Linux Tips

3 Advanced usages of Linux ‘date’ command

2011-12-02T19:09:00.001+08:00

In Linux shell, ‘date’ is a basic command to show current date, like:

$ date Sat Jan 8 15:34:16 CST 2011 $

From the result, it shows more than just date, and also time and time zone. It’s different from what DOS does, which use two command to get/set date and time. However the ‘date’ command can do much more than this, let’s first check the usage.
Standard version:

$ date --help Usage: date [OPTION]... [+FORMAT] or: date [-u|--utc|--universal] [MMDDhhmm[[CC]YY][.ss]] Display the current time in the given FORMAT, or set the system date. ...[skipped]

busybox version:

# date --help BusyBox v1.11.1 (2010-12-23 19:34:38 CST) multi-call binary
Usage: date [OPTION]... [+FMT] [TIME]
Display time (using +FMT), or set time ...[skipped]

The busybox version has less functions than standard version (of course!), however the following scenarios are supported on both versions.

1. Show current date/time in a specific format

If the ‘date’ command is followed by a plus sign (+) and immediately the specific format string, it can be sued to determine the date/time output format. Here is all the supported format for ‘date’ command.

FORMAT controls the output. The only valid option for the second form specifies Coordinated Universal Time. Interpreted sequences are:
%%   a literal % %a   locale's abbreviated weekday name (e.g., Sun) %A   locale's full weekday name (e.g., Sunday) %b   locale's abbreviated month name (e.g., Jan) %B   locale's full month name (e.g., January) %c   locale's date and time (e.g., Thu Mar 3 23:05:25 2005) %C   century; like %Y, except omit last two digits (e.g., 21) %d   day of month (e.g, 01) %D   date; same as %m/%d/%y %e   day of month, space padded; same as %_d %F   full date; same as %Y-%m-%d %g   last two digits of year of ISO week number (see %G) %G   year of ISO week number (see %V); normally useful only with %V %h   same as %b %H   hour (00..23) %I   hour (01..12) %j   day of year (001..366) %k   hour ( 0..23) %l   hour ( 1..12) %m   month (01..12) %M   minute (00..59) %n   a newline %N   nanoseconds (000000000..999999999) %p   locale's equivalent of either AM or PM; blank if not known %P   like %p, but lower case %r   locale's 12-hour clock time (e.g., 11:11:04 PM) %R   24-hour hour and minute; same as %H:%M %s   seconds since 1970-01-01 00:00:00 UTC %S   second (00..60) %t   a tab %T   time; same as %H:%M:%S %u   day of week (1..7); 1 is Monday %U   week number of year, with Sunday as first day of week (00..53) %V   ISO week number, with Monday as first day of week (01..53) %w   day of week (0..6); 0 is Sunday %W   week number of year, with Monday as first day of week (00..53) %x   locale's date representation (e.g., 12/31/99) %X   locale's time representation (e.g., 23:13:48) %y   last two digits of year (00..99) %Y   year %z   +hhmm numeric timezone (e.g., -0400) %:z +hh:mm numeric timezone (e.g., -04:00) %::z +hh:mm:ss numeric time zone (e.g., -04:00:00) %:::z numeric time zone with : to necessary precision (e.g., -04, +05:30) %Z   alphabetic time zone abbreviation (e.g., EDT)
By default, date pads numeric fields with zeroes. The following optional flags may follow `%':
- (hyphen) do not pad the field _ (underscore) pad with spaces 0 (zero) pad with zeros ^ use upper case if possible # use opposite case if possible

It looks a very long table, it’s mandatory to meet any requests. Among these usage, the red color part are not supported by busybox version.
Let’s discover the format by presenting some examples:

$ date
Sat Jan 8 16:32:05 CST 2011
$ date +%x
01/08/2011
$ date %X
date: invalid date `%X' #you’ll get such prompt when missing ‘+’$ date +%X
04:32:22 PM
$ date +%a
Sat
$ date +%A
Saturday
$ date +%D
01/08/11
$ date +%F
2011-01-08
$

