Unbuffered memory is memory where the memory controller module drives the memory directly, instead of using a store-and-forward system like registered memory. Buffered memory modules have built-in registers on their address and control lines. A register is a very small temporary holding area (usually 64 bits) for data. These registers act as buffers between the Glossary Link CPU and the memory.

Some systems do not support unbuffered memory, others require unbuffered memory, and many more give you the option to use unbuffered or registered memory. The use of unbuffered memory is reasonable for gaming systems. It is not recommended for server-class systems.

Kingston registered memory

The use of registered memory increases the reliability of the data, that is why most true servers motherboards require registered memory. Registered memory is slightly slower than unregistered memory (1 clock cycle) as data must be moved through the registers. The use of registered memory is recommended for server-class systems. It is not useful on gaming systems.

Finally, the unregistered RAM is cheaper than the registered RAM.

Related posts:

1. How about the memory limitations for linux?
2. Clean up buffer and cached memory by defrag_mem

Please note that this post does not cover the topics on password strength. you can use some online tools to check password strength.

GMail Password Strength Check

You can use the following four commands directly in most of Linux distributions with default install.

joseph@test:~$ < /dev/urandom tr -dc _A-Z-a-z-0-9 | head -c8
3saMDFXB

I personally prefer to use this command. The other commands work great too, check below:

cat /dev/urandom | tr -cd 'a-f0-9' | head -c 16
dd if=/dev/random count=10 bs=1 | hexdump  | cut -d   -f 2-| head -n 1 | tr -d " "

uuencode which is supplied by sharutils is needed by this command (Sharutils is not installed by default):

head -c 12 /dev/random | uuencode -m - | tail -n 2 | head -n 1

By using the following command, you need to get mkpasswd installed. It’s man page is available here. It can be run like this:

mkpasswd -l 16 -s 0 -C 0|sed 's/[0ol1]/f/g'

Finally, if you have pwgen installed, you can also generate a random password in linux command line like this:

pwgen -ycn 16 1

PWGen supports Windows platform as well, you just need to install pwgen-win instead, it has a nice GUI.

Related posts:

1. Ways to reset MySQL user password
2. Generate MAC Address by Scripts
3. Dmidecode: Report hardware information from command line

Before we get into the article, here are a few factoids like Haystack and Cassandra (note that this article was created in Middle, 2009) to give you an idea of the scaling challenge that Facebook has to deal with.

Let’s go back here at http://royal.pingdom.com to check the full version of this article Exploring the software behind Facebook, the world’s largest site!

Related posts:

1. Cassandra: Structured Key-Value based storage system
2. Haystack new storage solution for billions of photos
3. Flashcache: caching data in SSD disks

Deprecated: Assigning the return value of new by reference is deprecated in jv_eoty/jv_menus/jv.common.php on line 61

Apparently in PHP5 all objects are passed as reference and 5.3.x formally deprecated the feature, which has forced me to turn off error reporting until I can update all my php code.

So the problem is that traditionally objects would be passed by value and you had to explicitly declare when you wanted the object to be passed by reference, but in PHP 5.3.x you don’t need to do that any more.  For instance, a simplified example would be:

class myClass{
  public $temp = array();
  function new_construct() {
    $this->temp = array[1,2,3];
  }
}

$someArray = & new myClass();

The & sign tells PHP to return the value as a reference. In PHP 5.3.x you don’t need to specify the &, otherwise, you’ll get to warning message that we mentioned here!

Related posts:

1. Avoid script running multiple times by file lock
2. Details about TCP_MISS/000 in Squid access.log

1, rcconf - http://packages.debian.org/lenny/rcconf

This tool configures system services in connection with system runlevels. It turns on/off services using the scripts in /etc/init.d/. Rcconf works with System-V style runlevel configuration. It is a Text User Interface (TUI) frontend to the update-rc.d command.

To start rcconf, login as root and type the command “rcconf“, then select the service that you would like to enable or disable, and press Tab to apply a change.

A similar command is sysv-rc-conf, you can make more specified choices in this TUI based configuration tool.

2, sysv-rc – http://packages.debian.org/lenny/sysv-rc

Sysv-rc provides a terminal interface for managing “/etc/rcX.d/” symlinks. With the help of update-rc.d, you can allow turning services on or off simply. Actually it’s much what chkconfig does in Redhat/CentOS, which means you can edit startup scripts for any runlevel.

By default sysv-rc is installed even in a minimal installation. Otherwise you need to run “apt-get install sysv-rc” to install it. Here’re some common usage examples for update-rc.d:

To remove the service SERVICE-NAME:

update-rc.d -f SERVICE-NAME remove

To add the service SERVICE-NAME:

update-rc.d SERVICE-NAME defaults

update-rc.d SERVICE-NAME defaults 29

The argument 29 is optional, which ensures that SERVICE-NAME is called after all scripts whose number is less than 29 have completed, and before all scripts whose number is 30 or greater.

