ACSLS Management

Log in as or su to ACSSA for cmd_proc:
ssh acssa@acsls-1.loni.ucla.edu
or
su - acssa

Bring CAP or other device offline and back online (in this case CAP 1,1,0) =

vary cap 1,1,0 offline force
vary cap 1,1,0 online

Querying Volumes, Cap, etc

q vol volid
query volume volid
q cap cap id
or simply use q or query and ACSLS will as for input

Eject Tapes (example: eject volumes 500831, 500837 and 500855 through 500865 to CAP 1,1,0)

eject cap ID volIDs
eject 1,1,0 500831 500837 500855-500865

Use volrpt to find wrongly placed tapes to be ejected (i.e. 500 tapes in StorageTek^? 8500):

Log in as ACSSS for shell commands (don't use su to ensure proper environment variables):
volrpt -d -l -1,{0,1,2,3,4,5,6,7,9} |grep 500

No permission to read topic ActiveDirectoryBackup - perhaps you need to log in?

No permission to read topic AddUser - perhaps you need to log in?

No permission to read topic AdminScripts - perhaps you need to log in?

No permission to read topic AgentSmith - perhaps you need to log in?

hasdhshahjasd h asdh jkhkjasdhjh

Install Apache 2.0.58:

cd /tmp wget http://apache.mirrormax.net/httpd/httpd-2.0.58.tar.gz sudo tar -xvzf httpd-2.0.58.tar.gz cd httpd-2.0.58 sudo ./configure --prefix=/usr/local/apache2 --enable-static-support sudo make sudo make install

If you want to move apache to a system different from the compiling system, /usr/local/apache2 should be transportable.

Startup scripts: sudo ln -s /usr/local/apache2/bin/apachectl /etc/init.d/httpd

For each applicable rc*.d (i.e. init 3, 5, etc) add a symlink to have apache start on boot: sudo ln -s ../init.d/httpd S50httpd

No permission to read topic Infrastructure.CR_411Config - perhaps you need to log in?

Cranium: Compilers

This page contains instructions on how to use the different compilers we have installed in the Cranium cluster like the GCC, Intel and PGI.

GCC Intel PGI

---

GCC

Here is a guide on how to use the GCC compiler. Here is a general description on how to compile with C++:

How do I use gcc, g++, and gdb? If you want to use the GNU C (gcc) compiler to compile your code from the command line just type: % gcc file.c or % g++ file.c This compiles file.c into an executable binary named a.out. If you use Makefiles to compile your programs, then put the following line at the top of your Makefile: CC=gcc or CC=g++ By default our GCC compiler build executables for 64 bits, but if you need to build an executable for 32 bits, set the following variables in the command line or in your Makefile: setenv CFLAGS "-O2 -m32" setenv CXXFLAGS "-O2 -m32" Here are a few options to gcc and g++: -o outputfile To specify the name of the output file. The executable will be named a.out unless you use this option. -g To compile with debugging flags, for use with gdb. -L dir To specify directories for the linker to search for the library files. -l library This specifies a library to link with. -I dir This specifies a directories for the compile to search for when looking for include files. For a complete set of options that control compile optimization, use this link. Now, to check errors on a "correctly" compiled binary you can use the debugger 'gdb'. Here is a typical example of a gcc/gdb session: % cat hello.c #include<stdio.h> main() { int count; for (count=0;count<10;count++) printf("Hello from LONI!\n"); } % gcc -g hello.c % gdb ./a.out GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.13 (sparc-sun-solaris2.3), Copyright 1994 Free Software Foundation, Inc... (gdb) b main Breakpoint 1 at 0x10784: file hello.c, line 6. (gdb) r Starting program: /home1/b/bozo/./a.out Breakpoint 1, main () at hello.c:6 6 for (count=0;count<10;count++) (gdb) s 7 printf("Hello from LONI!\n"); (gdb) p count $1 = 0 (gdb) disp count 1: count = 0 (gdb) set count=8 (gdb) s Hello from LONI! 6 for (count=0;count<10;count++) 1: count = 8 (gdb) 7 printf("Hello from LONI!\n"); 1: count = 9 (gdb) c Continuing. Hello from LONI! Program exited with code 01. (gdb) q % Here are a few gdb commands: help Will give you help on most gdb functions. If you wish for help on a specific command, type help command. b function-name To set a breakpoint at a function. r args To run the program. It will run until it reaches a breakpoint. s To single-step through lines of code. c To continue until the next breakpoint. p variable To print a variable's value. q To quit gdb.

Intel

The Intel C++ Compiler lets you build and optimize C/C++ applications for Intel IA-32, Intel Extended Memory 64 Technology (Intel EM64T), and Intel Itanium-based systems running Linux operating systems. You can check a detailed user guide by following this link.

For the 64 bit compiler (please check the appropriate paths),

setenv ICCPATH /usr/local/intel_cc-10.1.008_64bit
setenv PATH bin:${PATH}

This should enable you to compile and link programs using the Intel compiler. In case you are using Makefiles, it's also helpful to set up the path as

Use the Intel compiler,

CC = /usr/local/intel_cc-10.1.008_64bit/bin/icc

Use the Intel Linker,

LD = /usr/local/intel_cc-10.1.008_64bit/bin/icpc

Optimizing with the Intel compiler, most gcc options such as -g, -O2, -O1 should work. Additionally, exclusive optimizing options are provided as follows:

For parallelizing loops,

-parallel

Enables the auto-parallelizer to generate multithreaded code for loops that can be safely executed in parallel.

General faster optimization,

-fast

Provides a single, simple optimization that enables a collection of optimizations that favor run-time performance.

For targetting specific processors,

-mtune=itanium2

or

-mtune=itanium

For generating vectorized code,

-xS

Can generate SSE4 Vectorizing Compiler and Media Accelerators instructions for future Intel processors that support the instructions. Can generate SSSE3, SSE3, SSE2, and SSSE4 instructions and it can optimize for future Intel processors.

-xT

Can generate SSSE3, SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for the Intel Core Duo processor family.

-xP

Can generate SSE3, SSE2, and SSE instructions for Intel processors, and it can optimize for processors based on Intel Core microarchitecture and Intel NetBurst microarchitecture, like Intel Core Duo processors, Pentium 4 processors with SSE3, and Intel Xeon processors with SSE3.

Linking with the intel compiler,

icpc -L$(LIBDIR) <.o files> -o

Additionally, if one is using C++ and requires -lstdc++, it's better to specify -lguide -lpthread before -lstdc++ to avoid linker errors.

PGI

To compile PGI applications for 32 bits, please set the following variables (if you are using a tcsh shell):

Set the PGI variable,

setenv PGI /usr/local/pgi-7.1.3_64bit

Set the variable for the license file,

setenv LM_LICENSE_FILE ${PGI}/license.dat

Include PGI in your current path,

setenv PATH ${PGI}/linux86/7.1-3/bin:${PATH}

Set the compiler variables,

setenv CC ${PGI}/linux86/7.1-3/bin/pgCC
setenv CFLAGS "-O3 -Bstatic_pgi"

If you have to set additional include directories,

INCLUDES = -I/needed_include_path

And, for additional libraries needed when compiling your code,

INCLUDES = -L/needed_library_path

For 64 bits, follow a similar procedure like with 32 bits:

setenv PGI /usr/local/pgi-7.1.3_64bit setenv PGI /usr/local/pgi-7.1.3_64bit
setenv LM_LICENSE_FILE ${PGI}/license.dat
setenv PATH ${PGI}/linux86-64/7.1-3/bin:${PATH}
setenv CC ${PGI}/linux86-64/7.1-3/bin/pgCC
setenv CFLAGS "-O3 -tp x64 -Bstatic_pgi"

A complete guide for PGI can be found by clicking here. Check also:

PGI User's guide,
PGI Tools guide,
PGI Compiler suite.

Cerebro Rocks: Cluster's Applications

We have a variety of applications to be used by the LONI users who access the Cranium cluster. We use cerebro-rcc.loni.ucla.edu to build/install/upgrade a needed application. Once the application is ready to be used on production, we proceed to copy it into cerebro-rocks.loni.ucla.edu and then, propagate it into all the compute nodes on the cluster. We use rsync as method of propagation having cerebro-rocks as the rsync server. The applications installed on the Cranium (Cerebro Rocks) cluster are shown as follow:

Analyis of Functional Neuro Images

• URL: http://afni.nimh.nih.gov/afni
• Current Version: AFNI_2007_05_29_1644
• Installed Version: AFNI_2007_05_29_1644
• Installed at: Aug. 8th, 2007 (hugo.hernandez@loni.ucla.edu)

AFNI is a set of C programs for processing, analyzing, and displaying functional MRI (FMRI) data - a technique for mapping human brain activity. It runs on Unix+X11+Motif systems, including SGI, Solaris, Linux, and Mac OS X. It is available free (in C source code format, and some precompiled binaries) for research purposes.

Building the application

1. Go to build dir
        # cd /tmp/hugo

2. Download binaries from http://afni.nimh.nih.gov/afni/download/afni/releases/latest
        o Download AFNI/SUMA Software latest for Linux (Linux gcc33_64)
        o filename: linux_gcc33_64.tar.gz

3. Uncompress the tar file,         # tar -zxvf linux_gcc33_64.tar.gz
        # cd linux_gcc33_64
        # cp -f * /usr/local/afni-2007529-1644_64bit/bin/.
        # mv /usr/local/afni-2007529-1644_64bit/bin/README* /usr/local/afni-2007529-1644_64bit/.

4. Add AFNI Matlab library
        o Download AFNI Matlab Library latest for all platforms
        o filename: afni_matlab.tgz
        o http://afni.nimh.nih.gov/afni/download/afnimatlab/releases/latest

5. Uncompress the tar file,
        # tar -zxvf afni_matlab.tgz
        # cd afni_matlab
        # mv matlab /usr/local/afni-2007529-1644_64bit/.

6. Add AFNI Sample Data
        o Download AFNI/SUMA Sample Data latest for all platforms
        o filenames: AFNI_data1.tgz, AFNI_data2.tgz
        o http://afni.nimh.nih.gov/afni/download/data/releases/latest

7. Uncompress the tar file,
        # tar -zxvf AFNI_data1.tgz
        # cp -rf AFNI_data1 /usr/local/afni-2007529-1644_64bit/dataSample/.
        # tar -zxvf AFNI_data2.tgz
        # cp -rf AFNI_data2 /usr/local/afni-2007529-1644_64bit/dataSample/.

8. Add SUMA Sample Data
        o Download AFNI/SUMA Sample Data latest for all platforms
        o filenames: SUMA_demo.tgz, SUMA_StandardMeshes.tgz

9. Uncompress the tar file,
        # tar -zxvf SUMA_demo.tgz
        # cp -rf suma_demo /usr/local/afni-2007529-1644_64bit/sumaSample/.
        # tar -zxvf SUMA_StandardMeshes.tgz
        # cp -rf std_mashes /usr/local/afni-2007529-1644_64bit/sumaSample/.

go Up

---

Automated Image Registration

• URL: http://bishopw.loni.ucla.edu/AIR5/index.html
• Current Version: 5.2.5
• Installed Version: 5.2.5
• Installed at: Aug. 7th, 2007 (hugo.hernandez@loni.ucla.edu)

AIR allows automated registration of 3D (and 2D) images within and across subjects and within and sometimes across imaging modalities. AIR source code written in C is available to the research community free of charge. The code can be compiled for UNIX, PC or Macintosh platforms. Only source code is available (no executables).

Building the application

1) Adquired code from /cxfs/tmp/sysadm/air5.2.5
        - Copied to /tmp/air5.2.5

2) ./configure --prefix=/tmp/air5.2.5/bin
        o To compile the extended binaries, edit
                - /tmp/hugo/air5.2.5/Makefile
                - /tmp/hugo/air5.2.5/src/Makefile

         as in Makefile.txt and Makefile-src.txt respectively.
         Be careful with the TAB spaces on the Makefiles.

3) make all

4) make install

5) Copy binaries
        # cp /tmp/air5.2.5/bin/* /usr/local/air5.2.5_64_16/bin/.

6) For 8 and 16 bits binaries, edit src/config.h
        #define AIR_CONFIG_OUTBITS 16
   to
        #define AIR_CONFIG_OUTBITS 8

7) For binaries compiled at 32 bits, set the CFLAGS and CXXFLGS to "-02 -m32" before step (2)

go Up

---

Brainsuite

• URL: http://brainsuite.usc.edu
• Current Version: brainsuite2
• Installed Version: brainsuite2
• Installed at: Aug. 29th, 2007 (hugo.hernandez@loni.ucla.edu)

BrainSuite is a magnetic resonance (MR) image analysis tool designed for identifying tissue types and surfaces in MR images of the human head.

Building the application

1. Copied exec files from David Shattuck's directory:
        # cd /usr/local/brainsuite_32bit/bin
        # cp /nethome/users/shattuck/experimental/i386-redhat-linux-gnu/bfc07a_i386-redhat-linux-gnu .
        # cp /nethome/users/shattuck/experimental/i386-redhat-linux-gnu/bse07a_i386-redhat-linux-gnu .
        # cp /nethome/users/shattuck/experimental/i386-redhat-linux-gnu/pvc07a_i386-redhat-linux-gnu .

go Up

---

Fast Lexical Analyzer Generator

• URL: http://flex.sourceforge.net
• Current Version: 2.5.4a
• Installed Version: 2.5.31
• Installed at: Aug. 29th, 2007 (hugo.hernandez@loni.ucla.edu)
• Package needed by netbpm

Flex is a tool for generating programs that recognize lexical patterns in text. There are many applications for Flex, including writing compilers in conjunction with GNU Bison. Flex is a free implementation of the well known Lex program. It features a Lex compatibility mode, and also provides several new features such as exclusive start conditions.

Building the application

# used GNU Make 3.80

1. Uncompress the tar file,
        # cd /data/hugo
        # tar -zxvf flex-2.5.31.tar.gz

2. Build the package,
        # ./configure --prefix=/usr/local/flex-2.5.31_64bit
        # make
        # make install

3. Copy executable into /usr/local/bin
        # cp -rf /usr/local/flex-2.5.31_64bit/bin/flex /usr/local/bin/.

go Up

---

Fast Light Toolkit

• URL: http://www.fltk.org
• Current Version: v1.1.7
• Installed Version: v1.1.7
• Installed at: Aug. 23th, 2007 (hugo.hernandez@loni.ucla.edu)

FLTK (pronounced "fulltick") is a cross-platform C++ GUI toolkit for UNIX®/Linux® (X11), Microsoft® Windows®, and MacOS^?® X. FLTK provides modern GUI functionality without the bloat and supports 3D graphics via OpenGL^?® and its built-in GLUT emulation.

Building the application

1. uncompress tar file,
        # cd /data/hugo
        # tar -zxvf fltk-1.1.7.tar.gz
        # cd fltk-1.1.7

2. configure the package
        # mkdir /usr/local/fltk-1.1.7
        # ./configure --prefix=/usr/local/fltk-1.1.7 --libdir=/usr/local/lib

3. build the package,
        # make ; make install

4. copy libraries to the $prefix
        # cp /usr/local/lib/libfltk* /usr/local/fltk-1.1.7_64bit/lib

go Up

---

Free OpenGL Utility Toolkit

• URL: http://freeglut.sourceforge.net
• Current Version: 2.4.0
• Installed Version: 2.4.0
• Installed at: Aug. 31th, 2007 (hugo.hernandez@loni.ucla.edu)
• Package needed by NeuroLibs

freeglut is a completely OpenSourced alternative to the OpenGL Utility Toolkit (GLUT) library. GLUT was originally written by Mark Kilgard to support the sample programs in the second edition OpenGL 'RedBook'. Since then, GLUT has been used in a wide variety of practical applications because it is simple, widely available and highly portable. GLUT (and hence freeglut) allows the user to create and manage windows containing OpenGL contexts on a wide range of platforms and also read the mouse, keyboard and joystick functions.

Building the application

1. Download the code from http://freeglut.sourceforge.net/index.php#download,

2. uncompress the source code,
        # cd /tmp/hugo
        # tar -jxvf freeglut-2.4.0.tar.bz2

3. configure the package,
        # ./configure --prefix=/usr/local/freeglut-2.4.0_64bit

4. build the package,
        # make
        # make install

5. copy libraries into /usr/local/lib
        # cp -f /usr/local/freeglut-2.4.0_64bit/lib/* /usr/local/lib/.

go Up

---

FreeSurfer

• URL: http://surfer.nmr.mgh.harvard.edu
• Current Version: 4.0.0
• Installed Version: 3.0.5
• Installed at:

FreeSurfer is a set of automated tools for reconstruction of the brain’s cortical surface from structural MRI data, and overlay of functional MRI data onto the reconstructed surface.

Building the application

1. Uncompress the application,
        # cd /tmp/hugo
        # tar -C /usr/local -zxvf freesurfer-3.0.5.tar.gz

2. A license file is required to enable the tools. One can be obtained for free by registering at: https://surfer.nmr.mgh.harvard.edu/registration.html

go Up

---

FMRIB Software Library

• URL: http://www.fmrib.ox.ac.uk/fsl
• Current Version: 4.0
• Installed Version: 4.0
• Installed at: Sep. 10th, 2007 (hugo.hernandez@loni.ucla.edu)

FSL is a comprehensive library of analysis tools for FMRI, MRI and DTI brain imaging data, written mainly by members of the Analysis Group, FMRIB, Oxford, UK. FSL runs on Apple, PCs (Linux and Windows) and Sun, and is very easy to install.

Building the application

1. download code from http://www.fmrib.ox.ac.uk/fsl,

2. uncompress the tar file
        # tar -C /usr/local -zxvf /cxfs/tmp/sysadm/fsl-4.0.1-sources.tar.gz

3. to build the application
        # mv /usr/local/fsl /usr/local/fsl-4.0.1_64bit
        # cd /usr/local/fsl-4.0.1_64bit

4. set the main FSL environment variable
        # export FSLDIR=/usr/local/fsl-4.0.1_64bit

5. source the FSL environment setup script
        # . ${FSLDIR}/etc/fslconf/fsl.sh

6. check if your machine/compiler is supported by default
        # ls $FSLDIR/config/$FSLMACHTYPE

        ---> if not, select the closest match from the directories in $FSLDIR/config and do the following

        # cp -r $FSLDIR/config/linux_64-gcc4.1 $FSLDIR/config/$FSLMACHTYPE

7. build the application
        # ./build

8. if you want to re-make a particular package in $FSLDIR/src (e.g. flirt) then first set the FSLDEVDIR environment variable to the same as FSLDIR and then just type
        # make install

inside the relevant directory.

go Up

---

GNU Compiler Collection

• URL: http://gcc.gnu.org
• Current Version: 4.2.1
• Installed Version: 4.2.0
• Installed at: Sept. 5th, 2007 (hugo.hernandez@loni.ucla.edu)
• Package needed by Matlab

The GNU Compiler Collection includes front ends for C, C++, Objective-C, Fortran, Java, and Ada, as well as libraries for these languages (libstdc++, libgcj,...).