If you want a more complicated output (like including a single quote (‘) in your output, you’ll need the double quote (“) to enclose your format string. Here is the example:

$ date +"Hello world. It's: %Y/%m/%d %H:%M:%S"
Hello world. It's: 2011/01/08 16:37:34
$

Yet there is a very special format, normally used when writing a shell script. It’s “%s”, it will show the elapsed time in seconds since January first 1970 (1970/1/1). If you holds two different time with above seconds, then substrate these two number will get the elapsed time (We’ll talk about this later in details). Here is the output of “%s”:

$ date +%s
1294476056
$

    2. Set system time
First of all, one thing you should know: Only root has the permission to set system time. Any other users trying to do this will get a error message “Operation not permitted”, and system time keep unchanged:

$ date 01020101
date: cannot set date: Operation not permitted
Sun Jan 2 01:01:00 CST 2011
$

Normally we don’t do this in a PC Linux, maybe you’ll use NTP client to synchronize with NTP server to get a precision time. Sometimes we’ll set system time in a embedded device, so let’s check the command syntax on busybox:

# date --help
    ...[skipped]
Recognized formats for TIME:
        hh:mm[:ss]
        [YYYY.]MM.DD-hh:mm[:ss]
        YYYY-MM-DD hh:mm[:ss]
        MMDDhhmm[[YY]YY][.ss]
#

busybox provide 4 main formats to set system time, for each format you can yet skipped some parts. Frankly speaking, it’s kind of lousy. So the practical exercise is knowing how to get the help and follow it.Here are some examples:
1. Set system time to 5:35

# date
Mon Jan 10 07:03:27 UTC 2011
# date 5:35
Mon Jan 10 05:35:00 UTC 2011
#

2. Set system date/time to May 8th 0:00

# date
Mon Jan 10 05:36:32 UTC 2011
# date 05080000
Sun May 8 00:00:00 UTC 2011
#

3. Set date/time to 2011 Jan 11st 8:00 PM

# date
Mon Jan 10 05:36:32 UTC 2011
# date 2011.01.11-20:00
Tue Jan 11 20:00:00 UTC 2011
#

As the 3 examples show, you can set just time, or date without year but with time, or a complete setting on year, month, day and time.

    3. Calculate elapsed time
Sometimes we’ll use shell script to do a test or monitor something on the system. It’s good to know how much time has been elapsed.
Utilize “date +%s” mentioned above, and a little calculation, we can generate a shell function to calculate elapsed time. Here is an example:

$ cat elap.bash
#!/bin/bash
function elapsed_time()
{
    s_time=$1
    e_time=$2
    elap_s=$((e_time-s_time))
    ss=$((elap_s%60))
    mm=$(((elap_s/60)%60))
    hh=$((elaps/3600))
    printf "%i:%02i:%02i" $hh $mm $ss
}
begin_time=`date "+%s"`
for i in {1..3};
do
    now_time=`date "+%s"`
    elap_time=`elapsed_time $begin_time $now_time`
    echo "elapsed time=$elap_time"
    sleep 1
done

The function “elapsed_time” reads two arguments generated by “date +%s”, they present two different time point. After a simple calculation, hh, mm and ss, represent elapsed hours, minutes and seconds respectively. Finally the formatted output function “printf” is used to generate the output. One thing to know: this is a simple example, so it doesn’t cover the “day” concept, modify it yourself if you need it.
This simple script will call “elapsed_time” function every second, and print elapsed time.
Here is the execution result:

$ source elap.bash
elapsed time=0:00:00
elapsed time=0:00:01
elapsed time=0:00:02
$

Summary

Although ‘date’ is a very common command that we usually used, it may also reach a useful function.

2 ways to generate random numbers in Bash shell

2011-11-24T19:50:00.001+08:00

Have you ever tried to write Bash shell script and need a random number generator? Here provides 2 ways to achieve this.