To start the script SERVICE-NAME in runlevels 2345 and stop in 456, run (as root):

update-rc.d SERVICE-NAME start 2 3 4 5 . stop 0 1 6 .

update-rc.d SERVICE-NAME start 30 2 3 4 5 . stop 70 0 1 6 .

You can find more details of this command by running “man 8 update-rc.d”.

Related posts:

1. Set system variables in Debian and CentOS
2. Postfix: Manage transport mapping table
3. How to run 32bit application under x64 system

• Physical Package: The physical package is a microprocessor. For each physical package, it plugs into a physical socket on a mainboard, and may contain one or more processor cores.
• Processor Core: The circuitry that provides ability to decode and execute instructions. A processor core may contain one or more logical processors.
• Logical Processor: The basic unit of processor hardware that allows the software executive in the operating system to dispatch a task or execute a thread context. Each logical processor can execute only one thread context at a time.
• Hyper-Threading: A feature within the IA-32 family of processors, where each processor core provides the functionality of more than one logical processor.
• Multi-Core Processor: A physical package that contains more than one processor core.

(The above definitions come from Intel’s website, it would be decent to identify what we’re talking.)

As we know, on a running Linux/UNIX system, a detailed information of CPU can be retrieved from /proc/cpuinfo, but how can we identify the logical processors and the physical cores smartly?  Basically, there are two rules that will help you achieve these info.

1. Any CPU with the same “core id” are hyperthreads in the same processor core, which means this process core has Hyper-Threading enabled.
2. The “physical id” identifies physical package. Any cpu with the same physical id are threads or cores in the same physical socket.

Let me take Dell PowerEdge 2950 as an example, we’re running Red Hat Enterprise Linux AS release 4 Update 7 on this server.

$ cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 0
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3992.87
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 1
siblings        : 4
core id         : 4
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.03
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 2
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 2
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.01
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 1
siblings        : 4
core id         : 6
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.03
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 4
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 1
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3989.99
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 5
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 1
siblings        : 4
core id         : 5
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.02
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 6
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 3
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.01
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:                                                                   

processor       : 7
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping        : 10
cpu MHz         : 1995.005
cache size      : 6144 KB
physical id     : 1
siblings        : 4
core id         : 7
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3990.03
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management:

From the outputs, we can see that, there’re 8 processor-cores in 2 physical packages (in other words, there’re two quad-cores CPU on this server), and none of them has Hyper-Threading enabled. Here’s a quick explanation for some keywords that listed above:

• The processor entry contains a unique identifier for this logical processor.
• The physical id entry contains a unique identifier for each physical package.
• The core id entry holds a unique identifier for each core.
• The siblings entry lists the number of logical processors that exist on the same physical package.
• The cpu cores entry contains the number of cores that exist on the same physical package.
• The vendor id entry holds the string. GenuineIntel if the processor is an Intel processor, and AuthenticAMD if it’s an AMD processor.

There are some commands here to help identify these type of information quickly on a running Linux:

1. Total amount of processors including the logical ones:

   # cat /proc/cpuinfo | grep "processor"| wc -l

2. Total amount of CPU package:

   # cat /proc/cpuinfo | grep "physical id" | sort | uniq | wc -l

3. How many cores for each CPU. (Here we should call CPU as physical package as we mentioned above):

   # cat /proc/cpuinfo | grep "cpu cores" | wc -l

4. For each physical CPU, the amount of Logical CPU (the logical CPU can be physical cores, soft cores like threaded, or both).

   # cat /proc/cpuinfo | grep "siblings"

Finally, please note that the msg from /proc/cpuinfo depends on your operating systems, which means it’s not always right or up-to-date. If you faced any issues, please create a thread at our support forum, thanks!

Related posts:

1. How to bind a certain process to specified core?

As mentioned at Facebook.com:

We built Flashcache to help us scale InnoDB/MySQL, but it was designed as a generic caching module that can be used with any application built on top of any block device. For InnoDB, when the working set does not fit in the InnoDB buffer pool, read latency is significantly improved due to caching more of the working set in faster media, such as SSD’s. We also improve write performance by first caching writes in SSD’s and lazily flushing the data back to disk.

It looks like its based on the dm-cache work. It’s nice to see that moving again.

Read more details about Flashcache here at Facebook’s announcing page, and Flashcache’s source code is available at GitHub.com.

Related posts:

1. Haystack new storage solution for billions of photos
2. MPB: InnoDB vs MyISAM vs Falcon
3. What data is cached by operation system

Someone probably will say that they can make a simple script in some minutes to do this kind of things. Actually you don’t need to recreate the wheel, the operating system have supplied such an utility already, the command is rename, it’s part of the util-linux package.