Building the application

1. download code from http://gcc.gnu.org/mirrors.html,

2. uncompress the tar file
        # tar -C /tmp/hugo -jxvf /cxfs/tmp/sysadm/gcc-4.2.0.tar.bz2

3. build the application
        # cd /tmp/hugo/gcc-4.2.0
        # ./configure --prefix=/usr/local/gcc-4.2.0_64bit

        ---> be sure you have installed glibc-devel for both 32 and 64bits
                currently: glibc-devel-2.3.4-2.36

        # make

        ---> create backups for the executables of the current gcc version in /usr/bin

        # make install

4. create symlinks or copy new execs into /usr/bin

go Up

---

Insight Applications

• URL: http://www.itk.org/HTML/Download.php
• Current Version: 3.2.0
• Installed Version: 3.2.0
• Installed at: Aug. 23th, 2007 (hugo.hernandez@loni.ucla.edu)

The following applications illustrate the use of ITK in real-world medical imaging applications. Note these application are found in the InsightApplications module. They differ from the Insight/Examples examples in that they use other systems such as VTK, FLTK and Qt to create turn-key applications. Please make sure that you can compile and link the examples found in Insight/Examples prior to building these applications. The applications are often difficult to compile and run.

Building the application

1. Uncompress the tar file,
        # cd /tmp/hugo
        # tar -zxvf _InsightApplications_-3.2.0.tar.gz

2. Create build directory,
        # mkdir /usr/local/_InsightApplications_-3.2.0_64bit

3. Configure the package,
        # ccmake .

        Use flags:
                BUILD_SHARED_LIBRARIES ON
                ITK_DIR /tmp/hugo/InsightToolKit-3.2.0 (source code)
                CMAKE_INSTALL_PREFIX /usr/local/InsightApplications-3.2.0_64bit

        Hit [c] to configure,
        Hit [g] to generate and exit,

4. Build the package,
        # make
        # make install

5. Copied libraries into /usr/local/lib
        # cp -vf /usr/local/InsightApplications/lib/* /usr/local/lib/.

go Up

---

Insight Toolkit

• URL: http://www.itk.org/HTML/Download.php
• Current Version: 3.2.0
• Installed Version: 3.2.0
• Installed at: Aug. 22th, 2007 (hugo.hernandez@loni.ucla.edu)

ITK is an open-source software system to support the Visible Human Project. Currently under active development, ITK employs leading-edge segmentation and registration algorithms in two, three, and more dimensions.

Building the application

# used ccmake version 2.4-patch 6 (2.4-patch 5 used on cerebellum)
# used GNU Make 3.80

1. Uncompress the tar file,
        # cd /data/hugo
        # tar -zxvf _InsightToolkit_-3.2.0.tar.gz

2. Create build directory,
        # mkdir /usr/local/InsightToolkit-3.2.0

3. Configure the package,
        # ccmake /data/hugo/InsightToolkit-3.2.0

        Use flags:
                BUILD_SHARED_LIBRARIES ON
                BUILD_EXAMPLES 0N
                BUILD_TESTING ON
                CMAKE_INSTALL_PREFIX /usr/local/InsightToolkit-3.2.0

        Hit [c] to configure,
        Hit [g] to generate and exit,

4. Build the package,
        # make
        # make install

5. Copied execs from build dir to $prefix dir
        # cp /data/hugo/InsightToolkit-3.2.0/bin/* /usr/local/InsightToolkit-3.2.0/bin/.

6. Copied library dir into /usr/local/lib
        # cp -rvf /usr/local/InsightToolkit/lib/InsightToolkit /usr/local/lib/.

go Up

---

Java SE Development Kit

• URL: http://java.sun.com/javase/downloads/index.jsp
• Current Version: 6u2
• Installed Version: 6u2
• Installed at: Jul. 27th, 2007 (rico.magsipoc@loni.ucla.edu)

The Java SE Development Kit (JDK) includes the Java Runtime Environment (JRE) and command-line development tools that are useful for developing applets and applications.

Building the application

1. download the application from http://java.sun.com/javase/downloads/index.jsp,

2. install the application,
        # cd /tmp/hugo
        # chmod +x jdk-6u2-linux-amd64-rpm.bin
        # ./jdk-6u2-linux-i586-rpm.bin

        ---> follow the instructions

3. copy /usr/java into /usr/local

go Up

---

LONI Applications

• URL: Unknown
• Current Version: Unknown
• Installed Version: Unknown
• Installed at: Unknown

go Up

---

LONI Data

• URL: Unknown
• Current Version: Unknown
• Installed Version: Unknown
• Installed at: Unknown

go Up

---

LONI Jar Files

• URL: Unknown
• Current Version: Unknown
• Installed Version: Unknown
• Installed at: Unknown

go Up

---

LONI Scripts

• URL: Unknown
• Current Version: Unknown
• Installed Version: Unknown
• Installed at: Unknown

go Up

---

Matlab

• URL: http://www.mathworks.com/products/matlab
• Current Version: 7.5 (2007b)
• Installed Version: 7.4.0.287 (2007a)
• Installed at: Sept. 5th, 2007 (hugo.hernandez@loni.ucla.edu)

MATLAB is a high-level language and interactive environment that enables you to perform computationally intensive tasks faster than with traditional programming languages such as C, C++, and Fortran.

Building the application

1. download code from http://www.mathworks.com,

2. uncompress the tar file
        # cd /hugo/tmp/build-matlab
        # tar -xvf boot.ftp

3. build the application
        # ./install

        ---> follow all the instructions from the GUI

4. use SPM5
        ---> copy all the content of the SPM5 directory into the Matlab directory
        # cp -rf /usr/local/spm5 /usr/local/matlab7.4_64bit/.

        ---> recompile the mex files with the current gcc compiler
        # cd /usr/local/matlab7.4_64bit/spm5/src
        # make & make install

5. Include SPM5 into the Matlab path
        setenv MATLABPATH /usr/local/matlab7.4_64bit/spm5

go Up

---

Montreal Neurological Institute

• URL: http://mni.mcgill.ca/aboutmni.html
• Current Version: Unknown
• Installed Version: 03.15.2006
• Installed at: Feb., 2007

The MNI is a teaching and research institute of McGill University in which multidisciplinary teams of basic and clinician scientists work to generate fundamental information about the nervous system and to apply that knowledge to understanding and treating neurological diseases.

Building the application

System:
        Linux x86_64

Compiler:
        3.4.5

Procedure:
1 netcdf-3.5.0
        o setenv CPPFLAGS "-Df2cFortran"
        o setenv FC /usr/bin/g77
        o setenv FFLAGS "-Wno-globals"
        o setenv CXX /usr/bin/g++
        o ./configure --prefix=/usr/local/mni
        o make ; make install

2. minc-1.4.0
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

3. mni_autogen-0.98v
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

4. ebtks-1.5.0
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

5. N3-1.10
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

6. bicpl-1.4.2
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

7. conglomerate-1.5
        o ./configure --prefix=/usr/local/mni --with-build-path=/usr/local/mni
        o make ; make install

8. mni_perllib-0.07
        o setenv PERLLIB /usr/local/mni/lib
        o perl Makefile.PL PREFIX=/usr/local/mni INSTALLDIRS=perl
        o make ; make install

go Up

---

netpbm

• URL: http://netpbm.sourceforge.net
• Current Version: 10.26.44
• Installed Version: 10.26.42
• Installed at: Aug. 29th, 2007 (hugo.hernandez@loni.ucla.edu)

Netpbm is a toolkit for manipulation of graphic images, including conversion of images between a variety of different formats. There are over 300 separate tools in the package including converters for about 100 graphics formats. Examples of the sort of image manipulation we're talking about are: Shrinking an image by 10%; Cutting the top half off of an image; Making a mirror image; Creating a sequence of images that fade from one image to another.

Building the application

# used GNU Make 3.80

# requires:
        o flex/lex (generate programs for lexical tasks)
        o tiff libraries

1. Uncompress the tar file,
        # cd /data/hugo
        # tar -zxvf netpbm-10.26.42.tgz

2. Build the package. See doc/INSTALL.
        # ./configure
                o use static library
                o define tiff libraries and paths when required
        # make
        # make package pkgdir=/usr/local/netpbm-10.26.42_64bit

3. To gather all the installable parts into a specified directory, and finally
        # installnetpbm

go Up

---

nifticlib

• URL: http://sourceforge.net/projects/niftilib
• Current Version: 0.6
• Installed Version: 0.3
• Installed at: Aug. 28th, 2007 (hugo.hernandez@loni.ucla.edu)

niftilib is a collection of i/o routines for the nifti1 neuroimage data format. C (nifticlib), Java (niftijlib), Matlab (niftimatlib), and Python (pynifti) code is available. Information about the nifti format is available at http://nifti.nimh.nih.gov/df

Building the application

# used GNU Make 3.80

1. Uncompress the tar file,
        # cd /data/hugo
        # tar -C /usr/local -zxvf nifticlib-0.3.tar.gz
        # cd /usr/local ; mv nifticlib-0.3 nifticlib-0.3_64bit

2. Build the package,
        # make all

3. Copy libraries into /usr/local/lib
        # cp -rf /usr/local/nifticlib-0.3_64bit/lib /usr/local/lib/.

go Up

---

R Project for Statistical Computing

• URL: http://www.r-project.org
• Current Version: 2.5.1
• Installed Version: 2.4.1
• Installed at: Aug. 21th, 2007 (hugo.hernandez@loni.ucla.edu)

R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS.

Building the application

# based on R. Magsipoc's compilation on cerebellum (3/6/07)
#
# Build latest version of readline, 64-bit, first with default flags

1. Set flags,
        o setenv CFLAGS "-O2 -m64"
        o setenv CXXFLAGS "-O2 -m64"
        o setenv CPPFLAGS "-I/usr/local/include -I/usr/include"
        o setenv LDFLAGS "-L/usr/local/lib -L/usr/lib64"

2. Uncompress package,
        # cd /tmp/hugo ; tar -zxvf R-2.4.1.tar.gz
        # cd /tmp/hugo/R-2.4.1

3. Configure and build package,
        # ./configure --prefix=/usr/local/R-2.4.1_64bit
        # make
        # make check
        # make install

go Up

---

readline

• URL: http://tiswww.case.edu/php/chet/readline/rltop.html
• Current Version: 5.2
• Installed Version: 5.2
• Installed at: Aug. 21th, 2007 (hugo.hernandez@loni.ucla.edu)

The GNU Readline library provides a set of functions for use by applications that allow users to edit command lines as they are typed in. Both Emacs and vi editing modes are available. The Readline library includes additional functions to maintain a list of previously-entered command lines, to recall and perhaps reedit those lines, and perform csh-like history expansion on previous commands.

Building the application

1. Set flags,
        o setenv CFLAGS "-O2 -m64"
        o setenv CXXFLAGS "-O2 -m64"
        o setenv CPPFLAGS "-I/usr/local/include -I/usr/include"
        o setenv LDFLAGS "-L/usr/local/lib -L/usr/lib64"

2. Uncompress package,
        # cd /tmp/hugo ; tar -zxvf readline-5.2.tar.gz
        # cd /tmp/hugo/readline-5.2

3. Configure and build package,
        # ./configure --prefix=/usr/local/readline-5.2_64bit
        # make
        # make install

go Up

---

Shape Tools

• URL: unknown
• Current Version: 1.3.2
• Installed Version: 1.3.2
• Installed at: Aug. 24th, 2007 (hugo.hernandez@loni.ucla.edu)

Building the application

Copied directory from Craig Schwartz.

go Up

---

Statistical Parametric Mapping

• URL: http://www.fil.ion.ucl.ac.uk/spm/software/spm5
• Current Version: 5
• Installed Version: 5
• Installed at: Sept. 5th, 2007 (hugo.hernandez@loni.ucla.edu)

Statistical Parametric Mapping refers to the construction and assessment of spatially extended statistical processes used to test hypotheses about functional imaging data. These ideas have been instantiated in software that is called SPM. The SPM is a MATLAB software package and has been designed for the analysis of brain imaging data sequences. The sequences can be a series of images from different cohorts, or time-series from the same subject. The current release is designed for the analysis of fMRI, PET, SPECT, EEG and MEG.

Building the application

1. download code from http://www.fil.ion.ucl.ac.uk/spm/software/download.html,

2. uncompress the tar file
        # tar -C /usr/local -zxvf /cxfs/tmp/sysadm/MatLab/spm5.tar.gz

3. recompile the mex files to work with the current GCC version
        # cd /usr/local/spm5/src
        # make && make install

4. To enable this package for Matlab, copy the complete directory into the Matlab directory and include it into the Matlab Path
        # cp -rf /usr/local/spm5 /usr/local/matlab/.

go Up

---

Tagged Image File Format

• URL: http://www.remotesensing.org/libtiff
• Current Version: 3.8.2
• Installed Version: 3.8.2
• Installed at: Aug. 29th, 2007 (hugo.hernandez@loni.ucla.edu)

This software provides support for the Tag Image File Format (TIFF), a widely used format for storing image data. Included in this software distribution is a library, libtiff, for reading and writing TIFF, a small collection of tools for doing simple manipulations of TIFF images, and documentation on the library and tools. Libtiff is a portable software, it was built and tested on various systems: UNIX flavors (Linux, BSD, Solaris, MacOS X), Windows, OpenVMS. It should be possible to port libtiff and additional tools on other OSes.

Building the application

# used GNU Make 3.80

1. Uncompress the tar file,
        # cd /data/hugo
        # tar -zxvf tiff-3.8.2.tar.gz

2. Build the package,
        # ./configure --prefix=/usr/local/tiff-3.8.2_64bit
        # make
        # make install

3. Copy libraries into /usr/local/lib
        # cp -rf /usr/local/tiff-3.8.2_64bit/lib/* /usr/local/lib/.

go Up

---

Visualization Toolkit

• URL: http://www.vtk.org
• Current Version: 5.0.3
• Installed Version: 5.0.2
• Installed at: Aug. 23th, 2007 (hugo.hernandez@loni.ucla.edu)

The Visualization Toolkit (VTK) is an open source, freely available software system for 3D computer graphics, image processing, and visualization used by thousands of researchers and developers around the world. VTK consists of a C++ class library, and several interpreted interface layers including Tcl/Tk, Java, and Python.

Building the application

1. uncompress the tar file,
        # cd /tmp/hugo
        # tar -zxvf vtk-5.0.2.tar.gz
        # cd VTK

2. configure package,
        # ccmake .

        Use the flags:
                BUILD_SHARED_LIBS ON
                CMAKE_INSTALL_PREFIX /usr/local/vtk-5.0.2
                VTK_DATA_ROOT /tmp/hugo/VTK
                VTK_USE_RENDERING ON
                VTK_USE_RPATH OFF
                VTK_WRAP_TCL ON

3. build package,
        # mkdir /usr/local/vtk-5.0.2
        # make ; make install

go Up

---

No permission to read topic Infrastructure.CR_BasicConfig - perhaps you need to log in?

No permission to read topic Infrastructure.CR_DNS - perhaps you need to log in?

Disabled Cranium Production Nodes (Hardware and Testing)

This page is meant to serve as formal notification of disabled nodes, which we do on a near-daily basis. The page will allow us to keep track of reasons for node disablings, while also allowing us to catch nodes that should be returned into production much more easily.

Please file the node under the proper rack category, give a brief justification, the date of the disable, and the responsible SysAdmin^? (to be contacted if the node is believed to remain disabled by oversight).

Nodes with hardware issues are denoted in RED.

R1 nodes	Reason	Next Action	Date	Sysadmin
cerebro-1-[101-125]	Moved to pipeline-ext.q	HOLD
cerebro-1-109	Machine reboots before Boot Agent when instructed to PXE boot	HOLD	04.22.09	jpierce

R2 nodes	Reason	Next Action	Date	Sysadmin

R3 nodes	Reason	Next Action	Date	Sysadmin
cerebro-3-134	Scrapped for parts	permanently offline	12.15.08	jpierce
cerebro-3-135	Scrapped for parts	permanently offline	12.15.08	jpierce

R4 nodes	Reason	Next Action	Date	Sysadmin
cerebro-4-114	Issue: No life over serial console, SP, or VGA on boot.	Good parts taken: 2x RAM	05.05.09	jpierce
cerebro-4-138	DHCP requests not seen by frontend	HOLD	04.23.09	jpierce

R5 nodes	Reason	Next Action	Date	Sysadmin

R6 nodes	Reason	Next Action	Date	Sysadmin

R7 nodes	Reason	Next Action	Date	Sysadmin

R8 nodes	Reason	Next Action	Date	Sysadmin
cerebro-8-114	Bad DIMM reported in slot 1, stops boot-up	replace with memory from another machine	04.23.09	jpierce
cerebro-8-137	Being used as pipeline-lsh1	HOLD	12.15.08	hdezmora

R23 nodes	Reason	Next Action	Date	Sysadmin
cerebro-23-101	Being used as cerebro-lmh3	Pending 6.2 going live	12.04.08	hdezmora
cerebro-23-102	Being used as cerebro-lmh4	Pending 6.2 going live	12.04.08	hdezmora
cerebro-23-103	Being used as cerebro-lsh5	Pending 6.2 going live	12.04.08	hdezmora
cerebro-23-104	Being used as development exec node	HOLD	02.05.09	jpierce
cerebro-23-105	Being used as pipeline-lmh1	HOLD	12.15.08	hdezmora
cerebro-23-115	Testing Panasas storage performance	HOLD	04.07.09	jpierce
cerebro-23-116	Testing Panasas storage performance	HOLD	04.07.09	jpierce
cerebro-23-129	Test install /usr/sge -> /ifs/sge-dev	HOLD	02.18.09	jpierce

R24 nodes	Reason	Next Action	Date	Sysadmin
cerebro-24-111	Temporarily repurposed to upgrade mgmt servers to CentOS? 5.2	Configure OS	04.28.09	jpierce
cerebro-24-112	Temporarily repurposed to upgrade mgmt servers to CentOS? 5.2	Install OS	04.28.09	jpierce
cerebro-24-136	Repurposed as hsqldb	HOLD	02.18.09	jpierce
cerebro-24-137	Repurposed as Rocks 5 frontend	HOLD	02.19.09	jpierce

No permission to read topic Infrastructure.CR_ExtendCompute - perhaps you need to log in?

No permission to read topic Infrastructure.CR_GrubScheme - perhaps you need to log in?

Cerebro Rocks: Pre-Node Frontend Configuration

Use these links (in order) to complete the basic frontend configuration:

• Basic Configuration
• 411 Configuration
• SGE Configuration
• Node Kickstart Configuration

No permission to read topic Infrastructure.CR_NIS - perhaps you need to log in?

No permission to read topic Infrastructure.CR_NodeInstall - perhaps you need to log in?

No permission to read topic Infrastructure.CR_PartitionScheme - perhaps you need to log in?

No permission to read topic Infrastructure.CR_PostConfig - perhaps you need to log in?

No permission to read topic Infrastructure.CR_SGEConfig - perhaps you need to log in?

No permission to read topic Infrastructure.CR_Troubleshoot - perhaps you need to log in?

No permission to read topic Infrastructure.CR_qmasterFailover - perhaps you need to log in?

No permission to read topic CatOS4Levitt - perhaps you need to log in?

No permission to read topic CerebroRocks - perhaps you need to log in?

No permission to read topic CerebroRocksCommands - perhaps you need to log in?

No permission to read topic ChangeLogs - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_general - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_loni-core-1 - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_loni-core-2 - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_loni-fwsm - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_loni-sw-1 - perhaps you need to log in?

loni-sw-2: changelog

Overview - loni-sw-2 is a Cisco 2890, located in Reed Datacenter, Rack 36. It provides network access to the Reed Datacenter, and is connected to loni-core-1 via GigE Fiber LX <-> Reed DC Fiber Panel <-> NRB 225S Rack 26 Fiber Panel <-> NRB Datacenter, Rack 13 Fiber Panel <-> loni-core-1.

Dates are in the format YY_MMDD

08_0313 - dhasson

• Start of changelog

-- DavidHasson - 12 Mar 2008

loni-sw-3: changelog

Overview - loni-sw-3 is a Cisco 2970-24TS located in the "Production 3" rack, or rack 19. It's connected via dual GigE^? Copper <-> loni-core-1. It's a production switch, which means it's trunked and is intended to support all vlans except management 280.

Dates are in the format YY_MMDD, entries go latest to oldest

08_0710 - dhasson

• Zoned gi0/17 -> dessi-local, vlan 210
• renamed to "nrb-r19-tor1"

08_0313 - dhasson

• Start of changelog

-- DavidHasson - 13 Mar 2008

No permission to read topic Infrastructure.Changelog_loni-sw-4 - perhaps you need to log in?

loni-sw-5: changelog

Overview - loni-sw-5 is a SMC 10/100/1000 located in the "Production 3" rack, or rack 19. It's connected via single GigE^? SX Fiber <-> loni-core-1. It's a production switch, which means it's trunked and is intended to support all vlans except management 280.

Dates are in the format YY_MMDD, entries go latest to oldest

08_0313 - dhasson

• Start of changelog

-- DavidHasson - 13 Mar 2008

loni-sw-6: changelog

Overview - loni-sw-6 is a Linksys SRW2024 located in Reed C Floor IDF. It's connected via GigE^? Fiber SX <-> loni-sw-7 (Reed DC). It's a distribution switch, which means it's trunked and is intended to support the 210 and 104 vlans for windows and linux clients.

Dates are in the format YY_MMDD, entries go latest to oldest

08_0313 - dhasson

• Start of changelog

-- DavidHasson - 13 Mar 2008

loni-sw-7: changelog

Overview - loni-sw-7 is a Linksys SRW2016 located in Reed DC, Rack 36. It's connected via GigE to loni-sw-2 in the same rack. It is currently only used as a fiber transciever to interconnect loni-sw-6 downstairs with the LONI network.

Dates are in the format YY_MMDD, entries go latest to oldest

08_0313 - dhasson

• Start of changelog

-- DavidHasson - 13 Mar 2008

RicoMagsipoc^? - 23 Jul 2008

Overview - loni-edge is comprised of 2 Cisco Cat3750 in failover mode. These are located in the NRB1 DC, Rack 13.

Dates are in the format YY_MMDD, entries go latest to oldest

08_0723 - rico

• shutdown port 1/0/2. Originally configured for BIRN VPN.
  • interface GigabitEthernet1^?/0/2
  • description loni-edge <-> birn-vpn UPLINK
  • switchport access vlan 200

-- RicoMagsipoc^? - 23 Jul 2008

No permission to read topic Infrastructure.Changelog_nessus - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_viola-rs - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_viola-rs1 - perhaps you need to log in?

No permission to read topic Infrastructure.Changelog_viola-rs2 - perhaps you need to log in?

Everyday Systems Administration Tasks

Modifying DNS - Adding & Changing DNS Records for loni.ucla.edu using nsupdate

Networker Client Setup - How to setup a new client (e.g., server or workstation) to be backed up by networker

-- DavidHasson - 14 Oct 2007

No permission to read topic ComputerCenterSchema - perhaps you need to log in?

No permission to read topic ControllingCerebroCluster - perhaps you need to log in?

No permission to read topic CvsRepository - perhaps you need to log in?

No permission to read topic DHCPSetUp - perhaps you need to log in?

FreeBSD^? NFS: allow all for portmapper http://www.freebsddiary.org/nfs-portmap.php

No permission to read topic DelUser - perhaps you need to log in?

No permission to read topic DellIPMIMangaement - perhaps you need to log in?

No permission to read topic DellKVM - perhaps you need to log in?

No permission to read topic DellRacManagment - perhaps you need to log in?

No permission to read topic DellTricks - perhaps you need to log in?

No permission to read topic Infrastructure.Drac3SOL - perhaps you need to log in?

No permission to read topic EditingComputersDB - perhaps you need to log in?

Configuring your Email client for IMAP

To access your access your email via IMAP you must configure your email client to use SSL.

When configuring your server settings set your Server Name to be:

webmail.loni.ucla.edu

Set your User Name to be your LONI logon name, mine would be:

jtrout

Finally make sure that the secure connection has SSL checked and the default port is set to 993.

If you need a step by step walk through check the links below.

Outlook

http://www.bol.ucla.edu/help/howto/apps/outlook/outlookconfig.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Thunderbird

http://www.bol.ucla.edu/software/mac/thunderbird/configure.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Mail (Mac)

http://www.bol.ucla.edu/help/howto/apps/mailapp/mailconfig.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Entourage

http://www.bol.ucla.edu/help/howto/apps/entourage/entxconfig.html

http://www.bol.ucla.edu/help/howto/apps/entourage/ent2004config.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Mozilla

http://www.bol.ucla.edu/software/mac/mozilla/config.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Eudora

http://www.bol.ucla.edu/software/mac/eudoraosx/config.html

note: this walk through is for UCLA make sure to change their server names to LONI server names.

Configuring your Email client for POP

To access your access your email via IMAP you must configure your email client to use SSL.

When configuring your server settings set your Server Name to be:

webmail.loni.ucla.edu

Set your User Name to be your LONI logon name, mine would be:

jtrout

Finally make sure that the secure connection has SSL checked and the default port is set to 995.

If you need a step-by-step walkthrough check the links above. The setup is the same for POP as it is for IMAP. Remember to check the UCLA Server Name mail.ucla.edu to LONI’s Server Name webmail.loni.ucla.edu

Configuring your Email client for SMTP

SMTP also requires secure authentication; however, the method differs depending on the email client. If you are using outlook, outlook express or Mac's mail SSL will work.

*note if you are using Mac’s Mail program set the SSL authentication to password, do not leave it on the default which is Kerberos 5.

At this point the only other email client that has been tested is Thunderbird. Thunderbird's SSL does not work with Exchange so under Authentication use TLS not SSL. Make sure that the Server Name is webmail.loni.ucla.edu and your User Name is your LONI account name and everything should work fine.

*note Even though we have not tested other email clients they should work. However, you might have to play around with the configuration to make it work. Try SSL first and if that fails use TLS. Remember SMTP requires your username and password, so make sure not to leave those fields blank or you will not be able to send out emails. For help check out the step-by-step walkthroughs above.

No permission to read topic ExchangeCluster - perhaps you need to log in?

No permission to read topic FwsmSop - perhaps you need to log in?

Accessing the LONI Network Remotely

SSH

SSH is short for Secure Shell. It provides a secure, encrypted method of transmitting information over TCP/IP to ensure the confidentiality, integrity and authenticity of transferred data. This is the preferred method of connectivity to LONI's Unix and Linux computational resources.

To access the LONI network via SSH:

1. Obtain an SSH client for Windows, Linux, or MacOSX
   • To find out if you already have ssh on a Unix/Linux based machine, type which ssh
2. SSH to ssh.loni.ucla.edu over port 22 using the your LONI username and password.
3. If accessing LONI for the first time, please change your password by following the instructions here.

About the SSH Servers

How to access LONI's SSL VPN

A virtual private network (VPN) is an encrypted connection established between an external user and LONI's network. Once a user successfully establishes a VPN tunnel, they are given full access to internal resources, including network shares, printers and secure compute servers. We have a new, robust SSL VPN to address firewalls that prohibit IPSEC/PPTP traffic. There is minimal setup required (client webapp installation only). This service is accessible here. See AccessingtheVPN.pdf below for a comprehensive walk-through.

How to set up a PPTP VPN in Windows

Deprecated. This is meant to be a secondary means of establishing a VPN connection, only if you cannot use the SSL VPN.

32-bit, 64-bit Windows XP

You must have a working internet connection for this procedure to work. Under Network Connections in the Control Panel:

1. Select "Create a New Connection."
2. Select "Connect to the network at my workplace."
3. Select "Virtual Private Network connection."
4. Choose an appropriate Company Name. (i.e. LONI VPN)>
5. Select "Do not dial an initial connection."
6. Use hostname vpn2.loni.ucla.edu
7. Select appropriate level of usability.
8. Select "Finish."

Use your LONI Windows username and password.

32-bit, 64-bit Windows Vista, Windows 7 or Windows 2008

You must have a working internet connection for this procedure to work. Under the Start Menu:

1. Select "Connect To"
2. Select "Set up a Connection or Network"
3. Select "Connect to a workplace"
4. Select "Use my Internet connection (VPN)"
5. Use Internet address vpn2.loni.ucla.edu
6. Choose an appropriate Destination Name. (i.e. LONI PPTP VPN)
7. Select appropriate level of usability.
8. Use your LONI username and password and the domain LONI
9. Select "Connect"

How to set up a PPTP VPN in Mac OSX

Deprecated. This is meant to be a secondary means of establishing a VPN connection, only if you cannot use the SSL VPN.

1. Open Internet Connect in the /Applications folder.
   
2. In Internet Connect, select New VPN (PPTP) Connection Window from the File menu.
   
3. Enter the PPTP server address vpn2.loni.ucla.edu and the appropriate credentials.
   
4. Once you have everything filled out, click the Connect button. Watch the Status field to be sure you are establishing a connection.
   
5. After the connection is up, you can check in the Network pane of System Preferences for a PPTP device.
   

Use your LONI Windows username and password.

Wireless and VPN on the UCLA campus

To access the network wirelessly while on the UCLA campus:

1. Connect to UCLAWLAN.
   
2. Login to LONI VPN using your login credentials.
   You do not need to authenticate via the BOL Access Page or the BOL VPN to access the LONI Network

-- RicoMagsipoc - 29 Jan 2009

LONI Grid Computing Notes

To facilitate the submission and execution of compute jobs in this heterogeneous compute environment, SUN's Grid Engine (SGE) is used to virtualize the resources above into a compute service. A grid layer sits atop the compute resources and submits jobs to available resources according to user-defined criteria such as CPU type, processor count, etc.

---

Cranium Grid Technical Details

The Cranium cluster has a total of 1152 available processors, with a total of 1152 defined slots to run compute jobs. The cluster is comprised of two types of compute nodes, 296 Sun Microsystems V20z servers with dual 64-bit 2.4GHz AMD Opteron processors and 8Gb of memory, and 80 Sun Microsystems X2200 M1 servers with a dual 64-bit Quad-Core 2.2GHz AMD Opteron processor and 16GB of memory. The cluster is configured using Rocks Cluster 4.5 and Sun Grid Engine 6.1u4 to schedule, distribute and administer the submitted jobs. All compute nodes and servers are running CentOS release 4.5 as the operating system. The cluster is structured as follows:

• 2 SGE Master nodes in fault-tolerant shadowed configuration (not accessible to the general public)
• 1 SGE Submit node
  • qsub.loni.ucla.edu (gui.loni.ucla.edu)
• 1 Pipeline Server node
  • cranium.loni.ucla.edu
• 1 Compiler node
  • cerebro-rcc.loni.ucla.edu
• SGE Execution nodes
  • cerebro-1/8-101 through cerebro-1/8-137 (V20z servers)
  • cerebro-23/24-101 through cerebro-23/24-137 (X2200 M1 servers)
• Mounted NFS filesystems
  • All /ifs directories (e.g. /ifs/tmp, /ifs/four_d, etc.)
  • /ifshome for user home directories

Submitting to Cranium Grid Engine

Summary: Compile code for GNU/Linux, kernel 2.6.9-55.ELsmp. Put executable and data files in /ifs directory. Run qsub on script containing job arguments.

The Sun Grid Engine (SGE) is a job management service intended to accept, schedule, dispatch, and manage remote execution of a large number of standalone, parallel or interactive user jobs. The current version for the Cerebro cluster is 6.1u4. Documentation can be found at http://gridengine.sunsource.net.

• Place all code and data files you wish to use into the appropriate /ifs NFS directory. By default, you can use /ifs/tmp. (NOTE: files older than 28 days are deleted off /ifs/tmp nightly)
• Compile code on cerebro-rcc.loni.ucla.edu using GCC, PGI or Intel compiler
  • GCC can be found in /usr/local/bin/gcc or /usr/bin/gcc4,
  • Compiler tools for GCC v4.2 can be found in /usr/local/gcc-4.2.0_64bit
  • PGI compilers can be found in /usr/local/pgi
  • Intel compilers can be found in /usr/local/intel*
• Login to qsub.loni.ucla.edu via SSH. Note that an external user must
  a) Log into an SSH proxy such as ssh.loni.ucla.edu first, prior to establishing an SSH connection to a submit node or
  b) VPN into LONI network. Instructions on how to establish a VPN connection can be found in http://www.loni.ucla.edu/twiki/bin/view/Infrastructure/GettingOnline
  
• Source the configuration script
  • 'source /usr/sge/loni/common/settings.sh' if your shell is sh/bash
  • 'source /usr/sge/loni/common/settings.csh' if your shell is tcsh/csh
• Wrap the executable (line 4) into a shell script containing Grid interpreter primitives (lines 2, 3, and 4)
  • Example shell script:
    1. ) #!/bin/sh
       
    2. ) #$ -S /bin/sh
       
    3. ) #$ -j y -o /tmp
       
    4. ) #$ -l h_vmem=3G
    5. ) /ifs/tmp/myexecutable /ifs/tmp/inputfile /ifs/tmp/outputfile
       
       
       
• If your job requires more than 2G of memory (up to 8G), specify the amount using "-l h_vmem=[memory]" -- units of K(ilobytes), M(egabytes) or G(igabytes) as in line 3 of the above script. If your job uses 2G or less, you need not specify the h_vmem resource. Jobs exceeding their expected (or default, if unspecified) memory usage, will be terminated immediately. This ensures that we track memory usage of every job carefully, mitigating the likelihood that one process takes down a compute node running multiple jobs from multiple users.

  • Queues: containers for different categories of jobs. Queues provide the corresponding resources for concurrent execution of multiple jobs that belong to the same category. A single queue can group a large set of hosts and a particular host can belong to different queues. The following queues are defined for the Cranium cluster:

    1) short.q: There is a 2-CPU-hour hard limit set for this queue. If the job exceeds this time limit, it will be automatically terminated. Any one user can only user 75% of all the slots (CPUs) in this queue. There are approximately 160 slots alloted to short.q.
    
         Submitting jobs:
         prompt> qsub -q short.q {name of the script}
    
    

    2) medium.q: There is a 12-CPU-hour hard limit set for the medium.q. As with the short.q above, if the job exceeds the time limit, it will be automatically terminated. Any one user can only user 75% of its slots (CPUs). There are approximately 104 slots alloted to the queue.
    
         Submitting jobs:
         prompt> qsub -q medium.q {name of the script}
    
    

    3) long.q: There is NO CPU time limit set for the long.q. Subsequently, it has the least number of slots, with 64 CPUs available. Any one user can only user 75% of all the CPUs in this queue.
    
         Submitting jobs:
         prompt> qsub -q long.q {name of the script}
    
    

    4) pipeline.q: This queue is used by the Pipeline server ONLY. There is NO CPU time limit set for this queue.
    
         Submitting jobs:
         Using the Pipeline Execution Environment application found here.
    
    

The Cranium cluster does not oversubscribe its CPUs; there is a 1-to-1 relationship between a compute job and a CPU. Experience showed that CPU oversubscription, where a single CPU is shared by multiple jobs, caused unpredictability and instability. By default, each CPU (thus, each process) gets 2GB of memory.

Useful Grid Engine commands:

• 'qsub' - Allows grid users to submit job to the cluster.
  
  Options:
  • -cwd: Run from current working directory,
  • -o {output file}: where to send the standard output,
  • -e {error file}: where to send the standard error,
  • -N (job name}: Job name for reporting,
  • -j y: To merge the output and error messages into a specific file
  • -p {value}: Priority level of the submitted job(s).
  • -l h_vmem={value}: Memory limit of the job (if unspecified, {value}=2G by default); units of K(ilobytes), M(egabytes) or G(igabytes)
• 'qstat' - Allows grid users to see the status of their submitted job.
  
  Options:
  • -f: Display full information of the job,
  • -u {username}: Show only username's jobs.
  • -j {job-ID} -explain E: Why is my job in Eqw state?
• 'qdel' - Allows grid users to delete jobs from the queue. Options:
  • -u {username}: Allows grid users to delete all their jobs from the queue.
  • -f {job-ID}: Allows grid users to delete a job from the queue by specifying the SGE job-ID.
• 'qacct -o {username}' - Allows grid users to extract more specific, runtime status of jobs including error messages
• 'qalter' - Allows grid users to modify a pending batch job,
• 'qhold' - Allows grid users to hold back a pending submitted job,
• 'qhost' - Allows grid users to get information about execution hosts,
• 'qconf' - Allows grid users to get information about the cluster and the queues.

Jobs Status:

• 'qw' - Queued and waiting,
• 'w' - Job waiting,
• 's' - Job suspended,
• 't' - Job transferring and about to start,
• 'r' - Job running,
• 'h' - Job hold,
• 'R' - Job restarted,
• 'd' - Job has been marked for deletion,
• 'Eqw' - An error occurred with the job.

The state E(rror) appears for pending jobs that couldn't be started due to job properties. Why is my job in a 'Eqw' state? Use the commands:

        qstat -j job_ID -explain E

        or

        qacct -j job_ID

Check My Job Status:

• qstat -s prs -u $USER --- check jobs that are pending, running, or suspended
• qstat -t -u $USER --- display the nodes where the job is running
• qstat -ext -s p -u $USER --- display extended information for my pending jobs
• qstat -s h --- jobs in hold status
• qstat -s r --- jobs in running status
• qstat -s r -u $USER --- jobs that are mine and running
• qstat -s s -u $USER --- jobs that are mine and suspended

Host State:

• 'au' - Host is in alarm and unreachable,

• 'u' - Host is unreachable. Usually SGE is down or the machine is down. Check this out.

• 'a' - Host is in alarm. It is normal on if the state of the node is full, it means, if on the node is using most of its resources.

• 'aS' - Host is in alarm and Suspended. If the node is using most of its resources, SGE suspends this node to take any other job unless resources are available.

• 'd' - Host is disabled,

• 'E' - ERROR. This requires the command `qmod -c` to clear the error state.

Common reasons for a job to fail

• Submission script not formatted properly,

• SGE cannot find the binary file specified in the job script,

• Required input files are missing from the startup directory,

• Environment variables are not set or are set incorrectly,

• The program is not running out of time,

• The program's ever starting - it's easy to make typing errors when naming your program,

• Hardware failure.

To be considered when using the Cranium cluster

• It is not permitted to run jobs locally on the submit nodes, cerebro-lsn1 and cerebro-lsn2, as well as the master nodes. Doing so will negatively impact the entire cluster. Jobs that inadvertently run on these machines will be subject to immediate termination to ensure proper functioning of the Cranium cluster.

