32-bit ====   
==== 32-bit ====   
Line 126: Line 124:
$ sudo /bin/bccd-reset-network
$ sudo /bin/bccd-reset-network
        
        
-
Network setup workd smoothly. No Errors. PXE-Booting has the same issues as 64-bit.  
+
Network setup workd smoothly. No Errors. but still listing on its external nic for dhcp queries. Same fix for it as mentioned above.
-
     - all the Moduels do not compile or run on the 32-bit.  
+
     - all the Moduels do not compile or run on the 32-bit.
-
      - With openmpi error is: ld: Cannot fine -lmpi_cxx
+
-
      - With mpich2 error is: make: mpicxx: Command not found
+
      
      
-
     - CUDA: compiles fine, but does not generate the right results. for example, ./hello-world 10 10, both comlumns are 0 0.  
+
     - CUDA: compiles fine, but does not generate the right results. for example, ./hello-world 10 10, both comlumns are 0 0 on the livemode. but on the liberated mode it works fine. 
== Final: Comments/Suggestions/Objection/Question/?/? ==
== Final: Comments/Suggestions/Objection/Question/?/? ==
-
I will work on in tomorrow and start working on milestone tickest and get ready for the new release.  
+
We seem to be ready for a release Candidate.  
-
 
+
I could generate a Release Candidate when I get the thumbs up...
-
link to Error Pictures:  http://cluster.earlham.edu/bccd-ng/testing/mobeen/ERROR-PICS/
+

Latest revision as of 15:49, 9 July 2012

Contents

BCCD ISO TESTING REPORT

Author Mobeen Ludin
Date Generated July 7 / 2012
Date Tested July 8 / 2012
Release Revision # r3742

Summary

--This iso was generated as testing for Release Canadidate and soon new BCCD release. Forllowing tickets are bring resolved in these Testing ISOs:

        ISOs that fix issues:  #715, #716, #717, #718, #720, #721

Downloading and Making Bootable USB

First Downloading the ISO form a repository on earlham cs cluster:

first downloading the script for making bootable USB drive

  $ svn co svn+ssh://mmludin08@cluster.earlham.edu/cluster/svnroot/bccd-ng/trunk/bin BCCD
  $ cd BCCD 

This is where the the script exits for making bootable USB from BCCD ISO

Download the ISO from appropriate links

--64-Bit

  $ wget http://cluster.earlham.edu/bccd-ng/snapshot/bccd_latest.amd64.iso.md5
  $ wget http://cluster.earlham.edu/bccd-ng/snapshot/bccd_latest.amd64.iso

--32-Bit

  $ wget http://cluster.earlham.edu/bccd-ng/snapshot/bccd_latest.i386.iso.md5
  $ wget http://cluster.earlham.edu/bccd-ng/snapshot/bccd_latest.i386.iso

CheckSum Report:

To see if the ISO was downloaded 100% correctly

  $ md5sum bccd_latest.amd64.iso

92af26a2499d88f67149aa90b6bad055 bccd_latest.amd64.iso

  $ cat bccd_latest.amd64.iso.md5 

MD5 (bccd_latest.amd64.iso) = 92af26a2499d88f67149aa90b6bad055

  $ md5sum bccd_latest.i386.iso

443eb59f7a41439cd7c953d5d00c3a86 bccd_latest.i386.iso

  $ cat bccd_latest.i386.iso.md5 

MD5 (bccd_latest.i386.iso) = 443eb59f7a41439cd7c953d5d00c3a86

Second Making Bootable USB

The next step will be to insert the 4G USB drive (the Intel one) into the machine, wait a couple seconds, and then do

   $ sudo ./build_bootable_USB.sh /dev/sdb bccd_latest.amd64.iso

This is the usual procedure for running the script. The thing that can vary is /dev/sdb; usually this will be the location of the USB, but if there is another USB connected it might actually be /dev/sdc.

NOTE: To see find out the memory partitions run the following command:

   $ ls /dev/sd*

or

    $ dmesg

ISO TESTING REPORT

LIVEMODE --From USB

64-bit

Most of the Modules such as: GalaxSee, Life, Param_space etc.. worked fine. CUDA had path issue. Error = ld: cannot find -lcudart.

It was solved by setting up the right LDFLAGS paths to -L/bccd/software/cuda/3.2.16/Linux/BCCD/x86_64/lib64. This should be fixed for the Module also.

Network Managment

     $ bccd-nic-setup 
     $ sudo /bin/bccd-reset-network 

worked fine: and generated no errors. although the issue of listing for dhcp queries on the external nic still existed. Which we found a fix for it

The Default for dhcp initscript interfaces was set to NULL. by changing it to eth0, now it only listens to queries on its internal nic.


Liberating BCCD

$ sudo perl /root/liberate.pl --libdev /dev/sda

Liberation pased 77 out of 78 tests. Test # 8 failed.

not ok 8 - Set LVM PV partition type.

LIBERATED -- From Hard Disk

64-bit

Network Setup Managment

$ bccd-nic-setup $ sudo /bin/bccd-reset-network

Network setup workd smoothly. No Errors. Same fix as above.

PXE-BOOTING

- PXE-Booting is working.

- All the Modules both compiles and run fine with openmpi and mpich2.

32-bit

Network Setup Managment

$ bccd-nic-setup $ sudo /bin/bccd-reset-network

Network setup workd smoothly. No Errors. but still listing on its external nic for dhcp queries. Same fix for it as mentioned above.

   - all the Moduels do not compile or run on the 32-bit.
   
   - CUDA: compiles fine, but does not generate the right results. for example, ./hello-world 10 10, both comlumns are 0 0 on the livemode. but on the liberated mode it works fine.  

Final: Comments/Suggestions/Objection/Question/?/?

We seem to be ready for a release Candidate. I could generate a Release Candidate when I get the thumbs up...

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox