INIT failed status and Problem communicating with BSMP on IBM Blade Chassis

I have a fully loaded Blade Chassis H (ie 14 Blades). I have the following symptoms similar to what was written in IBM Site INIT failed status - IBM BladeCenter HS22, HS22V

   

0X0021600 / Symptom:

BladeCenter HS22 or HS22V status is reported as Init Failed and the following error is posted in the Advanced Management Module (AMM) event log:

0x00216000 Problem communicating with BSMP

The chassis has 14 blades installed.

Workaround:

Remove one blade from the chassis.

Solution:
Install AMM firmware v3.54L (BPET54L) or later.

The file is available by selecting the appropriate machine type on the 'Software and device drivers - IBM BladeCenter' web page, at the following URL:

IBM  Blade Chassis Product View

Patching

Select the correct Blade Chassis Product. Since I am using Blade Chassis H, my fixes were

ibm_fw_amm_bpet62o_anyos_noarch
(IBM BladeCenter Advanced Management Module Firmware Update)

The patching will require restart of the AMM. Restarting the Advanced Management Module will disconnect your current session. 

CUDA Toolkits, Webinars and Downloads

Useful Information for those who are interested in Nvidia GP-GPU

1. GPU Computing Webinars 
2. CUDA Downloads 
3. NVIDIA GPUDirect
4. NVIDIA GPU Computing Documentation
5. CUDA Tools & Ecosystem
6. CUDA GPUs
7. CUDA Newsletter

Listing of Hardware Address of Network Cards

To quickly list Hardware Address of all your Network Cards in a nice listings, you can just use a simple commands

# ifconfig -a |grep HWaddr

eth0      Link encap:Ethernet  HWaddr 00:00:00:00:00:1a
eth1      Link encap:Ethernet  HWaddr 00:00:00:00:00:1b
eth2      Link encap:Ethernet  HWaddr 00:00:00:00:00:1c

"Device eth0 does not seem to be present" on cloned CentOS VM

After cloning the a CentOS 6 Virtual Machine from a VMware VM Template. On powering, I encountered the error "Device eth0 does not seem to be present".

Basically, in a nutshell, the OS remembers the old NIC of the template and new NIC provided after the clone are assigns as though they were new NICs. To solve the problem, remove the "SUBSYSTEM" entry that contains the old entry and change the eth1 to eth0 accordingly.

# vim /etc/udev/rules.d/70-persistent-net.rules

# PCI device 0x15ad:0x07b0 (vmxnet3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", 
ATTR{address}=="00:50:56:ae:00:1a", 
ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"

# PCI device 0x15ad:0x07b0 (vmxnet3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", 
ATTR{address}=="00:50:56:ae:00:1b", 
ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"

# PCI device 0x15ad:0x07b0 (vmxnet3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", 
ATTR{address}=="00:50:56:ae:00:1b", 
ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"

Remember to modify the /etc/sysconfig/network-scripts/ifcfg-eth0 to reflect the correct changes.

Once you are done, reload the udev configuration. See Updating the udev configuration on CentOS

There is a very good and complete information written on this whole issue.
Cloned Red Hat/CentOS/Scientific Linux Virtual Machines and “Device eth0 does not seem to be present” Message

No module named numpy when executing alpspython on CentOS 5

I was using testing out alps-2.0.2 on CentOS 5.4 .When I typed the following:

$ /usr/local/alps2.0.2/bin/alpspython

$ import pyalps

I got this error instead

Traceback (most recent call last):

  File "", line 1, in ?
  File "/usr/local/alps-2.0.2/lib/pyalps/__init__.py", line 37, in ?
    from dataset import *
  File "/usr/local/alps-2.0.2/lib/pyalps/dataset.py", line 29, in ?
    import numpy as np
ImportError: No module named numpy

The resolution was easier than I thought, just do a

# yum install python-numpy

The respective libraries will be installed

Using iptables to allow compute nodes to access public network

Have written a tutorial on Using iptables to allow compute nodes to access public network on LinuxCluster. Do read it.

Error when directly upgrading hdf5 1.6.10 directly on CentOS 5.4

On my CentOS 5.4 Linux Server

I was updating hdf5-1.6.10 to the latest version hdf5-1.8.7-1.el5.rf, however I have a dependency issue. I have got an error

R-hdf5-1.6.9-1.el5.x86_64 from installed has depsolving problems
  --> Missing Dependency: libhdf5.so.0()(64bit) is needed by package 
R-hdf5-1.6.9-1.el5.x86_64 (installed)
Error: Missing Dependency: libhdf5.so.0()(64bit) is needed by package 
R-hdf5-1.6.9-1.el5.x86_64 (installed)
 You could try using --skip-broken to work around the problem
 You could try running: package-cleanup --problems
                        package-cleanup --dupes
                        rpm -Va --nofiles --nodigest

This is due to the mismatched version of hdf5.x86_64 (version 1.6.10-1.el5) and R-hdf5.x86_64 (1.6.9-1.el5).

To solve this is pretty straightforward, you have to remove the installed hdf5 and R-hdf5 totally and reinstall

# yum remove hdf5
# yum install hdf5 hdf5-devel

Intel one-stop shop resource for Visual Computing

Do take a look at Intel Visual Computing Resource.  You can get free downloads of Intel® Graphics Performance Analyzers (Intel GPA) 2012, the Intel® Media SDK, and the Intel® OpenCL* SDK, along with access to case studies, white papers, community forums, and more...... 