• Use the appropriate queue when submitting jobs. Remember, the short.q is only for jobs of 2 CPU hours (or less) in duration. If you submit a job longer than 2 CPU hours into this queue, the system automatically kills your job.

• If your job hasn't started for days because someone has dozens of jobs ahead of yours, it's worth mailing the SysAdmin team,

• Do not run jobs on the cluster except through SGE,

• Do not hog large memory machines.
  
  

Feel free to contact the SysAdmin team to the email: clusteradmins@loni.ucla.edu if you have any question or concern.

No permission to read topic HardwareAndSoftwareOrderTracker - perhaps you need to log in?

No permission to read topic Infrastructure.IPTracking - perhaps you need to log in?

-- DavidHasson - 21 Aug 2007

No permission to read topic InstallingClusterNodes - perhaps you need to log in?

No permission to read topic IssueRequest - perhaps you need to log in?

No permission to read topic JumpStart - perhaps you need to log in?

How to Configure a DHCP + JumpStart Server

DHCP Server Configuration

Configure the DHCP server by using the service configuration utility

# dhcpconfig -D -r SUNWbinfiles -p /var/dhcp -a 192.168.4.1 -d data -l 84600 -h dns -y data
Created DHCP configuration file.
Created dhcptab.
Added "Locale" macro to dhcptab.
Added server macro to dhcptab - cerebro-sm.
DHCP server started.

where
-D: configuring DHCP service
-r: data storage module type
-p: DHCP directory path
-a: DNS server IP
-d: DNS domain name
-l: lease length used for address
-hresource in which to place host data, i.e., NIS or DNS
-y: DNS or NIS domain server

Configure the network for DHCP service and create the corresponding macro for that network

# dhcpconfig -N 192.168.0.0 -t 192.168.7.254
Added network macro to dhcptab - 192.168.0.0.
Created network table.

where
-N: network IP
-t: router

Add the PXE client macros to the DHCP configuration, by including vendor options

# dhtadm -A -m PXEClient:Arch:00000:UNDI:002001 -d ':BootSrvA=192.168.4.4:'
# dhtadm -A -s SinstNM -d 'Vendor=SUNW.i86pc,11,ASCII,1,0'
# dhtadm -A -s SinstPTH -d 'Vendor=SUNW.i86pc,12,ASCII,1,0'
# dhtadm -A -s SinstIP4 -d 'Vendor=SUNW.i86pc,10,IP,1,1'
# dhtadm -A -s SrootNM -d 'Vendor=SUNW.i86pc,3,ASCII,1,0'
# dhtadm -A -s SrootPTH -d 'Vendor=SUNW.i86pc,4,ASCII,1,0'
# dhtadm -A -s SrootIP4 -d 'Vendor=SUNW.i86pc,2,IP,1,1'
# dhtadm -A -s SjumpsCF -d 'Vendor=SUNW.i86pc,14,ASCII,1,0'
# dhtadm -A -s SsysidCF -d 'Vendor=SUNW.i86pc,13,ASCII,1,0'
# dhtadm -A -s SbootURI -d 'Vendor=SUNW.i86pc,16,ASCII,1,0'
# dhtadm -M -m 192.168.0.0 -e 'SinstNM="cerebro-sm"'
# dhtadm -M -m 192.168.0.0 -e 'SinstPTH="/export/install/x86_10"'
# dhtadm -M -m 192.168.0.0 -e 'SinstIP4=192.168.4.4'
# dhtadm -M -m 192.168.0.0 -e 'SrootNM="cerebro-sm"'
# dhtadm -M -m 192.168.0.0 -e 'SrootPTH="/export/install/x86_10/Tools/Boot"'
# dhtadm -M -m 192.168.0.0 -e 'SrootIP4=192.168.4.4'
# dhtadm -M -m 192.168.0.0 -e 'BootFile=nbp.SUNW.i86pc'
# dhtadm -M -m 192.168.0.0 -e ':Subnet=255.255.252.0:'
# dhtadm -M -m 192.168.0.0 -d ':Subnet=255.255.252.0:
      Router=192.168.7.254:Broadcst=192.168.0.255:NISdmain="loni":
      NISservs=192.168.4.2:SinstNM="cerebro-sm":SinstPTH="/export/install/x86_10":
      SinstIP4=192.168.4.4:SrootNM="cerebro-sm":SrootPTH="/export/install/x86_10/Tools/Boot":
      SrootIP4=192.168.4.4:BootFile="nbp.SUNW.i86pc":DNSdmain="data":DNSserv=192.168.4.1:' (all in a single line)
# dhtadm -M -m cerebro-sm -d ':Include=Locale:Timeserv=192.168.4.4:
      LeaseTim=84600:LeaseNeg:DNSdmain="data":DNSserv=192.168.4.1:
      SinstNM="cerebro-sm":SinstPTH="/export/install/x86_10":SinstIP4=192.168.4.4:
      SrootNM="cerebro-sm":SrootPTH="/export/install/x86_10/Tools/Boot":
      SrootIP4=192.168.4.4:BootFile="nbp.SUNW.i86pc":NISdmain="loni":
      NISservs=192.168.4.2:' (all in a single line)

BootSrvA: the IP address of the boot (TFTP) server
"Vendor" refers to vendor related options
"Sinst" refers to the JumpStart installing directory
"Sroot" refers to the JumpStart booting directory
BootFile: the filename of the NBP executable located into the boot server, executed by the PXE

To print the macro

# dhtadm -P

Now, verify if the DHCP daemon is running and the service is enabled

# ps -efl | grep -i dhcp
0 S root 1855 1 0 40 20 ? 1612 ? 17:42:56 ? 0:01 /usr/lib/inet/in.dhcpd
# svcs -a | grep dhcp

Create the named table

# pntadm -C 192.168.4.0

List the coonfigured DHCP networks

# pntadm -L
192.168.0.0
192.168.4.0

Add the clients into the DHCP server

# pntadm -A 192.168.5.56 -c cerebro-B-056 -f PERMANENT -i 0100093D118F85 -m 192.168.0.0 192.168.4.0

Here, we are adding node 'cerebro-B-056' as a DHCP client.

-A: IP of the node
-c: name of the node
-f: dynamic or static IP (PERMANENT is for static)
-i: Mac Address (there is a 01 before to the mac address; exclude the ":")
-m: macro (192.168.0.0) + subnet (192.168.4.0)

To check if the machine is already a DHCP client.

# pntadm -P 192.168.4.0 | grep -i b-056
0100093D118F85 01 192.168.5.56 192.168.4.4 Forever 192.168.0.0 cerebro-B-056

If you would like to use dynamic IPs

# pntadm -A 192.168.5.56 -c cerebro-B-056 -f DYNAMIC -i 0100093D118F85 -m 192.168.0.0 192.168.4.0

Add the corresponding entries in the /etc/ethers file. Declare the following variables into the /etc/inet/dhcpsvc.conf file

# echo "INTERFACES=bge0" >> /etc/inet/dhcpsvc.conf
# echo "HOSTS_RESOURCE=dns" >> /etc/inet/dhcpsvc.conf
# echo "HOSTS_DOMAIN=data" >> /etc/inet/dhcpsvc.conf
# echo "ICMP_VERIFY=TRUE" >> /etc/inet/dhcpsvc.conf
# echo "RESCAN_INTERVAL=15" >> /etc/inet/dhcpsvc.conf
# echo "UPDATE_TIMEOUT" >> /etc/inet/dhcpsvc.conf
# echo "BOOTP_COMPAT=automatic" >> /etc/inet/dhcpsvc.conf

Restart the DHCP daemon

# pkill -HUP in.dhcpd

To list the configured DHCP networks

# pntadm -L

JumpStart Server Configuration

First, let's load the Solaris OS distribution into the server. Create the parent install directory

# mkdir /export/install

Create a subdirectory for the Solaris OS distribution

# mkdir /export/install/x86_10

Mount the Solaris OS distribution. In this procedure, we assume an image of the OS is stored on disk. Copy the Solaris OS distribution into the created installation directory

# mount -F hsfs -o ro `lofiadm -a /path_to_isoImage` /mount_point_path
# cd /mount_point_path/Solaris_10/Tools
# ./setup_install_server -b /export/install/x86_10

Verifying target directory...
Calculating the required disk space for the Solaris_10 product
Calculating space required for the installation boot image
Copying the DVD image to disk...
Copying Install Boot Image hierarchy...
Install Server setup complete

the -b option specifies that the server is both, installing and boot servers. Now, create the configuration server directory

# mkdir /export/install/jumpstart
# cd /mount_point_path/Solaris_10/Misc/jumpstart_sample
# cp check /export/install/jumpstart
# cd /export/install/jumpstart
# cat > rules

# rule_keyword	rule_value	begin	profile	finish
any	-	preInstall	cerebroData	postInstall

Configure preInstall and postInstall scripts, the corresponding profile (cerebroData) as well as the jumpstart configuration script, sysidcfg. Then, verify the configuration file syntax

# ./check

Validating rules...
Validating profile cerebroData...
The custom JumpStart configuration is ok.

Now, share the installation directory and check if the NFS service is running

# cat >> /etc/dfs/dfstab
share -F nfs -o ro,anon=0 -d "Install Server" /export/install
# /etc/init.d/nfs.server start (or)
# svcadm enable -s svc:/network/nfs/server:default
# showmount -e localhost
export list for localhost:
/export/install (everyone)
# shareall
# share
-            /export/install ro,anon=0 "Install Server"

Next step is to configure client access

# cd /export/install/x86_10/Solaris_10/Tools
# ./add_install_client -d -s cerebro-sm:/export/install/x86_10 -c cerebro-sm:/export/install/jumpstart -p cerebro-sm:/export/install/jumpstart -b "output-device=ttya" -b "input-device=ttya" SUNW.i86pc i86pc

LONI SSH Gateways

SSH is short for Secure Shell. It provides a secure, encrypted method of transmitting information over TCP/IP to ensure the confidentiality, integrity and authenticity of transferred data. This is the preferred method of connectivity to LONI's Unix and Linux computational resources.

To access the LONI network via SSH:

1. Obtain an SSH client for Windows, Linux, or MacOSX
   • To find out if you already have ssh on a Unix/Linux based machine, type which ssh
2. SSH to ssh.loni.ucla.edu over port 22 using the your LONI username and password.
3. If accessing LONI for the first time, please change your password by following the instructions here.

-- DavidHasson - 19 Mar 2009

Configuring a Linux box during OS Installation:

• Add Linux box to NIS domain:
  • domain: loni
  • servers: 128.97.134.82,128.97.134.43,128.97.134.45,128.97.134.46

• Configure NTP
  • time server: 128.97.134.82

• Allow only SSH connections in Firewall Config

Configuring a Linux box post OS Installation:

• visudo
  • uncommment line '%wheel ALL=(ALL) ALL' to allow sudo
  • add line 'Defaults logfile=/var/log/sudo.log' to log sudo use

• add usernames w/ sudo privileges in /etc/group:wheel

• copy doig:/etc/fstab LONI entries into machine's /etc/fstab

• create NFS mountpoint directories

• mount using 'mount -va'

• edit /etc/ssh/sshd_config to EXPLICITLY disable PermitRootLogin^?
  • 'PermitRootLogin no'

• edit /etc/hosts.deny by adding line 'ALL: ALL'

• edit /etc/hosts.allow by adding line 'sshd: ALL'

No permission to read topic LinuxVirtualServer - perhaps you need to log in?

Darwin Streaming Server

Wirecast

-- DavidHasson - 22 May 2008

LONI Services account

To get a LONI Services account, go to the LONI Services login page.

Calendars

LONI calendars are available here

Pipeline accounts

Pipeline accounts depend upon a LONI Services account. You must have your account enabled for Pipeline access. Please contact Cornelius Hojatkashani (chojatk@loni.ucla.edu) for more details

The following articles should outline instructions for common tasks involving the use of LONI resources.

Access the LONI Network

Access LONI E-mail Accounts^?

Use a printer within the LONI Lab

Changing your LONI password

Set up your machine with X11 - If you need to display graphical output from a Unix/Linux machine that you are remotely logged into, you need a X11 Server (a program that accepts connections to display on your screen). This guide applies to Windows, Mac OS X, and Linux.

-- DavidHasson - 27 Sep 2007

No permission to read topic MacOSXHints - perhaps you need to log in?

• Download from Carbon Emacs site
• Mount disk image
• Copy to Applications folder

• Download and install Fink
• Download and install Fink Commander
• Run Fink Commander

• Install the "tetex" package binary (current version 3.0-1) by clicking on the blue "+" at the top left corner. See screenshot above. This will take about 30-60 minutes

• Download from Smart CVS site
• Mount disk image
• Copy to Applications folder

Once you have it installed, the LONI CVS information can be found at here

No permission to read topic MantisLDAP - perhaps you need to log in?

LONI Support Request

Name	Type	Size	Values	Tooltip message
TopicClassification?	select	1	NoDisclosure?, PublicSupported?, PublicFAQ?	blah blah...
OperatingSystem?	checkbox	3	OsHPUX?, OsLinux?, OsSolaris?, OsWin?	blah blah...
OsVersion?	text	16		blah blah...

NRB Datacenter

-- DavidHasson - 05 Aug 2008

NRB Datacenter

-- DavidHasson - 5 Aug 2008

Data Center Power Distribution

MarcSycip - 11 Sep 2008

Marc Sycip

MarcSycip - 11 Sep 2008

Matlab Installation Instructions

INSTALLING MATLAB 2008a

• Navigate to \\achilleos\software\Windows\Mathematics\MatLAB

• Run "setup.exe" in the "Matlab2008a-win32" folder for 32-bit Windows, or the "Matlab2008a-win64" folder for 64-bit.
  • Note: You don't need to mount it, or copy the files like you might for some other installs

• Matlab 2008a now uses a FLK "File License Key" - which is basically just a renamed PLP. In both cases, it just uses it to figure out what you're licensed to install, as there are tons of toolkits, etc.
  • This key is located here: \\achilleos\software\Windows\Mathematics\MatLAB\Licenses\matlab2008a-flk.txt
  • The current key is "00418-05741-12681-23448-02257-40048" for all platforms. If it gets updated, it will be current in the file above.

• The license file to use for all versions of Matlab using the network license is here: \\achilleos\software\Windows\Mathematics\MatLAB\Licenses\matlab-client-license.dat

• You can paste the above URL directly into the box when setup asks for the location. You might need to copy it or use an fqdn in the path if your computer isn't local or doesn't resolve LONI names correctly when on the VPN.

• Install everything to get the desired toolboxes.

MarcSycip - 26 Aug 2008

Infrastructure Meeting Archives

Current Meetings - Meeting notes for the current year (2007).

2006

August 7, 2006

August 14, 2006

August 21, 2006

August 29, 2006

September 5, 2006

September 13, 2006

September 18, 2006

September 25, 2006

October 27, 2006

November 3, 2006

-- DavidHasson - 20 Aug 2007

No permission to read topic MeetingSummary - perhaps you need to log in?

No permission to read topic MeetingSummaryProfile - perhaps you need to log in?

No permission to read topic MeetingSummarySite - perhaps you need to log in?

No permission to read topic MgmtNetwork - perhaps you need to log in?

Exchange 2007

Exchange 2007 no longer has the frontend/backend topology. It is divided up into server roles: Edge Transport, Hub Transport, Mailbox, Client Access, and Unified Messaging. A good website describing these roles can be found here (http://www.msexchange.org/tutorials/Introduction-Exchange-2007-Server-Roles.html) Also Exchange 2007 requires a 64-bit version of windows to run. Mailbox Servers (bellow are some bullet points about the exchange mailbox servers only) Prerequisite: Before installing exchange a few programs need to be installed: IIS, .NET 2.0 Framework, Microsoft Management Console 3.0, and Microsoft Windows PowerShell^?.

• Clustering:
  • Only the mailbox servers can be clustered. This means that the servers in the cluster can only have the mailbox exchange serve role and none of the others.

For instructions on how to setup a windows cluster for exchange see Chapter 31 in “SAMS Exchange Server 2007 UNLEASHED.”

• Storage:
  • The two mailbox servers are connected to the Sun3510 storage array (See the 3510 write-up for more info) via fibre channel. The servers have a duel port fibre card which allows for load balancing across the two Brocade switches.

• MPIO:
  • In order for the mailbox server to address the disks properly MPIO software needs to be installed. Without the MPIO software each server would see double the amount of disks that it should. The reason is that disk info is being mapped to both ports on the fibre card. What the MPIO software does is it recognizes that the same disks sets are being sent through on both ports and it classifies one port as a failover port. This way the server can see the appropriate amount of disks and the server can function as it should. The MPIO software that is being used can be found on Sun’s website. Also keep in mind that these are 64-bit machines and require MPIO software that is compatible with the 64-bit architecture. When you run the MPIO software you can type “?” for a list of all the commands. The two commands that matter are devinfo and pathinfo. These two commands allow you to manipulate the way in which MPIO functions. For example, to set all the disk to run in failsafe mode you would type: devinfo all FailOver^?.

If you type devinfo or pathinfo you can receive an output of the state of the disks. Online documentation is rather sparce. I found something helpful here: http://209.85.165.104/search?q=cache:iD_QajataMsJ:www.dothill.com/assets/pdfs/SN_MPIO_83-00004226_RevC.pdf+dsmcli&hl=en&ct=clnk&cd=1&gl=us#6 However, I am not sure how long this website will be up. Another option would be to google “dsmcli”.

No permission to read topic ModifyDNS - perhaps you need to log in?

JonathanPierce^? - 07 Aug 2008

first rev, no permissions

second rev, adding correct permissions

third rev, now i messed things up horribly

• Set ALLOWTOPICVIEW = TwinGeneticsGroup^?
• Set ALLOWTOPICCHANGE = SysadmGroup

No permission to read topic MySQL - perhaps you need to log in?

NetWorker Troubleshooting

Connection refused / Connection timed out

This indicates that the Networker server wasn't able to connect to the Networker client on the target machine.

1. Verify that the client machine is online and reachable over the network.
2. Check whether Networker client is installed and running on the client machine.
3. Check Networker's 'nsrports' setting: type 'nsrports' on the client, it should return '7937-7967'
4. Check iptables configuration on the client, it should allow incoming traffic on ports 7937-7967.
5. Telnet from Networker server to the client on port 7937 to verify that connection to the client can be made from server.

Probe job had unrecoverable failure(s), please refer to daemon.raw

1. Check for connectivity issues between the Networker server and the client.
2. Move the client to the "test" savegroup (or create a new subgroup and move the client to it). Then, try to probe the savegroup with command "savegrp -vvvp test" (substitute the new savegroup name if necessary). If probe succeeds, try to run the savegroup with command "savegrp -vv test". All commands must be run from the Networker server.

Mount point X already exists

This error only occurs when backing up a Virtual Machine through a VCB proxy (for LONI, the proxy is fiske.loni.ucla.edu). The error indicates that previous backup job did not dismount the Virtual Machine completely (it could have been interrupted, etc).

1. Login to the VCB proxy and delete the folder X that is referenced in the error message. Also, there is an option in VCB settings file that can take care of that issue by deleting previously mounted folders before attempting a backup job.

For more information about VCB, see this page.