1. Use the Built-in variable $RANDOM of Bash

$RANDOM is a built-in variable of Bash, it can be used to generate a random number. It's very easy to use, we can use echo to get its value. Here is the example:

$ echo $RANDOM
4371
$ echo $RANDOM
20269
$

Below is another example, which will generate 3 random numbers automatically:

$ for i in {1..3}; do echo $RANDOM; done
11434
9696
32237
$

The number generated by $RANDOM is between 0 and 32767, here is the explanation in detail.

If you want another range, you can use the following method to generate it. Here we create a shell function random_range, which can accept up to 2 arguments. If there is only one argument, it will generate random number between 0 and first argument. Yet if there is 2 arguments, these two numbers will be the lower bound and upper bound of generated numbers. (To keep it simple in the example, there is no strict checking mechanism, please make sure the second argument is larger than the first one).

function random_range()
{
    if [ "$#" -lt "2" ]; then
        echo "Usage: random_range <low> <high>"
        return
    fi
    low=$1
    range=$(($2 - $1))
    echo $(($low+$RANDOM % $range))
}

Here is the execution results:

$ random_range 10
Usage: random_range <low> <high>
$ random_range 10 30
18
$ random_range 10 30
13
$ random_range 10 30
10
$ random_range 10 300
85
$ random_range 10 300
221
$ random_range 10 300
103
$

2. Use the special file /dev/urandom

There are 2 special files in Linux-like system: dev/random and dev/urandom. They are used to provide the random number for Kernel.urandom is the one which is proper to act as a normal use in user space.

The following command utilizes the od command to generate a random number between 0 and 32767, which is exactly the same as above $echo $RANDOM command where the random number is 16-bits wide.

$ od -vAn -N2 -i /dev/urandom
6484
$ od -vAn -N2 -i /dev/urandom
25989
$

The od command alone with /dev/urandom can yet generate a much larger random number. Below is an example with 32-bits wide random number which is located in the range of 0 ~ 2147483645.

$ od -vAn -N4 -tu4 /dev/urandom
2981514769
$ od -vAn -N4 -tu4 /dev/urandom
1055285400
$

Summary

Today 2 ways are provided to generate random numbers in Bash. Apparently the first way with $RANDOM is a easier way. However if you need a very large number, the second way with /dev/urandom could be your better choice.

Linux shell IO redirect

2011-11-17T12:26:00.001+08:00

IO redirect is fundamental in Linux shell -- one basic thing you should pay your attention when learning/using Linux.

3 Basic File descriptors:

File Descriptor	Name	Description
0	stdin	standard input
1	stdout	standard output
2	stderr	standard error

Below I'll introduce 3 advanced usage, well they are also very common scenarios.

1. Redirect stderr to stdout

$ ls non-exist 2>&1

Note the "2>&1" sequence, please keep them there. This command will redirect stderr to stdout.

2. Output stdout and stderr to a file simultaneously

Think about this: you are compiling a huge project which uses Makefile, the compilation failed and a lot of error message shown on the scree. It could be easier for you to analyze these error messages in your favorite editor.

Here is the solution. Either of below ways will achieve this. And of course the show-up sequence is exactly the same as on the screen.

$ find /var -name run &>file

$ find /var -name run >file 2>&1

Please note:(1) The sequence of "&" and ">" on two ways is different, don't mix them up!
(2) For the second one, you must assign stdout first, then stderr.

3. Ignore stdout or stderr

Sometimes you don't want to see some message just because they are trifling or you want to keep a clean screen. You can put everything into the bottomless pit: /dev/null:

This example ignores stderr:

$ find /var -name run 2>/dev/null

This example ignores both stdout and stderr:

$ secret-app 1>/dev/null 2>/dev/null

How to check Linux distribution version

2011-11-16T18:08:00.001+08:00

Do you know how to check Linux distribution version by a shell command?

In Linux, the most common command to show current version is "uname-r" to check the kernel version:

[jackieyeh@dhcppc15 ~]$ uname -r
2.6.18-6-686
[jackieyeh@dhcppc15 ~]$

What if you'd like to know the distribution version in stead of kernel version?

Since there are some many "flavors" in Linux, the Linux version is distinguished not only by kernel version, but also distribution versions, like Fedora, Debian, Ubuntu, etc...

Here is the summary of the ways to get the distribution version. Although different distributions may have slight difference, luckily you can get the answers with couple trials.

1. cat /proc/version

By typing "cat /proc/version", you'll get similar results as "uname", but it's more complete. This is a generic command for every distribution, you may or may not get the distribution information, but anyway you can get some "clues". Below are 2 examples of Fedora and Debian:

Fedora:

[jackieyeh@vega ~]$ cat /proc/version
Linux version 2.6.23.17-88.fc7 (mockbuild@xenbuilder2.fedora.redhat.com) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-27)) #1 SMP Thu May 15 00:35:10 EDT 2008
[jackieyeh@vega ~]$

Debian:

packetship:~# cat /proc/version
Linux version 2.6.18-6-686 (Debian 2.6.18.dfsg.1-24etch2) (dannf@debian.org) (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) #1 SMP Tue May 5 00:40:20 UTC 2009
packetship:~#

2. cat /etc/*-release

By typing "cat /etc/*-release", you may get the distribution version. The "*" is a wild-card character as we don't know the exact versions (that's what we try to get, remember?). This command is effective on Fedora and Ubuntu. Below is an example of Fedora.

Fedora:

[jackieyeh@vega src]$ cat /etc/*-release
Fedora release 7 (Moonshine)
Fedora release 7 (Moonshine)
[jackieyeh@vega src]$

3. cat /etc/issue

By tying "cat /etc/issue", you can always get the result -- This is effective for every distribution since it's a standard file for Linux system and always there. Normally this file will show distribution version by default. However this file is usually edited by the system administrator to show welcome message when a user log in the system -- therefore the distribution version may be removed.

Below are 2 examples of Fedora and Debian:

Fedora:

[jackieyeh@vega ~]$ cat /etc/issue
Fedora release 7 (Moonshine)
Kernel \r on an \m
[jackieyeh@vega ~]$

Debian:

packetship:~# cat /etc/issue
Debian GNU/Linux 4.0 \n \l
packetship:~#

4. cat /etc/debian_version

Obviously, typing "cat /etc/debian_version" is exclusively for Debian. It will show Debian distribution version without any problem.

Debian:

packetship:~# cat /etc/debian_version
4.0
packetship:~#

Summary:
Generally speaking, by typing "cat /etc/issue" normally you can get your answers. And it's effective cross every distribution, so it should be your first choice. Once it's modified by the system administrator and prevent you from getting the answer, then you can try other ways as suggested above.

Linux Tips

3 Advanced usages of Linux ‘date’ command

1. Show current date/time in a specific format

2. Set system time

3. Calculate elapsed time

Summary

2 ways to generate random numbers in Bash shell

Have you ever tried to write Bash shell script and need a random number generator? Here provides 2 ways to achieve this.

1. Use the Built-in variable $RANDOM of Bash

2. Use the special file /dev/urandom

Summary

Today 2 ways are provided to generate random numbers in Bash. Apparently the first way with $RANDOM is a easier way. However if you need a very large number, the second way with /dev/urandom could be your better choice.

Linux shell IO redirect

3 Basic File descriptors:

Below I'll introduce 3 advanced usage, well they are also very common scenarios.

1. Redirect stderr to stdout

Note the "2>&1" sequence, please keep them there. This command will redirect stderr to stdout.

2. Output stdout and stderr to a file simultaneously

3. Ignore stdout or stderr

How to check Linux distribution version

1. cat /proc/version

2. cat /etc/*-release

3. cat /etc/issue

4. cat /etc/debian_version