Intel® Graphics Performance Analyzers (Intel® GPA) are an easy-to-use suite of optimization tools for analyzing and optimizing games, media, and other graphics intensive applications. For more information, see Intel® Graphics Performance Analyzers 2012 R1

Yum cannot retrieve repository metadata for CentOS 5

If you encountered this type of errors, do look at your repo configuration at /etc/yum.repos.d. In my situation, I was trying to update a package which falls under Centos-Base.repo

http://mirror.centos.org/centos/5/addons/x86_64/repodata/repomd.xml: [Errno 12] 
Timeout: 
Trying other mirror.
Error: Cannot retrieve repository metadata (repomd.xml) for repository: 
addons. Please verify its path and try again

http://mirror.centos.org/centos/5/addons/x86_64/repodata/primary.sqlite.bz2: 
[Errno 12] Timeout: Trying other mirror.
http://mirror.centos.org/centos/5/addons/x86_64/repodata/primary.sqlite.bz2: 
Errno 12] Timeout: Trying other mirror.
Error: failure: repodata/primary.sqlite.bz2 from addons: 
[Errno 256] No more mirrors to try.
 

3 things you need to do

1. Check that your mirriorlist in /etc/yum.repo.d is uncommented. It should look something like
   mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=updates
2. Check that your baseurl is uncommented. (Usually this is not needed, but this is a last resort
   baseurl=http://mirror.centos.org/centos/$releasever/os/$basearch/
3. Type the parameter in the repo list
   enabled=1

ARISTA Switch - Error Detection Automation

From ARISTA site. This video clip is quite useful in explaining AEM Use Case: 'Error Detection Automation'

Advanced Event Management (AEM), a sub-system of Arista EOS, the world's most advanced network operating system, is a powerful and flexible tool to automate tasks and customize the behavior of EOS and the operation of the overall data center switching infrastructure.

1. Event-driven controls bring simplicity to network operations: Exporting massive amounts of data from the switch to a server just to find some small subset of information is inefficient and unnecessarily complex. Why not let the individual network elements do the work of parsing and alerting based on configurable events. By reducing the 'noise' operations staff increase their visibility to critical events.
2.  AEM allows operators to fully utilize the intelligence within EOS to respond to real-time events, automate routine tasks, and take local automated action based on real-time network conditions for improved network performance. 
3. According to Gartner Group 80% of mission-critical systems outages through 2015 will be caused by people and process issues: AEM is purpose-built to enhance system up-time and enable distributed systems automation to simplify complex tasks, react programmatically to network changes, and to simplify complex troubleshooting tasks—thus minimizing systems outages.

Flushing DNS Cache in CentOS

If you need to flush dns, you can use the following command to:

# service nscd restart

Usually you do not need to modify the /etc/nscd.conf and the current default should suffice for most situation.

According to Linux Manual Page
Nscd  is a daemon that provides a cache for the most common name  service  requests.  The  default  configuration file,  /etc/nscd.conf,  determines  the  behavior  of the cache daemon. See nscd.conf(5).
 
Nscd provides cacheing for  accesses  of  the  passwd(5), group(5),  and  hosts(5)  databases through standard libc interfaces,  such  as  getpwnam(3),  getpwuid(3),  getgrnam(3), getgrgid(3), gethostbyname(3), and others.......

andLinux - Run Linux natively on Windows

According to the Project Site,

What is andLinux?

andLinux is a complete Ubuntu Linux system running seamlessly in Windows 2000 based systems (2000, XP, 2003, Vista, 7; 32-bit versions only)........

andLinux uses coLinux as its core which is confusing for many people. coLinux is a port of the Linux kernel to Windows.Although this technology is a bit like running Linux in a virtual machine, coLinux differs itself by being more of a merger of Windows and the Linux kernel and not an emulated PC, making it more efficient. Xming is used as X server and PulseAudio as sound server.

System Requirements:

• OS: Windows 2000 / XP / 2003 / Vista / 7 (currently only the 32-bit versions are supported)
• Memory: at least 128 MB (192 MB or more is recommended)
  
• Hard disk space: 2.5 GB (XFCE version) / 4.5 GB (KDE version)
  Note that you need an NTFS file system (which is default since Windows 2000) because you can't create files larger than 2 GB on FAT(32) file systems!

Limitations:

• andLinux is not suitable for High-Performance realtime graphics
• Only 1 core will be used even though you may have multi-core machine. This is due to andlinux processes are encapsulated in one Windows process.

CIO Awards 2012

CIO 2012 Awards Listing....including the CIO 100 Index (Pdf format) are in.

Intel Recommended Readings

I have always enjoyed Intel Press Books and Technology Journal. Their motto says it all - By developer for Developer. You can download the Recommended Reading List or look at some technical articles from Intel Industry . With the title or the ISBN, you can purchase the book from Amazon.com

Benefits from reading books on the Intel Recommended Reading List:

• Reduce the risk in your development cycle
• Leverage the latest technology & best practices
• Avoid costly design errors
• Create higher performance designs
• Accelerate your development process

Using wget to download recursively from the FTP site

The content of the articles is taken from Man Page and  How to download recursively from an FTP site

Here is some useful commands from the articles.

Option 1: Recursive with depth

$ wget -r --level=99 ftp://userid:password@ftp.mysite.com/

* The default is --level=5

Option 2: Mirroring

$ wget -m ftp://userid:password@ftp.mysite.com/

This option turns on recursion and time-stamping, sets infinite recursion depth and keeps FTP directory listings.

Option 3:  Mirroring in the background

$ nohup wget -m ftp://userid:password@ftp.mysite.com &

Option 4: Preserve local files and avoid overwritting. Use --no-clobber option

$ wget -nc -m ftp://userid:password@ftp.mysite.com/