The virtual machine identified by X is unknown

This error only occurs when backing up a Virtual Machine through a VCB proxy (for LONI, the proxy is fiske.loni.ucla.edu). The error indicates that the VCB proxy could not find and contact the virtual machine on the ESX server. More detailed description of the error suggests that the virtual machine may be powered off, or VMware Tools might not be installed on it. If the virtual machine is specified via its name (instead of its FQDN or IP address) then that name must be unique.

Troubleshooting the error:

1. Login to the VCB proxy, open command prompt, and navigate to the VCB folder
2. Run the 'vcbVmName' command to test whether the VCB proxy can reach the virtual machine:
   1. vcbVmName -h SERVER -u USER -p PASS -s ipaddr:VM
      1. SERVER is the ESX server (available in the VCB config file)
      2. USER is the username for the ESX server (available in the VCB config file)
      3. PASS is the password for the above (available in the VCB config file)
      4. VM is the FQDN or IP of the virtual machine (the X)

The above command should find the virtual machine on the ESX server if everything is configured correctly.

For more information about VCB, see this page.

Client X is not properly configured on the NetWorker Server

This error indicates that there is no configuration exists for the given client on the Networker server. This error may also come up when the client's IP address does not correctly resolve to its fully qualified domain name.

1. Check whether the Networker client is listed in the Networker Management Console, under Configuration->Clients.
2. Check whether the client's IP address correctly resolves to its fully qualified domain name, and vice versa.

1 retry attempted, (interrupted) exiting, aborted due to inactivity

~

Stuck on "starting session: state active"

~

Saveset "NMCASA:/gst_on_vodalus/lgto_gst" is stuck on "starting session: state active"

The workaround reported by the EMC tech. support is as follows:

1. Stop the group
2. Kill savepsm process
3. Stop gst service
4. Kill dbsrv9 as warrants
5. /etc/init.d/gst start
6. /etc/init.d/gst stop

At this stage, if kill is returned during the stop, then start gst and stop until kill is not returned.

There is already a machine using the name: "X"

The error indicates that there is a cached security certificate on Networker server that conflicts with the corresponding client certificate on the client. This can happen when the client's OS has been reinstalled or Networker software on the client has been reinstalled. The newly generated security certificate on the client does not match the one on the server.

The workaround reported by the EMC tech. support is as follows:

1. Login to the Networker server as root
2. Type "nsradmin -s SERVER -p nsrexec" (where SERVER is the FQDN or IP of the Networker server)
3. Type "print type: nsr peer information" to see cached security certificates
4. Type "delete name: X" to delete cached security certificate for host X

This deletes the old cached security certificate which causes these messages.

Networker Commands

• Add tape volumes to a newly configured jukebox:

for (( i = 500000 ; i <= 500049 ; i++ )) ; do nsrjb -a -s vodalus.loni.ucla.edu -v -T $i 2>&1 | tee -a nsrjb-a-T_500000-500049.log ; done

• Change location for multiple tapes:

for (( i = 500000 ; i <= 500049 ; i++ )) ; do mmlocate -u -n $i 'STK9310' 2>&1 | tee -a mmlocate_500000-500049.log ; done

• Run tape inventory on multiple slots:

for (( i = 1 ; i <= 50 ; i++ )) ; do nsrjb -I -S $i -v 2>&1 | tee -a nsrjb-I-S_1-50.log ; done

LinhNamVu^? - 20 Jul 2007

install red hat 5

- applications/games and entertainemnt

- applications/graphics

- applications/sound and video

- applications/text-based internet

+ development/dev libaries

+ development/dev tools

+ development/legacy software dev

+ servers/dns

+ servers/mail

+ servers/mail/postifx

+ servers/mail/mailman

- servers/mail/sendmail

- servers/print support

- servers/server config tools

on next boot:

set ntp servers

set dns information

turn SELINUX to permissive (will still report errors that selinux would normally not allow, but would still allow it)

copy over /etc/yum.conf

copy over /var/named/* (make sure there is a chroot directory)

copy over /var/mail/spamassasin/*

copy over /var/named.conf

copy over /var/spool/postfix/*

Installing Red Hat-based distributions over HTTP

This guide applies to Red Hat, CentOS^?, Fedora and possibly other distributions as well. This example is for Red Hat AS4 i386. Other distributions should be the same, except you will need to point to the appropriate HTTP repository.

• Begin by booting off Installation Disc 1 or a boot-only disc for the distribution you would like to install.
• At the introductory 'boot:' prompt enter 'linux askmethod'
  • The installation kernel loads as normal, except it enters text mode.
• For language select the default 'English'.
• Keyboard is typically the default 'us'.
• For Installation Method select 'HTTP'
• Configure an IP Address using either DHCP or static IP. This is the only the IP for the installation itself, so you will have another opportunity to change the system's permanent IP address later on.
• For 'HTTP Setup', point to the correct server and directory. In this case 'Web site name:' can be either 'yum' or 'yum.loni.ucla.edu' and 'Red Hat Enterprise Linux directory:' is 'rhel4as-i386/disc1'
  • To confirm that you have a proper installation directory, in a web browser check that topdir contains a RedHat^?/base directory.
• When asked if you would like to test the CD, you can select 'skip'. This is typically unnecessary especially since all packages will be fetched over HTTP instead of CD.
  • If all goes well you should see an indication that the local installation media was detected successfully.
  • Anaconda (the installation GUI) now loads.
• Proceed with the Linux installation as usual.

Network Maps

Equipment

NRB Datacenter: NRB1 225A

Auxiliary Equipment Locations: NRB1, Reed

Network

NRB Floor: NRB1 225^?

LONI Network: Layer 2/3^?

LONI Network: 10ge Phase II Vision^?

Power

NRB Datacenter: NRB1 225A Floor Power Distribution

NRB Datacenter: NRB1 225A Breaker Diagrams^?

NRB Datacenter: NRB1 225 General Power^?

-- DavidHasson - 13 Mar 2008

No permission to read topic NetworkPortScanning - perhaps you need to log in?

No permission to read topic NetworkerClient - perhaps you need to log in?

Tips for New Systems

Current systems:

Workstations

• Dell Precision 390
• Core 2 2.0Ghz or better
• 4Gb Memory or better
• Graphics card:
  • nVidia FX550 or better for standard users
  • nVidia FX3450 or better for graphics
• 1394 Card
• No floppy
• 5-Button Optical Mouse
• 160Gb SATA or better
• DVD/RW Drive

Desktops

• Dell Optiplex 755 Mini Tower
• XP Home Edition (we have a XP Pro VLK)
• Core 2 2.0Ghz or better
• 2Gb Memory or better
• Graphics card:
  • ATI Radeon 2400 XT or better
  • nVidia 8800 or better
• 1394 Card
• No floppy
• 5-Button Optical Mouse
• 160Gb SATA or better
• DVD/RW Drive

Monitors

• Dell 2007FP
• Dell 2407WFP

General Ideas

• Purchase with the price break. Usually prices go up linearly with performance, but with a jump right where the premium products start
• Don't price out a machine that is no better than spare machines we have in the lab.
• Always get the cheapest warranty, and the cheapest OS.
• Purchase nVidia if possible, our current software is developed on nVidia

-- DavidHasson - 05 Oct 2007

Special List for the Cranium's Compute Nodes

This page described the different lists used by agentSmith to monitor the entire Cranium cluster. agentSmith uses three host lists to work. All these lists are located in cerebro-admin:/usr/local/loni/admin/nodes and are described as follow:

• reportingList - This file contains information of all the compute nodes listed to serve in the Cranium cluster. Each line describes a compute node with its hostname, data IP and mgmt IP. This file is not needed to be edited unless we have new computers coming into the Cranium cluster or we decide to change any of the identifiers of the compute nodes (hostname or IPs).

• blackList - The file describing this list contains the hostname of the compute nodes reported with special issues like the ones powered off because of maintenance or disabled from SGE due to special purposes. These nodes are prevented to be monitored by agentSmith. As a consequence, agentSmith can't change the SGE status of the node or a down node can't be powered off or receive reinstall instructions. When editing this file, just add the hostname of the compute node you would like to include. The file group is owned by admin and its permissions are 664.

• hwIssuesList - Compute nodes contained on this list are the ones declared with hardware issues. The list prevents agentSmith to take care of them. All the nodes included on this list are supposed to be powered off. This list must be validated with the information we have on the Disabled Cranium Production Nodes page.

To edit any of the two lists, blackList and hwIssuesList, use this link. Note, this page is only accessible by the SysAdmin group members and within the LONI network. Each time the list is updated via web, an email report will be sent to the cluster mailing list. Also, remember there is the option to edit these files from the command line by accessing cerebro-admin.

Created by Hugo Hernandez-Mora on Mon Jun 15th, 2009 @ 13:53:26

DmitriyVinogradov^? - 15 Jun 2009

OSTicket Changes

Ticket Lock / Unlock Patch

The ticket unlocking functionality was added to OSTicket using the instructions found in http://www.osticket.com/forums/showthread.php?t=1066 (credit to webPragmatist).

The following files have been modified:

/scp/tickets.php

case 'process':
    $isdeptmanager=($deptId==$thisuser->getDeptId())?true:false;
    switch(strtolower($_POST['do'])):
        // *** Added by webPragmatist
        case 'unlock_exit':
            if($ticket->isLocked() && ($lock=$ticket->getLock()) && !$lock->isExpired()) {
                $lock->expire();
                header('Location: tickets.php');
            }
            break;
        // *** End addition by webPragmatist

/include/class.lock.php

function renew() {
    global $cfg;

    $sql='UPDATE '.TICKET_LOCK_TABLE.' SET expire=DATE_ADD(NOW(),INTERVAL '.$cfg->getLockTime().' MINUTE) '.
        ' WHERE lock_id='.db_input($this->getId());
    //echo $sql;
    if(db_query($sql) && db_affected_rows()) {
        $this->reload();
        return true;
    }
    return false;
}

// *** Added by webPragmatist
// Expire / remove existing lock.
function expire() {
    $sql='DELETE FROM '.TICKET_LOCK_TABLE.
        ' WHERE lock_id='.db_input($this->getId());
    //echo $sql;
    if(db_query($sql) && db_affected_rows()) {
        $this->reload();
        return true;
    }
    return false;
}
// *** End addition by webPragmatist

/include/staff/viewticket.inc.php

<select id="do" name="do" 
  onChange="this.form.ticket_priority.disabled=strcmp(this.options[this.selectedIndex].value,'change_priority','reopen','overdue')?false:true;">
    <option value="">Select Action</option>
    <!-- Added by webPragmatist -->
    <option value="unlock_exit">Unlock &amp; Exit</option>
    <!-- End addition by webPragmatist -->

Additional Columns in Ticket Summary Table

A new column labeled "Assigned To" was added to the ticket summary table that is seen from staff and admin control panel. For this to happen, the following changes were made:

/include/staff/tickets.inc.php

--- tickets.inc.php.old 2009-06-16 11:49:18.000000000 -0700
+++ tickets.inc.php     2009-06-16 12:53:19.000000000 -0700
@@ -158,10 +158,11 @@ $pagelimit=$pagelimit?$pagelimit:PAGE_LI
 $page=($_GET['p'] && is_numeric($_GET['p']))?$_GET['p']:1;


-$qselect = 'SELECT ticket.ticket_id,lock_id,ticketID,ticket.dept_id,ticket.staff_id,subject,name,email,dept_name '.
+$qselect = 'SELECT ticket.ticket_id,lock_id,ticketID,ticket.dept_id,ticket.staff_id,st.username,subject,name,ticket.email,dept_name '.
            ',status,source,isoverdue,ticket.created,pri.* ';
 $qfrom=' FROM '.TICKET_TABLE.' ticket LEFT JOIN '.DEPT_TABLE.' dept ON ticket.dept_id=dept.dept_id '.
        ' LEFT JOIN '.TICKET_PRIORITY_TABLE.' pri ON ticket.priority_id=pri.priority_id '.
+       ' LEFT JOIN '.STAFF_TABLE.' st ON ticket.staff_id=st.staff_id '.
        ' LEFT JOIN '.TICKET_LOCK_TABLE.' tlock ON ticket.ticket_id=tlock.ticket_id AND tlock.expire>NOW() ';

 //get ticket count based on the query so far..
@@ -328,7 +329,8 @@ $basic_display=!isset($_REQUEST['advance
                 <a href="tickets.php?sort=dept&order=<?=$negorder?><?=$qstr?>" title="Sort By Category <?=$negorder?>">Department</a></th>
                <th width="70">
                 <a href="tickets.php?sort=pri&order=<?=$negorder?><?=$qstr?>" title="Sort By Priority <?=$negorder?>">Priority</a></th>
-            <th width="150" >From</th>
+            <th width="150">From</th>
+            <th width="150">Assigned To</th>
         </tr>
         <?
         $class = "row1";
@@ -361,6 +362,7 @@ $basic_display=!isset($_REQUEST['advance
                 <td nowrap><?=Format::truncate($row['dept_name'],30)?></td>
                 <td class="nohover" align="center" style="background-color:<?=$row['priority_color']?>;"><?=$row['priority_desc']?></td>
                 <td nowrap><?=Format::truncate($row['name'],22,strpos($row['name'],'@'))?>&nbsp;</td>
+               <td nowrap><?=$row['username']; ?> </td>
             </tr>
             <?
             $class = ($class =='row2') ?'row1':'row2';

Brief explanation of changes:

1. First change modifies the SQL query that is used to retrieve tickets for display in the table. An additional LEFT JOIN statement is added that links ticket's staff ID to the staff table.
2. Second change modifies the table header, adding a new column labeled as "Assigned To".
3. Third change modifies the loop which outputs one table row at a time, adding a new cell that holds the username of the staff person to which the ticket is assigned (if any).

Fixing "From Address" Bug

OSTicket sent emails with a wrong "From" address, deviating from the settings. The fix is below.

/includes/class.misc.php

--- class.misc.php.old   2008-06-05 12:47:36.000000000 -0700
+++ class.misc.php   2009-02-13 17:35:36.000000000 -0800
@@ -64,9 +64,9 @@
         $subject=preg_replace("/(\r\n|\r|\n)/s",'', trim($subject));
         $message = preg_replace("/(\r\n|\r)/s", "\n", trim($message));
         #Headers
-        $headers .= "From: ".$fromname."<".$fromaddress.">".$eol;
-        $headers .= "Reply-To: ".$fromname."<".$fromaddress.">".$eol;
-        $headers .= "Return-Path: ".$fromname."<".$fromaddress.">".$eol;
+        $headers .= "From: <".$fromaddress.">".$eol;
+        $headers .= "Reply-To: <".$fromaddress.">".$eol;
+        $headers .= "Return-Path: <".$fromaddress.">".$eol;
         $headers .= "Message-ID: <".time()."-".$fromaddress.">".$eol;
         $headers .= "X-Mailer: osTicket v 1.6".$eol;
         if($xheaders) { //possibly attachments...does mess with content type

Brief explanation of changes:

The $headers string building code was simplified, after which the problem disappeared.

CURS Authentication for Staff

The capability to authenticate users through LONI's CURS system to allow access to staff/admin panel was implemented with a few changes to the Staff PHP class.

/includes/class.staff.php

--- class.staff.php.old   2008-03-13 23:53:32.000000000 -0700
+++ class.staff.php   2009-04-17 13:15:46.000000000 -0700
@@ -14,6 +14,19 @@
     vim: expandtab sw=4 ts=4 sts=4:
     $Id: OSTicketChanges.txt,v 1.4 2009/07/03 00:53:30 dvinogra Exp wwwprod $
 **********************************************************************/
+
+/*
+$fname = "/var/www/osticket/debug.txt";
+if (!$fp)
+   $fp = fopen($fname, "a");
+*/
+function dw($line)
+{
+   global $fp;
+   if ($fp)
+      fwrite($fp, date("D M j G:i:s") . ">\t" . $line . "\n");
+}
+
 class Staff {
     
     var $udata;
@@ -36,13 +49,45 @@
         return ($this->lookup($var));
     }
 
+
+// Start modifications by LONI
     function lookup($var){
 
+   dw("called lookup($var)");
+
+   if (empty($var))
+      return FALSE;
+
         $sql=sprintf("SELECT * FROM ".STAFF_TABLE." LEFT JOIN ".GROUP_TABLE." USING(group_id) WHERE %s ",is_numeric($var)?"staff_id=$var":"username='$var'");
      
         $res=db_query($sql);
-        if(!$res || !db_num_rows($res))
-            return NULL;
+
+        if(!$res || !db_num_rows($res))      // if user is not found, we can try something else
+   {
+      dw("\tuser not found in database");
+      $em = $this->lookup_access_file($var);
+      if ($em != NULL)   // check if the user is in a local access file
+      {
+         dw("\tlookup_access_file($var): user found in local access file");
+         if ($this->add_curs_user($var, $em) != NULL)   // if successful, try to add user to database or return NULL
+         {
+            dw("\tadd_curs_user($var, $em): user added to database");
+            $res = db_query($sql);   // query the database again, now that CURS user has been added to it
+            if(!$res || !db_num_rows($res))   // sanity check - at this point database should have CURS user added to it
+               return NULL;   // for some reason, database still can't find the newly added user...
+         }
+         else
+         {
+            dw("\tadd_curs_user($var, $em): user couldn't be added to database");
+            return NULL;   // user was found in local access file, but couldn't be added to database
+         }
+      }
+      else
+      {
+         dw("\tlookup_access_file($var): user not found in local access file");
+         return NULL;   // user was not found in database and in local access file
+      }
+   }
 