With the help of this command, it’s a very easy work to bulk rename files.
But, note that this command is a little different by systems.
Here we’ll show the usage examples for rename in both Redhat / CentOS and Debian / Ubuntu.

1, Batch rename files in Redhat / CentOS
For the scenario we mentioned above, it can be done like this:

$ rename .html .htm *

It renames all files in current working directory that match *.html* to *.htm*.
Note that only the first matched occurrence will be modified. If there’s a file called abc.html.abc.html, after renamed once, the new name is abc.htm.abc.html, not abc.html.abc.htm.

2, Bulk rename files in Debian / Ubuntu
The command here in Debian are quite different which makes it more powerful and professional.
I’ll show two examples with explaination:

$ rename -n 's/.html$/htm/' *

-n means no action will be applied, it will shows detailed information on what files would have been renamed:

$ rename -n 's/.html$/.htm/' abc.html.abc.html
abc.html.abc.html renamed as abc.html.abc.htm

The expression regunation here is in Perl style, which means you can apply multiple actions at one time, like this:

$ rename 's/.html$/htm/, s/css/css.backup/, s/jscript/js/' *

For more details you can raise a forum thread here at our support forum.

Related posts:

1. Rename a Vserver step by step
2. Set system variables in Debian and CentOS
3. Manage Debian auto startup scripts

VENet consists of two Ethernet devices — the one in physical server and another one in virtualized guest. These devices are connected to each other, so if a packet goes into one device it will come out from the other device.

In this post, we’ll share some tips on how to enable venet. The content is mainly from OpenVZ’s official guide.

The commands that we used are listed below with explanations.

Firstly, assuming that we’re in the physical server, We need to add a new device named eth0

vzctl set 150 --netif_add eth0,00:12:34:56:78:9A,veth101.0,00:12:34:56:78:9B --save

The second MAC address is from the mother side, and the first MAC address is self-generated. Click here if you’re not sure how to generate a MAC address.

Then enable forwarding and ARP proxy, and apply some changes in route table:

# cat veth150.sh
ifconfig veth150.0 0
echo 1 > /proc/sys/net/ipv4/conf/veth150.0/forwarding
echo 1 > /proc/sys/net/ipv4/conf/veth150.0/proxy_arp
echo 1 > /proc/sys/net/ipv4/conf/eth0/forwarding
echo 1 > /proc/sys/net/ipv4/conf/eth0/proxy_arp
ip route add 192.168.201.150 dev veth150.0

Warning: Before making these changes to your system, you’re suggested to verify what exactly you’re doing. For example, proxy_arp is problematic in a mixed network.

Secondly, in the VPS we need to config the new device eth0 like this:

/sbin/ifconfig venet0:0 0
/sbin/ifconfig eth0 0
/sbin/ip addr add 192.168.201.150 dev eth0
/sbin/ip route add default dev eth0

The IP address 192.168.201.150 can be in a different network from the physical side, you just need to make sure they are in the same VLAN.

When it’s tested OK, don’t forget to modify your network configuration files in /etc/sysconfig/network-scripts/ for a permanent change.

Related posts:

1. OpenVZ On CentOS 5.4 installation and configuration
2. How do I set up hugepages in RHEL?
3. What network issues should be monitored?

How can we download a RPM package using YUM? Generally, there are two ways, you can use either yum-downloadonly or yum-utils. 1, yum-downloadonly installation and usage example You just need to install a plugin called yum-downloadonly. This plugin supplies a new parameter called –downloadonly so that yum can download the package s without an further installation/upgrade. It’s much like apt-get‘s -d option in Ubuntu/Debian:

$ yum --help| grep download
Loaded plugins: downloadonly, fastestmirror
  --downloadonly        don't update, just download
  --downloaddir=DLDIR   specifies an alternate directory to store package

A further question is How can I install the download-only plugin for Yum? It’s as simple as a normal install, you just need to run:

$ sudo yum install yum-downloadonly

If you wanna download a package without installation, you just need to add the parameter –downloadonly. By default the downloaded package will be stored in /var/cache/yum, you can specifies an alternate directory to store this file such as /opt, like this:

yum update httpd -y --downloadonly --downloaddir=/opt

2, yum-utils.noarch installation and usage example

Yum-utils is a collection of utilities and plugins extending and supplementing yum in different ways, and all the utilities are contained in yum-utils package, including the one that we need which is called yumdownloader.

# yum -y install yum-utils.noarch

Now we can use the yumdownloader command to download packages from yum repositories. Here’s an example:

# yumdownloader httpd

When you get the RPM, you can query further information from it, or decompress it.

Related posts:

1. How to skip broken dependencies when Upgrading System
2. OpenVZ On CentOS 5.4 installation and configuration
3. Update CentOS 4 to CentOS 5 remotely