         $row=db_fetch_array($res);
         $this->udata=$row;
@@ -59,6 +104,126 @@
 
         return($this->id);
     }
+   
+   function lookup_curs($un, $pw){
+
+      dw("called lookup_curs($un, ***)");
+
+      if (empty($un) || empty($pw))
+         return FALSE;
+
+      $auth_url = 'https://utilities.loni.ucla.edu/services/user?service=authenticateUser'; 
+      $info_url = 'https://utilities.loni.ucla.edu/services/user?service=getUserIds'; 
+      $var_username = 'userEmail'; 
+      $var_password = 'userPassword'; 
+
+      $auth_data = array($var_password=>$pw, 'password'=>'usER4Mee'); 
+      $info_data = array('password'=>'usER4Mee'); 
+
+      $auth_url .= "&" . $var_username . "=" . $un; 
+      $info_url .= "&" . $var_username . "=" . $un; 
+      
+      $info_response = $this->http_dopost($info_url, $info_data); 
+      $info_array = split(":", $info_response); 
+      
+      if ((count($info_array) >= 4) && ($info_array[0] == "SUCCESS") && !empty($info_array[3])) 
+         return TRUE;
+      else
+         return FALSE;
+   }
+   
+   function http_dopost($url, $data) {
+
+      $params = array('http' => array( 
+         'method' => 'POST', 
+         'content' => http_build_query($data) 
+      )); 
+       
+      $ctx = stream_context_create($params); 
+      $fp = @fopen($url, 'rb', false, $ctx); 
+       
+      if (!$fp)  
+         return false; 
+
+      $response = @stream_get_contents($fp); 
+      fclose($fp); 
+
+      return $response;
+   }
+   
+   function lookup_access_file($un){
+
+      dw("called lookup_access_file($un)");
+
+      if (empty($un))
+         return NULL;
+
+      $access_file = "/etc/osticket-staff";
+
+      if (!file_exists($access_file))
+         return NULL;
+
+      $arr_entries = explode("\n", file_get_contents($access_file));
+
+      if (count($arr_entries) < 1)
+         return NULL;
+
+      for ($i = 0; $i < count($arr_entries); $i++)
+      {
+         if (empty($arr_entries[$i]))
+            continue;
+
+         $arr_entry = explode(":", $arr_entries[$i]);
+
+         if (count($arr_entry) < 3)
+            continue;
+
+         dw("\t\tcompare: " . $arr_entry[1] . " == " . $un);
+
+         if (($arr_entry[1] == $un) && ($arr_entry[2] == "1"))
+            return $arr_entry[0];
+      }
+
+      return NULL;
+   }
+   
+   function add_curs_user($un, $em){
+
+      dw("called add_curs_user($un, $em)");
+
+      if (empty($un) || empty($em))
+         return FALSE;
+
+      $sql=' SET updated=NOW() '.
+         ',isadmin=1'.
+         ',isactive=1'.
+         ',isvisible=1'.
+         ',onvacation=0'.
+         ',dept_id=1'.
+         ',group_id=1'.
+         ',username='.db_input(Format::striptags($un)).
+         ',firstname='.db_input(Format::striptags($un)).
+         ',email='.db_input($em).
+         ',passwd='.rand(100,999).
+         ',change_passwd=0';
+
+      $sql='INSERT INTO '.STAFF_TABLE.' '.$sql.',created=NOW()';
+      
+      if(!db_query($sql) or !($uID=db_insert_id()))
+      {
+         dw("\tcould not add user to database");
+         return FALSE;
+      }
+      else
+      {
+         dw("\tadded user to database");
+         return TRUE;
+      }
+
+      return FALSE;
+   }
+// End modifications by LONI
+
 
     function reload(){
         $this->lookup($this->id);
@@ -69,9 +234,44 @@
     }
 
     /*compares user password*/
+
+// Start modifications by LONI
     function check_passwd($password){
-        return (strlen($this->passwd) && strcmp($this->passwd, MD5($password))==0)?(TRUE):(FALSE);
+
+   dw("called check_passwd(<length=" . strlen($password) . ">)");
+
+   if (empty($password))
+      return FALSE;
+
+   $passwd_ok = (strlen($this->passwd) && strcmp($this->passwd, MD5($password))==0)?(TRUE):(FALSE);
+   
+   if (!$passwd_ok && ($em = $this->lookup_access_file($this->username)) && $em != FALSE)   // if first check failed, it could be a CURS user
+   {
+      dw("\tfirst passwd check failed, trying CURS");
+
+      $passwd_ok = $this->lookup_curs($em, $password);
+      if ($passwd_ok)
+      {
+         dw("\tCURS auth successful");
+
+         $sql = "UPDATE ".STAFF_TABLE." SET updated=NOW()".
+            ",passwd='".md5($password)."'".
+            " WHERE staff_id=".$this->id;
+         if (!db_query($sql))
+         {
+            dw("\tfailed to update database with passwd");
+         }
+      }
+      else
+      {
+         dw("\tCURS auth failed");
+      }
+   }
+
+        return $passwd_ok;
     }
+// End modifications by LONI
+
 
     function getTZoffset(){
         global $cfg;

Brief explanation of changes:

1. First change adds a helper function named dw($line) which writes a string to a debug file. By default, the function's file open statement is commented out, which disables this behavior.
2. Second change inserts additional logic into the lookup($var) function. If upon a call to the function the user is not found in the database, a local text file is checked, and if match is found, the user is added to the database (however, user's password is not yet set).
3. Third change adds four more functions to the file, named lookup_curs($un, $pw), http_dopost($url, $data), lookup_access_file($un), and add_curs_user($un, $em).
   1. lookup_curs authenticates given username and password with CURS, using http_dopost function.
   2. http_dopost sends an HTTP POST request to a given URL with provided data.
   3. lookup_access_file checks a text access file for a given username (email address).
   4. add_curs_user adds a user with given username and email address to the database.
4. Last change modifies the check_passwd($password) function by adding necessary logic to enable CURS logins into the system.

Adding a new user to OSticket with CURS authentication

1. Simply edit /etc/osticket-users to add a new CURS user to OSticket. The file asks for user's short login name, user's email (which is the login name for CURS), and 1/0 enabled/disabled setting. It is advised that you also create this user through OSticket's admin panel, giving it appropriate full name, group and department membership, etc, and specifying any random password. If the new user logs in for the first time and CURS authenticates him successfully, the password will be updated.

No permission to read topic ObservingMRIScans - perhaps you need to log in?

ShanrenZhou^? - 18 Dec 2006

1. Go to http://www.crucial.com/company/contacts.asp

2. Click on the 'Live Chat with an expert' picture

3. Prove them with the following information

- Batch number (should start with a "B")

- CT part number

- Quantity

- Name

- Address

- Phone number

- E-mail address

- Exchange Method

4. For Exchange Method, choose 'Wait-to-Receive Exchange'

5. Return the items to the address that Crucial provides and be sure to write down

the RMA number

Old Infrastructure Home Page

Systems Administration FAQs

ComputerCenterSchema - A brief description about the LONI Computer Center

GridComputing - How to submit command-line jobs to the Cerebro and the Cerebellum Clusters

GettingOnline - How to get connected to the LONI network

LoniServices - How to access various LONI services

Video Teleconferencing - How to initiate a VTC connnection

Weekly Sysadmin Meetings Summary

MeetingSummarySite - Chronological List of the SysAdmin Meetings.

Systems Administration Guides

ACSLS - Administrate StorageTek^? tape silos

ActiveDirectoryBackup - Backup and Restore Active Directory

AddUser - script used to add a new UNIX user account on Severian

AdminScripts - useful perl scripts used for administering purposes

AgentSmith - In charge of monitoring and diagnosing the Cerebro cluster

Apache - How to setup Apache (2.0.58)

CatOS4Levitt - ACL for Levitt's Lab

CerebroRocks - Documentation on installing Rocks 4.3 in the Cerebro cluster.

ControllingCerebroCluster - Powerful perl scripts to interact with the Cerebro cluster

Dashboard Prototype

DelUser - script used to remove a current UNIX user account on Severian

DHCPSetUp - Setting DHCP Servers

EditingComputersDB - How to edit the LONI Computers Database

ExchangeCluster - Exchange 2007 Writeup

GettingOnline - Ethernet, Wireless, VPN

InstallingClusterNodes - Different methods used to install Solaris 10 on the Cerebro cluster nodes

JumpStart - How to install Solaris 10 on the Sun Fire V20z nodes via the JumpStart method

JumpStartDHCPServer - How to configure a DHCP and JumpStart server

LinuxVirtualServer - Administrate Linux Virtual Server

ModifyDNS - Add/edit DNS entries

MySQL - How to setup MySQL (4.1.22)

NetservInstallation - netserv installation process

NetworkInstallRedHat - Howto perform a network-based installation Red Hat Linux over HTTP

NetworkPortScanning - Port scanning by using "Network Mapper” for a network exploration at LONI

OpenDellCase - calling dell for part replacement

ObtainCrucialRMA - Obtain RMA from Crucial.com

OrderTracker - Current orders that have not been received

PC2WindowsSetup - Steps taken to install Windows on PC2 from reclaimed LVM space.

PatchLink - Info on LONI's patch management software

PHPIISHowTo - How to install PHP on a Microsoft IIS server

PostfixSpamassassin - Configuring Postfix and Spamassassin

PowerUpOrder - Order to power up machines

PropagatingConfigFiles - How to propagate configuration files

RebuildRAID - Rebuiling RAID mirror image

RedHatSystems - A list of LONI's Red Hat Linux systems by version

SyncCluster - How to set up a rsync server and its client nodes to synchronize the installed packages on the cluster

SecondaryDNS - Setting up a slave name server

SetupCerebellum - Maya6.5, Maya7, Rush, LWSN and Rocks

SetUpNIS - How to set up NIS server

SetupSamba - Setup Samba on Linux servers

SetupUnixAccounts - Setting up a new user account

SolarisGCC - Steps taken to install GCC on cerebro-cc (running Solaris 10)

SPCommonActions - Commonly used SP commands for the v20z and X2200 machines

Subversion - Subversion Howto's and help for Admins and Developers

SystemSetup - Steps to be taken to initially setup a WinXP^?, Linux, or OSX system

SynchronizingCerebroCluster - Steps taken to synchronize the Cerebro cluster

UbuntuWriteup - How to configure and reinstall Ubuntu on the Boxx

UsingFTP - How to put dir/files to be share by using FTP

UpdatingV20zSPandBIOS - How to update the BIOS and SP firmware in the Sun Fire V20z

VizFAQ - FAQ on installing and running the Visualization Cluster

VncBanksy - How to set up a VNC account for a user on Banksy

WhiteBoard - Polyvison Whiteboard System

WindowsServerRoles - Roles of various Windows 2003 server

WindowsXpRemoteDeploy - How to perform an Unattended Install of Windows

Yum - How to configure and use yum, for clients

YumServer - How to setup a yum repository for Red Hat Linux distributions

To be organized...

BuildingSecurity^?

CvsRepository

DellTricks

FileSystemPolicies^?

IssueRequest

MacOSXHints

ObservingMRIScans

ProgramLocations

RenewTomcatHTTPSCertificate

SharedGroupCalendars

SunCaseDB - Database of all opened cases since 02.16.07

SystemAccounts

SecuringSQLServer2005

RaidSetupOnSolaris10r606

EmailConfig

HowardChou^? - 17 Jun 2006

Dell # to call: 1-800-456-3355 extension 7260301

They will ask for your name: " " company name: it's either UCLA or KST email address: sysadm@loni.ucla.edu phone number: 310-267-5706 address; 635 Charles E. Young drive Suite 225 Los Angeles, CA 90095

The technical support will ask you to perform all sort of tests trying to get the machine working. Explaining to them you did all sort of testings or else you will sit there with him/her on the phone for several hours!

enjoy =)

Shanren Zhou - 12 Dec 2006

Dell Live Chat: http://support.dell.com/support/topics/global.aspx/support/gen/chat

1. Enter service tag: ##XXX## (can be found in the back of the machine) and click submit

2. You will need to fill out your personal contact information in the next page

3. As with the call support, try to convince them that you have done all the testing so you can have

the problem solved in prompt manner

Hardware and Software Order Tracker

Date-Ordered	Date Received	Requestor	Description	Vendor	Price	Quantity	Version	Comments	PO#
10/27/06	10/30/06	Brian	SPSS 14 for Windows	Software Central	$62 each	3	14
10/18/06	11/03/06	Brian	Adobe Creative Suite 2 for Mac	TRC	$384.50	1		In Sysadm office
10/31/06	11/14/06	Rico	Dell Precision 390 for Office group hardware refresh	Dell	$2,511.39	4		For Office group hardware refresh
10/31/06		JD	Server: Dual Core Xeon Processor 5150 4MB Cache, 2.66GHz, 1333MHz FSB, PE 1950 (222-3382)	Dell	$9,243.30	1		QUOTE #: 3264014861580NHE90500 & 1580NHE90600
10/31/06		JD	Server: QLogic 2462 Dual Port 4Gbps PCI-E FC HBA (222-0423)	Dell	$3,753.76	1		QUOTE #: 326403667	1580NHE90300
11/01/06	11/06/06	Ivo	100 GB 2.5" Hard Drives	CDWG	$149.62	2			1580NHE86100
11/01/06	11/06/06	Ivo	2.5" Hard Drive Enclosures	CDWG	$25.67	3			1580NHE86100
11/01/06		Rico	DVD Drive/floppy drive for Sun Fire V20z.	Sun	$936.00	6		Product Number: X9260A	1580NHE86300
11/01/06		Rico	4000WATT AC POWER SUPPLY FOR US (CABLE ATTACHED)	Network Hardware Resale	$2,250.00	1			1580NHE86400
11/01/06		Amanda	Keynote Pavilion Software	KeynotePro?	$19.95	1			1580QHE803
11/03/06		Rico	1.25" Flexible Nylon Wire Loom 50'	Cable Organizer	$144.95	2			1580QHE899
11/03/06		Rico	1.25" Flexible Nylon Wire Loom 50'	Cable Organizer	$144.95	2			1580QHE899
11/06/06		Rico	Solaris Only support	Dynamic Systems Inc	$2160.00
11/08/06		Rico	Sliding Rapid/Versa Rails Universal, PE 2950, Customer Install (310-8194)	Dell	$112.48 EACH	2
11/09/06	Received	Rico	Adobe After Effects	TRC	$198.90 EACH	2 x Windows, 2 x Mac	7 Professional	In Sysadm office
11/13/06		Rico	PNY NVIDIA Quadro FX 4500 X2 1GB PCIe Graphics Card	Dell	$3099.99	1	For the backup DIVE box for the P41 visit	Order Number: 200611138006
11/14/06		Rico	Router Board For An Origin 300 System	Great Eastern Technology	$1200.00 EACH	2	To upgrade our SGI fileservers from 4 processors to 8 processors each.		1580NHF23000
11/15/06		Rico	CT530854: 1GB, 184-pin DIMM Upgrade for a Dell PowerEdge? 2650 (533MHz FSB) System	Crucial	$747.71 Total	4
11/15/06		Rico	4 PAIR CAT6 PATCH CABLE	GoCables?.com	$17.91 EACH, Total: $124.12	6
11/22/06		Rico	Reference Manager for Windows	digitalriver.com	$247.94
11/22/06		C. Schwartz	Intel iMac 2.0GHz 20” Display (7013381)	UCLA Computer Store	$999			To be delivered to Craig's office.
11/23/06		Queenie	VMware Workstation 5 (for Windows)	VMware	$189.00		Electronic Download Version
11/28/06		Rico	Mathematica Network License Mgr & Mathematica Network License for Windows	Software Central	$80, $185	1 of Each
12/06/06		Rico	NUMALINK-1M-SU: Numalink Cable 1 M	Great Eastern Technology	$95.00 EACH	3		For Origin 300 server upgrade.
12/07/06		JD	146 GB 10,000 RPM Cheetah 10K.7 Ultra320 SCSI Internal Hard Drive (A0451508)	Dell	$345.00 Each, Total: $690.00	2		For the new patchlink server.
12/11/06	12/11/06	JD	RHN Management Module	Red Hat	$96.00 EACH	2		Order Number: W2811171
12/14/06	Received	Jonathan	BFG GeForce? 6200 OC 256MB DDR PCI Graphics Card	Best Buy	$156.99	1		For the frontend machine on our Visualization Cluster
12/15/06		Linda	CMS400..net Ephox MAC editor (10 users only)	Ektron	$359.00	1		License Key Case #00045516	1580NHG06500
01/02/07	01/12/07	Linh-Nam	John Van Horn's New Computer	Dell	$4,375.74	1		E-Quote Number: E008539813
01/03/07		Cornelius	MacBook? Pro 15.4"	Apple	$1,999	1
01/04/07		HongWei?	Hitachi 7200 RPM 250GB IDE ATA130 HD	NewEgg?	$69.99	1
01/04/07	Received	HongWei?	PNY VCG76512SAPB GeForce? 7600GS 512MB 128-bit GDDR2 AGP 4X/8X Video Card	NewEgg?	$142.99	1
01/12/07		Rico	KTK-15 Fuse	www.drillspot.com	$8.64 EACH	3		For Multimeter.
01/22/07		Rico	PowerEdge? 1950	Dell	$2,621.25	1		For the mail transport and spam filter for our 2007 Exchange setup.
01/22/07		Jonathan	13W3 to VGA adapter	www.monoprice.com	$8.46	1
02/09/07	02/16/07	Rico	IDE Cables	http://store.cwc-group.com/18ulata33661.html	$0.75 EACH	5
02/09/07	02/16/07	Rico	SATA Cables	http://store.cwc-group.com/seatas90180d.html	$1.65 EACH	5
02/09/07	02/22/07	Rico	SCSI Cables	http://www.computercablestore.com/detail.aspx?ID=574	$1.65 EACH	5
02/13/07		Allan	500 GB external hard drive	CDWG	$249.37	1		CDW Part: 993387	1580 N HH713 00
02/13/07	02/23/07	JD	Switch: WS-C2970G-24TS-E	Network Hardware	$2350	1		For servers in new production rack.
02/15/07		SVG Group	500 GB external hard drive	CDWG	$249.37	1		CDW Part: 993387
02/20/07		Rico	UPS	Dell	$2057.00 + $90 SH	3 Separate Items		For production rack.
02/27/07		J. Morra	Keyboard	CDWG	$39.89	1		CDW Part: 947770
03/01/07		Linh-Nam	Rack Enclosure	CDWG	$1,173.99	1		CDW Part: 920138
03/01/07		Linh-Nam	PDU	CDWG	$273.99	1		CDW Part: 730516
03/01/07		Shanren	146 GB SCSI Hard Drive	Dell	$319.00	1		For Poweredge 2850. Upgrade to Dell cluster's primary root drive.

JonathanPierce^? - 10 Jul 2007

The following outlines the steps taken to install Windows XP post-Linux install on PC2. We recovered a LVM partition sized roughly 40GB for the Windows install. The issue in this setup is that A) Windows obviously can't install to the logical volume and B) Windows setup hangs indefinitely after "Setup is inspecting your hardware configuration..." because it doesn't recognize the LVM partition type.

For reference, this is the original output of fdisk -l: Device Boot Start End Blocks Id System /dev/hdc1 * 1 13 104391 83 Linux /dev/hdc2 14 24321 195254010 8e Linux LVM

• Removed var from LVM:
  • [root@tono Desktop]# /usr/sbin/lvremove /dev/VolGroup00/LogVol_var
  • Do you really want to remove active logical volume "LogVol_var"? [y/n]: y
  • Logical volume "LogVol_var" successfully removed

• Removed swap from LVM (only volume with higher PE range than var):
  • [root@tono Desktop]# /sbin/swapoff /dev/VolGroup00/LogVol_swap
  • [root@tono Desktop]# /usr/sbin/lvremove /dev/VolGroup00/LogVol_swap
  • Do you really want to remove active logical volume "LogVol_swap"? [y/n]: y
  • Logical volume "LogVol_swap" successfully removed

• Recreated swap (with new lowest available PE number):
  • [root@tono Desktop]# /usr/sbin/lvcreate VolGroup00^? -n LogVol^?_swap -L 1.94G
  • Rounding up size to full physical extent 1.97 GB
  • Logical volume "LogVol_swap" created
  • [root@tono Desktop]# /sbin/mkswap /dev/VolGroup00/LogVol_swap
  • Setting up swapspace version 1, size = 2113925 kB
  • [root@tono Desktop]# /sbin/swapon -va
  • swapon on /dev/VolGroup00/LogVol_swap
  • [root@tono Desktop]# cat /proc/swaps
  • Filename Type Size Used Priority
  • /dev/mapper/VolGroup00-LogVol_swap partition 2064376 0 -2

• Resized LVM less 40G:
  • [root@tono Desktop]# /usr/sbin/pvresize --setphysicalvolumesize 146.19G /dev/hdc2
  • Physical volume "/dev/hdc2" changed
  • 1 physical volume(s) resized / 0 physical volume(s) not resized

• fdisk reports 8225280 bytes per cylinder > resized less roughly [(38*1024*1024*1024)/8225280] cylinders:
  • [root@tono Desktop]# /sbin/fdisk /dev/hda2
  • Command (m for help): d 2
  • Command (m for help): n
    • 'primary' > First cylinder: '14' > Last cylinder: '19348'

• Added NTFS partition:
  • [root@tono Desktop]# /sbin/fdisk /dev/hda2
  • _New primary > Defaults are fine because it's the only remaining space on the disk > 't' to change type > '3' for new partition > '7' to change to NTFS

• Rebooted to verify everything was fine, then (IMPORTANT) saved the output of 'fdisk -l'
  • Added NTFS partition using remaining space:
  • Device Boot Start End Blocks Id System
  • /dev/hdc1 * 1 13 104391 83 Linux
  • /dev/hdc2 14 19348 155308387+ 8e Linux LVM
  • /dev/hdc3 19349 24321 39945622+ 7 HPFS/NTFS

• Using the same procedure to delete as above in the 'fdisk reports 8225280...' step (but without recreating this time!), deleted partitions 1 and 2. This hides the partitions from Windows.

• Rebooted and installed Windows on the third partition. It's important at this stage that what is now reported as free space is NOT touched. Declaring the third partition type NTFS in Linux makes this simple to visually double check in the Windows installer.

• After install is working, started from our Ubuntu Live CD, then restored the partitions in fdisk.
  • Command (m for help): n
  • primary > First cylinder: 14 > Last: 19348
  • Command (m for help): n
  • primary > First cylinder: 19349 > Lsat: 24321

• The partition table is out of order at this point. This is unavoidable because Windows was installed on what it saw as the first partition:
  • [root@tono ~]: /sbin/fdisk -l
  • Device Boot Start End Blocks Id System
  • /dev/hdc1 19349 24321 39945622+ 7 HPFS/NTFS
  • /dev/hdc2 * 1 13 104391 83 Linux
  • /dev/hdc3 14 19348 155308387+ 8e Linux LVM

then restored GRUB after adding an entry for windows(overwritten by the Windows MBR):

• • Edited [mount for /dev/hdc2]/grub/menu.lst, updating all references to the new partition scheme and adding:
    • title Microsoft Windows XP
    • rootnoverify (hd0,0)
    • makeactive
    • chainloader +1
  • root@ubuntu $: grub
  • grub > root (hd0,1)
  • grub > setup (hd0)

After this procedure, I was able to reboot and select either Windows or Linux.

1) Export from windows:

• From IIS Admin:
• From Certificates Snap-in (run->mmc.exe, add->certificates, computer store, local machine)

2) openssl pkcs12 -in c:\certs\yourcert.pfx -out c:\certs\cag.pem -nodes

No permission to read topic PHPIISHowTo - perhaps you need to log in?

Resetting your password

You can change your password using the VPN gateway or the LONI Webmail site

Using the change password page on LONI Webmail

• Go to http://password.loni.ucla.edu
• On the page you will see several fields, fill them out as follows:
  • Account – username (eg. Joe Bruin has the username jbruin )
  • Old password – your current windows password
  • New password – what you want your new password to be
  • Confirm new password – same as above.
• Click “OK”
• The password will be effective immediately for logging into Windows, accessing mail and using the VPN. It may take 5-10 minutes to propagate the changes to the unix services (e.g. inire, cerebellum, cerebro) or for linux and mac workstations.

-- JonathanTrout - 04 Oct 2007 -- DavidHasson - 04 Oct 2007 -- LinhNamVu - 25 Oct 2007

No permission to read topic PatchLink - perhaps you need to log in?

No permission to read topic PostfixSpamassassin - perhaps you need to log in?

No permission to read topic PowerUpOrder - perhaps you need to log in?

No permission to read topic PowerVault220S - perhaps you need to log in?

Polk Print Cluster

Network Information

• Polk printer Service 10.1.1.113
• DNS entry for polk.loni.ucla.edu is 10.1.1.113
• Cluster service will negotiate which server will be hosting the print service
• On failover, cluster service will switch to the other server.
• Shared Storage vmhba1:1:4:0 (through vmware)

Polk-rs1

• 10.1.1.114
• 192.168.10.5 (internal network interface used by the cluster service)
• [vmfs-local] prod.polk/prod.polk-rs1.vmdk

Polk-rs2

• 10.1.1.115
• 192.168.10.6 (internal network interface used by the cluster service)
• [vmfs-local] prod.polk-rs2/prod.polk-rs2_1.vmdk

Operation

• You can access(add new printers/change settings) the polk printers from any computer
• Administrative access is needed to add printers to it.

Access to Printers hosted the the Cluster Printing Service

• \\polk\Printers and Faxes

Add a Printer to the Cluster Printing Service

• goto \\polk\Printers and Faxes in explorer
• add printer there

-- LinhNamVu - 12 Mar 2008

No permission to read topic ProgramLocations - perhaps you need to log in?

No permission to read topic PropagatingConfigFiles - perhaps you need to log in?

LinhNamVu^? - 30 Jan 2007

Setting up RAID on a Solaris 10 6/06 Machine

Install OS

Insert disks (make sure you have the right architecture (sparc,x86)) Install OS (to boot into selected media use STOP+A, then use

• boot cdrom

More detailed instructions from sun. Also keep in mind how you want to setup your partitions/disks.

Partitioning

If you want to just mirror the same partitions as installed the OS: Copy partition layout to other disks with

• prtvtoc /dev/rdsk/c#t#d#s# | fmthard -s - /dev/rdsk/c#t#d#s#

If you want to create you're own partition setup then create it on one of the empty disks To format a disk use:

• format

Then make sure the other disks you want in the RAID to have the same partition layout as the one you have

• prtvtoc /dev/rdsk/c#t#d#s# | fmthard -s - /dev/rdsk/c#t#d#s#

Setup RAID

Choose a naming convention of metadevices for RAID

• dXY (X = disk number in raid) (Y = raid number)

Create devices

• metainit -f d10 1 1 /dev/dsk/c#t#d#s#
• metainit -f d20 1 1 /dev/dsk/c#t#d#s#
• metainit -f d30 1 1 /dev/dsk/c#t#d#s#
• etc...

Create RAID device.This command will create the initial metadevice for our raid

• metainit -f d0 d10

Attach the first device

• metattach -f d0 d10 (note: there is one less 'a' in metattach)

The created metadevices will be located in /dev/md/dsk/DEVICE_NAME_HERE

If you are creating a metadevice for the partition that the systems boots off of then you'll need to edit /etc/vfstab and reboot the system and boot into the new metadevice settings. If not you can ignore this step.

• vi /etc/vfstab
• lockfs -fa
• init 6

Attach remaining devices

• metattach -f d0 d20
• metattach -f d0 d30
• etc...

If you want to be able to boot off the other disk if the first one fails then

• installboot /usr/platform/`uname -i`/lib/fs/ufs/bootblk  /dev/rdsk/c0t1d0s0
• setenv boot-device disk disk1
• nvstore
• boot disk1

Maintanence

You can see the status of the disks using metastat . If you want to see the progress of the syncing then use metastat -c . In the event one of the drives goes missing, that command will show you. Once you re-insert the new drive you can create it with the above steps and attach it. Running metastat will show that the drive needs maintanence and running metasync will tell it that it can start syncing again. metastat -c will now show the sync progress.

1) Install Rails

FreeBSD Installation

• Install Ports:
  • www/apache22
  • lang/php5
  • lang/ruby19
  • devel/ruby19-gems
• Install Gems:
  • gem install rails
• Install Passenger (this needed CVS last time I checked, though)
  • gem install passenger-x.x.x.gem
  • passenger-install-apache2-module

Setup VirtualHosts and Apache

Make a rails project

• cd /www/hosts/loni.funtaff.com/docs
• rails loni
• config/database.yml
• mv public/index.html public/index.html.old

Authentication (Railscast)

• script/plugin source http://svn.techno-weenie.net/projects/plugins/ (only needs to be done once per machine)
• script/plugin install restful_authentication
• script/plugin install http://elitists.textdriven.com/svn/plugins/acts_as_state_machine/trunk/
• script/generate authenticated user session --include-activation --stateful
• rake db:migrate
• script/generate controller home
• config/routes.rb
  • map.home '', :controller => 'home', :action => 'index'
  • map.resources :users, :member => { :suspend => :put, :unsuspend => :put, :purge => :delete }
  • map.activate '/activate/:activation_code', :controller => 'users', :action => 'activate', :activation_code => nil
  • map.signup '/signup', :controller => 'users', :action => 'new'
  • map.login '/login', :controller => 'session', :action => 'new'
  • map.logout '/logout', :controller => 'session', :action => 'destroy'
• controllers/application.rb:class ApplicationController
  • include AuthenticatedSystem
• views/home/index.html.erb

Test and create demo user

• script/server

-- DavidHasson - 14 May 2008

No permission to read topic RebuildRAID - perhaps you need to log in?

Red Hat Systems

2.1AS-i386

• aragon-lvs1
• aragon-lvs2
• aragon-rs1
• aragon-rs2
• dessi-dev
• dessi-lvs1
• dessi-lvs2
• dessi-test
• viola-dev
• viola-rs1
• viola-rs2
• viola-test

3AS-i386

• netserv1

3AS-x86_64

• cerebellum

3WS-i386

• booch
• netulla

3WS-x86_64

4AS-i386

• corbu
• sevuis
• nash

4AS-x86_64

• ruscha

4ES-i386

• dessi-lvs3
• dessi-lvs4
• viola-local

4ES-x86_64

• aragon-rs3
• aragon-rs4

4WS-i386

• 10.1.2.44
• accardi
• doig
• heckle
• koba
• renior
• helm
• johnson
• kobayashi

4WS-x86_64

No permission to read topic RemoteManagement - perhaps you need to log in?

No permission to read topic RenewTomcatHTTPSCertificate - perhaps you need to log in?

No permission to read topic SPCommonActions - perhaps you need to log in?

No permission to read topic SecondaryDNS - perhaps you need to log in?

LinhNamVu^? - 07 Dec 2006

Securing an Installationg of SQL Server 2005

To determine the SQL version on SQL Server 2005 open SQL Server Management Studio, connect to the database, then run the following query

SELECT  SERVERPROPERTY('productversion'), SERVERPROPERTY ('productlevel'), SERVERPROPERTY ('edition')

Column 1. The product version (for example, "9.00.1399.06").
Column 2. The product level (for example, "RTM").
Column 3. The edition (for example, "Enterprise Edition").

Current versions can be found here

Taken from http://www.databasejournal.com/features/mssql/article.php/3461471

It is worth pointing out that the new version of SQL Server has been designed with the "secure by default" principle in mind, resulting in a system with optimum, from the security standpoint, settings. The typical setup avoids installing or activating non-essential components and features which can expose the server and its data to potential attacks. This applies, for example, to SQL Server Agent, Full-Text Search, and Data Transformation Services (all set to manual startup), Analysis, Reporting, and Notification Services, SQL Browser, Service Broker Network Connectivity, Database Mirroring, SQLMail, or SQL Debugging.

Taken from http://www.sqlsecurity.com/FAQs/SQLSecurityChecklist/tabid/57/Default.aspx

How to secure SQL Server

• Keep up to date on SQL Server service packs and patches
• Audit SQL Server accounts for weak passwords
• Restict access to the SQL Server to only trusted clients
• Use Windows Only authentication where possible
• Store SQL Server backup files in a secure location and encrypted
• Disable all netlibs if the SQL Server is local-only
• Regularly scan the installation with Microsoft's Baseline Security Analyzer

Use Windows Only authentication mode for security if possible By using integrated security, you can greatly simplify administration by relying on the OS security and saving yourself from maintaining two separate security models. This also keeps passwords out of connection strings.

Take the time to audit SQL logins for null or weak passwords

Use the following code to check for null passwords:

Use master Select name, Password from syslogins where password is null order by name

There are a multitude of free and commercial tools to check for weak passwords. SQLPing2 is free and can be used to check for weak and null passwords.

Frequently check group and role memberships While the SQL Server security model has many enhancements, it also adds the extra layer of permissions that we must monitor to make sure no one has been given more access than they need or that they’ve already circumvented security to elevate themselves. There's also the spectre of user's who have changed roles within the company but the SQL Server permissions structure has not been adjusted. This goes back to assigning object access to groups and not individuals.

Physically secure the SQL Server Lock it behind a door and lock away the key while you’re at it. Someone sitting in front of the server will always find a way.

Rewrite applications to use more user-defined stored procedures and views This minimizes the need to give direct access to tables. It gives the developer more control over how data can be accessed.

Enable logging of all user login events You can also do this via script but using the following code:

xp_instance_regwrite N'HKEY_LOCAL_MACHINE', N'SOFTWARE\Microsoft\MSSQLServer\MSSQLServer',N'AuditLevel', REG_DWORD,3

Check master..Sp_password for trojan code Compare your production scripts to the default script on a fresh installation and keep that code handy.

Check master..Sp_helpstartup for trojan procedures Make sure no one has placed a backdoor here. Use Sp_unmakestartup to remove any rogue procedures.

Disable SQL Mail capability unless absolutely necessary Leaving it open gives a potential attacker another means of delivering potential trojans, viruses, or simply launching a particularly nasty denial of service attack. By itself it is fairly harmless but it can be used to help an attacker.

Remove the Guest user from databases to keep unauthorized users out This is the default but vigilant in case some dbo gets loose with the access controls. The exception to this is the master and tempdb databases as the guest account is required

Restrict to sysadmins-only access to dangerous stored procedures and extended stored procedures

There are quite a few of them, and this could take some time. Be careful not to do this on a production server first. Test on a development machine so you don’t break any functionality. Below is a list of the ones we recommend you assess:

sp_sdidebug xp_availablemedia xp_cmdshell xp_deletemail xp_dirtree xp_dropwebtask xp_dsninfo xp_enumdsn xp_enumerrorlogs xp_enumgroups xp_enumqueuedtasks xp_eventlog xp_findnextmsg xp_fixeddrives xp_getfiledetails xp_getnetname xp_grantlogin xp_logevent xp_loginconfig xp_logininfo xp_makewebtask xp_msver xp_regread xp_perfend xp_perfmonitor xp_perfsample xp_perfstart xp_readerrorlog xp_readmail xp_revokelogin xp_runweb

Make sure all SQL Server data and system files are installed on NTFS partitions If someone should gain access to the OS, make sure that the necessary permissions are in place to prevent a catastrophe.

Use a low-privilege user account for SQL Server service rather than LocalSystem^? or Administrator This account should only have minimal privileges (a local user is best) and should help contain an attack to the server in case of compromise. Notice that when using Enterprise Manager or SQL Server Configuration Manager (SQL 2005) to make this change, the ACLs on files, the registry, and user rights are done for you automatically.

Secure the “sa” account with a strong password This assumes you are using the SQL Server and Windows security mode. If possible, use the Windows Only mode and don't worry about people brute-forcing your 'sa' accounts. Of course, even so you'll want to set a strong password in case someone changes modes on you.

Choose only the network libraries you absolutely require Better yet, if the SQL Server is local-only then why not disable all network libraries and use shared memory to access the SQL Server? Just use the name '(local)' as the server name. If your SQL Server requires connectivity from other hosts, use the TCP/IP netlib and then determine if SSL is needed.

Make sure the latest OS and SQL Server Service Packs/Hot-Fixes are applied

What different network protocol libraries should I use?

If the database is local-only then disable ALL netlibs and use shared memory to access the SQL Server. If you must connect from remote clients, use TCP/IP and enable encryption (non-trusted subnets) . Only use the alternative netlibs (NwLink^?, Names Pipes, Banyan, etc) if need be in those integration scenarios.

SQL Server 2000 includes the new Super Sockets net-lib which can do SSL over any of the protocols. It is highly recommended you make use of this in lieu of the multi-protocol encryption. Remember that a server certificate must be installed before the encryption can occur. The certificate must match the DNS name of the server and be issues by a certificate authority that the client trusts.

Yet another option if you are using Windows 2000 is IPsec. IPsec will allow you to encrypt ALL traffic between the client and server. This is a good solution when you have complete control over the server and the clients can take the time to learn IPsec policy deployment.

Additional Links

http://msdn2.microsoft.com/en-us/library/ms161948.aspx

Serial Numbers

VMWare Server - VMWare Server is free, but you need to register and keep track of serial numbers regardless. The list is maintained on the Wiki.

VMWare Trials^? - We're currently using trials to run our test VM network. Status of trial registrations is maintained on the Wiki.

-- DavidHasson - 14 Oct 2007

No permission to read topic SetUpNIS - perhaps you need to log in?

No permission to read topic SetupCerebellum - perhaps you need to log in?

How to set up a Printer in Windows XP

Under Printers and Faxes in the Control Panel:

1. ) select "Add a printer"
2. ) Click on "Next."
3. ) Select "A network printer...."
4. ) Select "Connect to this printer..." and type
   
   \\polk\{printer name}
   
   where {printer name} can be toad, grenoulle, mantella, etc.
   
5. ) Select whether this printer is to be the default printer with {Yes | No}.
6. ) Click on "Next."
7. ) Select "Finish."

If the username and password prompt comes up, use the following format to enter your credentials:

LONI\{username}
{password}

where {username} is your LONI username and {password} is your LONI password.

How to set up a Printer in OS X

Under Print & Fax panel in the System Preferences:

1. ) Click on the '+' sign to add a LONI printer
2. ) Select the protocol 'Line Printer Daemon - LPD'
3. ) The address is 'polk.loni.ucla.edu'
4. ) The queue is the printer name. e.g. toad, grenoulle, mantella, etc.
5. ) Select "Add"

No permission to read topic SetupSamba - perhaps you need to log in?

No permission to read topic SetupUnixAccounts - perhaps you need to log in?

No permission to read topic SetupUserAccounts - perhaps you need to log in?

These are the steps to tunnel an SSH connection

Apple Mac OS X

• Install X11
• Launch X11
• type "ssh ${user}@${host} -Y", where ${user} is your username, and ${host} is the hostname of the server you are trying to log into

Windows

• First method- redirect output by setting the DISPLAY variable of your console to your local IP address and having an X-server accept it
  • Start X-server (X-Win32)
  • Find your IP address
    • Open the command prompt
    • Type "ipconfig"
  • Set the DISPLAY environment variable
    • for bash, "set DISPLAY ${IP}:0"
    • for csh
• Second method- tunnel X session over SSH Secure Shell
  • Start X-server (X-Win32)
  • In SSH Secure Shell, click on "Edit" menu item
  • Click on "Settings"
  • Select "Tunneling" option in the left side of the window
  • Verify that the "Tunnel X11 connections" option is selected
  • Click "OK"
  • restart all SSH connections

No permission to read topic SharedGroupCalendars - perhaps you need to log in?

This probably won't come up too often, but it's good reference. Arash requested gcc4 on cerebro-cc. Unfortunately, the Sun Freeware edition does not have support for the "-m64" flag compiled in. There are quite a few caveats in compiling from source (note that this applies to Solaris 10 -- I don't know if they're going to replace gcc3 or include it alongside in future distributions):

• This step might be possible using Sun Studio's 'dmake', but just to be safe: download, compile, and install GNU make.
• Download the source for binutils. As of 2.17, it has issues with the amd64 library and ld will spew errors during the gcc build. This patch (loosely based on a fix from OpenSolaris^?) fixed for me:

Index: bfd/elf.c
===================================================================
RCS file: /cvs/src/src/bfd/elf.c,v
retrieving revision 1.347
diff -u -2 -r1.347 elf.c
--- bfd/elf.c   23 Jun 2006 02:58:00 -0000   1.347
+++ bfd/elf.c   28 Jun 2006 00:32:08 -0000
@@ -2104,4 +2104,9 @@
       break;
 
+    case SHT_GNU_dof:
+    case SHT_GNU_signature:
+    case SHT_GNU_syminfo:
+      return TRUE;
+
     case SHT_GNU_verdef:
       elf_dynverdef (abfd) = shindex;
Index: include/elf/common.h
===================================================================
RCS file: /cvs/src/src/include/elf/common.h,v
retrieving revision 1.77
diff -u -2 -r1.77 common.h
--- include/elf/common.h   17 Feb 2006 14:36:26 -0000   1.77
+++ include/elf/common.h   28 Jun 2006 00:41:58 -0000
@@ -341,14 +341,20 @@
 #define SHT_GNU_LIBLIST   0x6ffffff7   /* List of prelink dependencies */
 
-/* The next three section types are defined by Solaris, and are named
+/* The next six section types are defined by Solaris, and are named
    SHT_SUNW*.  We use them in GNU code, so we also define SHT_GNU*
    versions.  */
-#define SHT_SUNW_verdef   0x6ffffffd   /* Versions defined by file */
-#define SHT_SUNW_verneed 0x6ffffffe   /* Versions needed by file */
-#define SHT_SUNW_versym   0x6fffffff   /* Symbol versions */
-
-#define SHT_GNU_verdef   SHT_SUNW_verdef
-#define SHT_GNU_verneed   SHT_SUNW_verneed
-#define SHT_GNU_versym   SHT_SUNW_versym
+#define SHT_SUNW_dof       0x6ffffff4   /* Solaris DTrace Object Format */
+#define SHT_SUNW_signature  0x6ffffff6   /* Solaris Cryptographic Framework: Digital Signature */
+#define SHT_SUNW_syminfo    0x6ffffffc   /* Symbol information */
+#define SHT_SUNW_verdef       0x6ffffffd   /* Versions defined by file */
+#define SHT_SUNW_verneed    0x6ffffffe   /* Versions needed by file */
+#define SHT_SUNW_versym       0x6fffffff   /* Symbol versions */
+
+#define SHT_GNU_dof       SHT_SUNW_dof
+#define SHT_GNU_signature   SHT_SUNW_signature
+#define SHT_GNU_syminfo       SHT_SUNW_syminfo
+#define SHT_GNU_verdef       SHT_SUNW_verdef
+#define SHT_GNU_verneed       SHT_SUNW_verneed
+#define SHT_GNU_versym       SHT_SUNW_versym
 
 #define SHT_LOPROC   0x70000000   /* Processor-specific semantics, lo */
Index: ld/configure.tgt
===================================================================
RCS file: /cvs/src/src/ld/configure.tgt,v
retrieving revision 1.189
diff -u -2 -r1.189 configure.tgt
--- ld/configure.tgt   23 Jun 2006 18:11:47 -0000   1.189
+++ ld/configure.tgt   28 Jun 2006 00:42:40 -0000
@@ -164,10 +164,10 @@
 i[3-7]86-*-solaris2*)   targ_emul=elf_i386_ldso
                         targ_extra_emuls="elf_i386 elf_x86_64"
+         targ_extra_libpath=$targ_extra_emuls
                         ;;
 i[3-7]86-*-unixware)   targ_emul=elf_i386 ;;
 i[3-7]86-*-solaris*)   targ_emul=elf_i386_ldso
                         targ_extra_emuls="elf_i386"
+         targ_extra_libpath=$targ_extra_emuls
                         ;;
 i[3-7]86-*-netbsdelf* | Index: ld/emulparams/elf_x86_64.sh
===================================================================
RCS file: /cvs/src/src/ld/emulparams/elf_x86_64.sh,v
retrieving revision 1.17
diff -u -2 -r1.17 elf_x86_64.sh
--- ld/emulparams/elf_x86_64.sh   30 May 2006 16:45:32 -0000   1.17
+++ ld/emulparams/elf_x86_64.sh   28 Jun 2006 00:42:47 -0000
@@ -23,5 +23,5 @@
 fi
 
-# Linux modify the default library search path to first include
+# Linux/solaris modify the default library search path to first include
 # a 64-bit specific directory.
 case "$target" in
@@ -31,3 +31,7 @@
     esac
     ;;
+  *-*-solaris2*) 
+      LIBPATH_SUFFIX=/amd64
+      ELF_INTERPRETER_NAME=\"/lib/amd64/ld.so.1\"
+    ;;
 esac

There is a good chance you'll need to inspect and patch by hand beyond 2.17 (if necessary at all).

• (Still binutils:) ./configure --disable-nls, make, and make install.
• The gcc documentation claims bison isn't necessary, but it will error out constantly without it. Download, compile, and install.
• Download gcc4. Solaris doesn't use GNU tar, so you'll get an error during the untar process. I just untarred on my own machine and scp'd the directory.
• I used the following line to configure (this is more for posterity and less of a note): ../src/configure --prefix=/usr/local/gcc4 --with-local-prefix=/usr/local --with-gnu-as --with-as=/usr/local/i386-pc-solaris2.10/bin/as --with-gnu-ld --with-ld=/usr/local/i386-pc-solaris2.10/bin/ld --enable-threads=posix --enable-shared --enable-multilib --disable-nls --with-included-gettext --with-system-zlib --enable-languages=c,c++
• Build, go to lunch, and install.
• Path /usr/local/gcc4/bin and test.

SOPHOS ENTERPRISE CONSOLE WIKI PAGE

Note: There are two parts to Sophos Enterprise Console: The Management Components (Management Server, Management Console, EM Library) & The Database. The easiest way to proceed is to install both parts on the same computer. The parts may be installed on separate machines, but, if the hosts are on different subnets, the two components will not be able to communicate with one another. This guide assumes both the Management Components and the Database are on the same machine.

1. Install the Database and Management Components

• Run the appropriate Enterprise Console Installer
• When asked to populate the database, if host machine is running a 64-bit OS, you must select to populate the database later. Scripts for populating the database will be created in the installation directory (usually C:\Program Files\Sophos\Enterprise Console\DB)
• Specify the SQL Server host and the instance onto which the database will be installed (HOST\INSTANCE)
  • Note: The default instance name is MSSQLSERVER
  • Note: For simplicity, you should create a new instance called SOPHOS
• A database called SOPHOS3 should be created on the SQL Server.

2. Populate the database manually

• After the Database and Management Server have been installed, open the command prompt and navigate to C:\Program Files\Sophos\Enterprise Console\DB
• For 64-bit operating systems, download the file “install.bat” from the Sophos website and place it in this folder.
• Run install.bat
  • If the instance you chose is named SOPHOS, no parameters need to be entered. Otherwise use install.bat [INSTANCE] where [INSTANCE] is the name of the instance the database is installed on. If it’s the host is a domain controller, you must also specify the domain: install.bat [INSTANCE] [DOMAIN].
• This creates the stored procedures and tables for the SOPHOS3 database.

3. Open Enterprise Console (it should automatically connect to the database) for Configuration

• Download the “How to Use Enterprise Console” PDF and follow the instructions to:
  • Create and configure a Library for packages
  • Create groups and synch them with active directory
  • Start Managing Computers
• The primary update source should be: \\HOSTNAME\InterChk\ESXP
  • The account used to get updates: LONI\antivirus (Ask a sysadmin for the PW)
• The secondary update source should be directly from Sophos
  • Contact BOL or Sophos for UCLA’s Username and Passwords to obtain updates

4. Modify Domain Group Policy so Enterprise Console can properly communicate with clients

• The following exceptions need to be added to the LONI Default Domain Policy’s Firewall Rules (Computer Configuration > Administrative Templates > Network > Network Connections > Windows Firewall ) to allow Sophos Enterprise Console to push out new installations:

Note: In the example below, 128.97.134.20 must be replaced by the specific IP address of the machine on which Enterprise Console is installed.

DOMAIN PROFILE

Program Exceptions:

Allow Local Program Exceptions: Enabled

Define inbound program exceptions: C:\Program Files\Sophos\Enterprise Console\Remote Management System:enabled:Sophos

Allow Inbound File and Printer Sharing Exception (from): 128.97.134.20

Allow Inbound Remote Administration Exception (from): 128.97.134.20

Port Exceptions:

Allow Local Port Exceptions: Enabled

Define inbound port exceptions:

8192:TCP:128.97.134.20:enabled:sophos

8192:UDP:128.97.134.20:enabled:sophos

8193:TCP:128.97.134.20:enabled:sophos

8193:UDP:128.97.134.20:enabled:sophos

8194:TCP:128.97.134.20:enabled:sophos

8194:UDP:128.97.134.20:enabled:sophos

STANDARD PROFILE

Allow Inbound Remote Administration Exception (from): 128.97.134.20

Allow Inbound File and Printer Sharing Exception (from): 128.97.134.20

Port Exceptions:

8192:TCP:128.97.134.20:enabled:sophos

8192:UDP:128.97.134.20:enabled:sophos

8193:TCP:128.97.134.20:enabled:sophos

8193:UDP:128.97.134.20:enabled:sophos

8194:TCP:128.97.134.20:enabled:sophos

8194:UDP:128.97.134.20:enabled:sophos

• The following must be added to the LONI Computer GPO’s Firewall Rules to allow Sophos Enterprise Console to push out new installations:

DOMAIN PROFILE

Allow Inbound Remote Administration Exception (from): 128.97.134.20

Allow Inbound File and Printer Sharing Exception (from): 128.97.134.20

STANDARD PROFILE

Allow Inbound Remote Administration Exception (from): 128.97.134.20

Allow Inbound File and Printer Sharing Exception (from): 128.97.134.20

• Clients may need to be restarted for this change in GPO to take effect. Once these exceptions in the windows firewall have been made, the client should be able to contact the Enterprise Console Server and vice versa.

5. Start Managing Computers

• Open Enterprise Console
• Select the computer(s) you wish to manage, click PROTECT in the top toolbar
• A Protect Wizard will come up. Enter domain admin credentials to allow the installation to go through. If successful, the computer’s icon should no longer be “grayed out”.
  • Note: If the installation fails, you can double click the computer’s icon and get more details about why it failed. A “0000002e” error code is a generic error which is most likely caused by a blockage due to a network and/or windows firewall.

• FOR SERVERS: If you can access \\GEER\interchk\ESXP but cannot run setup.exe (e.g. “Path not valid…”)
  • Copy the ESXP folder to local C: * Run the setup.exe from there to install Sophos
  • Edit the C:\Program Files\Sophos\Autoupdate\iconn.cfg so that primary server points to \\GEER\interchk\ESXP

Installing Sophos on Read Only Domain Controllers

1. Create IPSec Tunnel from RODC to Sophos Server

2. Allow IPSec Tunnel through the Firewall

Note: Steps 3 and 4 only necessary if you get Sophos Error 3051 during installation.

3. Create LONI SophosSAU^?0 account for each RODC

4. For 64bit OS, add to [HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\Sophos\AutoUpdate\Service], otherwise add the following registry keys:

• [HKEY_LOCAL_MACHINE\SOFTWARE\Sophos\AutoUpdate\Service]
  • "Download User"="SophosUpdate”
• [HKEY_LOCAL_MACHINE\SOFTWARE\Sophos\AutoUpdate\Service]
  • "Download Password"="[enter the password]"
• [HKEY_LOCAL_MACHINE\SOFTWARE\Sophos\AutoUpdate\Service]
  • "UserPreset"=dword:00000001

MarcSycip - 17 Sep 2009

Configuring Spamassassin to Pick Up More Spam

• Subscribe SpamAssassin to various rule channels using the saupdate command
• Specifically MIT's OpenProtect^? compilation of rules along with the official ones
  • Grabs updates from updates.spamassassin.org and saupdates.openprotect.com
  • sa-update --gpgkey D1C035168C1EBC08464946DA258CDB3ABDE9DC10 --channel saupdates.openprotect.com --channel updates.spamassassin.org
    • Necessary PGP Keys needed to be downloaded and imported to get updates from openprotect
    • sa-update --import key_file

• Created a bash script in /usr/local/bin/saupdate to run
• sa-update --gpgkey D1C035168C1EBC08464946DA258CDB3ABDE9DC10 --channel saupdates.openprotect.com --channel updates.spamassassin.org

• Cron job to execute that script every day
• 0 0 * * * /usr/local/bin/saupdate

References http://saupdates.openprotect.com/

LinhNamVu^? - 09 Oct 2007

No permission to read topic SpamassassinConfig - perhaps you need to log in?

Special Lists for the Cranium's Compute Nodes

This page describes the special lists used by the Cranium's cluster monitoring script, agentSmith.

Hugo Hernandez - 15 Jun 2009

How to use the Star-P Software to Parallel Process on the Cranium Cluster

Star-P software allows users performing scientific, engineering or analytical computation on array or matrix-based data to use parallel architectures such as multi-core workstations, multi-processor systems, distributed memory clusters and/or utility/cloud-based environments.

The software delivers users of MatLab and NumPy (Numeric Python) the ability to:

• Transform serial applications for parallel deployment
• Develop and deploy new parallel applications and algorithms

Visit the official Star-P website for more information.

To use the Star-P software: Follow the next steps. As a note, users are required to set public/private SSH authentication keys on the Cranium cluster to use the Star-P client. This allows you to SSH into any LONI machine without the need for passwords. Use this link to configure your authentication keys or send an email to support@loni.ucla.edu for assistance.

• As the software is currently in evaluation, you need to login into one of the Cranium Cluster's compute nodes we configured to run Star-P. This machine is not accessible from all computers in the LONI network, so you need to login into a SSH proxy machine before accessing this compute node.

• SSH into ssh.loni.ucla.edu,

• Now, ssh into cerebro-24-114.data.cluster.loni.ucla.edu,

• Run the command: starpClient (if you don't have /usr/local/bin in your path, then run /usr/local/bin/starpClient). You will have the following output:

• Once, Star-P reserves the needed slots in the cluster to be used during your session, a MatLab prompt will be available for you to run your job(s):

• By typing the qstat command, you can see the number of slots reserved by Star-P to run your parallel MatLab jobs in the Cranium cluster, in this case, 10 slots as shown in your Star-P client configuration file in your home directory.

• Use the MatLab prompt to run your jobs as shown in the following example:

StreamingCompression

23 Jun 2008 - 21:50 - NEW

DavidHasson

SuN3510

24 Jul 2008 - 17:16 - r1.2

JonathanTrout

Subversion

20 Feb 2008 - 02:04 - r1.6

JonathanTrout

SummaryNo1

14 Aug 2006 - 19:36 - r1.7

JordanMendler

SummaryNo10

13 Nov 2006 - 19:41 - r1.5

BrianChin

SummaryNo11

02 Feb 2007 - 02:26 - r1.4

RicoMagsipoc

SummaryNo12

13 Feb 2007 - 19:04 - r1.5

RicoMagsipoc

SummaryNo13

20 Feb 2007 - 22:55 - r1.6

JonathanTrout

SummaryNo14

05 Mar 2007 - 21:38 - r1.5

LinhNamVu

SummaryNo15

06 Mar 2007 - 07:58 - r1.2

ShanrenZhou

SummaryNo16

18 Jul 2007 - 19:43 - r1.5

LinhNamVu

SummaryNo17

18 Jul 2007 - 20:52 - r1.4

JonathanPierce

SummaryNo18

01 Aug 2007 - 21:58 - r1.5

LinhNamVu

SummaryNo19

07 Aug 2007 - 18:54 - r1.2

JonathanTrout

SummaryNo2

21 Aug 2006 - 18:05 - r1.6

JonathanPierce

SummaryNo20

07 Aug 2007 - 21:04 - NEW

HugoHernandez

SummaryNo3

05 Sep 2006 - 06:02 - r1.6

RicoMagsipoc

SummaryNo4

05 Sep 2006 - 20:21 - r1.4

HugoHernandez

SummaryNo5

08 Sep 2006 - 19:47 - r1.5

JonathanTrout

SummaryNo6

18 Sep 2006 - 17:57 - r1.3

BrianChin

SummaryNo7

19 Sep 2006 - 18:39 - r1.3

BrianChin

SummaryNo8

28 Sep 2006 - 23:57 - r1.6

BrianChin

SummaryNo9

04 Nov 2006 - 01:19 - r1.2

BrianChin

SunCaseDB

16 Feb 2007 - 19:06 - NEW

JonathanPierce

SyncCluster

07 Aug 2007 - 22:48 - r1.2

HugoHernandez

SynchronizingCerebroCluster

29 Jan 2007 - 20:49 - NEW

HugoHernandez

SysAdminHome

12 Oct 2009 - 21:16 - r1.14

DmitriyVinogradov

SysAdminHowtos

27 Jul 2009 - 17:18 - r1.8

MarcSycip

SysAdminLinks

12 Feb 2009 - 19:17 - r1.2

DavidHasson

SystemAccounts

17 May 2006 - 18:24 - r1.4

JonathanTrout

SystemSetup

26 Oct 2009 - 18:27 - r1.23

DavidHasson

SystemSetup_Mac

14 Oct 2008 - 21:04 - r1.8

JonathanPierce

SystemSetup_MacLeopard

24 Sep 2009 - 20:54 - r1.9

MarcSycip

SystemSetup_RedHatClient

27 Oct 2009 - 17:57 - r1.27

MarcSycip

SystemSetup_Ubuntu

15 Jan 2008 - 01:48 - NEW

JonathanPierce

SystemSetup_WinXP

24 Jul 2009 - 21:54 - r1.9

MarcSycip

SystemSetup_WinXP_RIS

28 Feb 2008 - 20:17 - r1.2

RicoMagsipoc

TestEnvHome

28 Feb 2008 - 20:50 - r1.3

RicoMagsipoc

UbuntuWriteup

11 Jan 2007 - 00:51 - r1.4

JonathanPierce

UpdatingSPandBIOS

18 Aug 2006 - 21:27 - NEW

HugoHernandez

UpdatingV20zSPandBIOS

13 Oct 2006 - 19:30 - r1.6

HugoHernandez

UsingFTP

19 Nov 2008 - 00:10 - r1.4

JonathanPierce

VLANs

28 Feb 2008 - 20:46 - r1.2

RicoMagsipoc

VMWareSerials

28 Feb 2008 - 20:46 - r1.2

RicoMagsipoc

VMwareServer

09 Oct 2007 - 00:29 - NEW

JonathanTrout

VNCBanksy

27 Oct 2006 - 00:21 - NEW

JonathanPierce

VizFAQ

28 Mar 2007 - 20:49 - r1.4

JonathanPierce

VncBanksy

07 Nov 2006 - 01:14 - r1.2

JonathanPierce

WeatherBoy

28 Feb 2008 - 20:53 - r1.2

RicoMagsipoc

WebChanges

16 Aug 2001 - 19:58 - NEW

PeterThoeny

WebHome

12 Oct 2009 - 21:14 - r1.144

DmitriyVinogradov

WebIndex

24 Nov 2001 - 11:41 - r1.2

PeterThoeny