Pidgin Messenger @ Office

Got approval for using Fedora @ office. But what about the MS Office Communicator or Cisco Webex Connect?  How do I use it with Fedora?

No problem. Use Pidgin Messenger!

I need to configure Cisco Webex Connect.

Install Pigdin and Pidgin Sipe

Code:

yum install pidgin

Code:

yum install pidgin-sipe

For Ubuntu use "apt-get install" command instead.

Start Pidgin

Account > Manage Account > Add

In the Basic Tab enter the required info

Click Add and you are good to go!

Once added from the main window enable the account

Accounts > Enable Accounts > Select Your Account

You can also configure MS Communicator.

Protocol that needs to be selected is "Office Communicator"

IPSec Linux and Solaris (Cont...)

Continued from IPSec Linux & Solaris...

A working IPSec configuration between Solaris 9 and Fedora.

Solaris Machine <> 172.50.20.100 (bge0)

Tunnel IP <> 198.168.1.100 (tun0)

Linux Machine <> 172.50.20.200 (eth0)

Tunnel IP <> 198.168.1.200 (tun0)

Like I've already mentioned, keying can be done manually or auto. Here we will use auto keying i.e. with the help of the IKE daemons.

Linux Configuration

IPSec tools needs to downloaded and installed.

IPSec Installation

Installation of IPSec is fairly simple, just like installing any other application. You need to be root for installation or make sure your admin has given you sudo privileges.

Here, I will provide some info on installing the ipsec-tools on Fedora 16.

The command is ofcourse "yum install ipsec-tools" for Fedora, "apt-get install ipsec-tools" on Ubuntu etc

Code:

root > yum install ipsec-tools
Loaded plugins: langpacks, presto, refresh-packagekit
fedora/group_gz                                                                                                     | 431 kB     00:01     
fedora-cinnamon                                                                                                     | 2.9 kB     00:00     
updates/metalink                                                                                                    | 3.3 kB     00:00     
updates                                                                                                             | 4.5 kB     00:00     
updates/primary_db                                                                                                  | 4.0 MB     00:59     
updates/group_gz                                                                                                    | 431 kB     00:08     
Setting up Install Process
Resolving Dependencies
--> Running transaction check
---> Package ipsec-tools.i686 0:0.8.0-2.fc16 will be installed
--> Finished Dependency Resolution

Dependencies Resolved

===========================================================================================================================================
 Package                            Arch                        Version                                Repository                     Size
===========================================================================================================================================
Installing:
 ipsec-tools                        i686                        0.8.0-2.fc16                           fedora                        364 k

Transaction Summary
===========================================================================================================================================
Install       1 Package

Total download size: 364 k
Installed size: 364 k
Is this ok [y/N]: y
Downloading Packages:
ipsec-tools-0.8.0-2.fc16.i686.rpm                                                                                   | 364 kB     00:01     
Running Transaction Check
Running Transaction Test
Transaction Test Succeeded
Running Transaction
  Installing : ipsec-tools-0.8.0-2.fc16.i686                                                                                           1/1 

Installed:
  ipsec-tools.i686 0:0.8.0-2.fc16                                                                                                          

Complete!
root > 

Just do a verification if you have all these in your machine

Code:

root > which racoon
/usr/sbin/racoon
root > which setkey
/sbin/setkey
root > ls -lrtd /etc/racoon/
drwxr-xr-x. 4 root root 4096 Feb 26 13:22 /etc/racoon/

Note : Older versions of IPSec will have the /etc/ipsec-tools folder instead of /etc/racoon. It doesn't matter really!

All the configuration in Linux is present at /etc/racoon folder.

We will do 2 levels of configuration.

1. Setting up the raccoon

2. Setting up the policies.

Raccoon Configuration

We will need to add/modify the /etc/racoon/racoon.conf file to suit our requirements which is read by the racoon daemon.

/etc/racoon/racoon.conf

Code:

#Racoon will use this file for pre_shared_key autentication#
path pre_shared_key "/etc/racoon/psk.txt";

remote 172.50.20.100 {
   exchange_mode main;
   lifetime time 360 min;
   nonce_size 20;
   proposal {
     dh_group 5;
     encryption_algorithm 3des;
     hash_algorithm md5;
     authentication_method pre_shared_key;
   }
}

#sainfo any
sainfo address 198.168.1.100 any address 198.168.1.200 any
{
   pfs_group 5;
   encryption_algorithm aes;
   authentication_algorithm hmac_md5;
   compression_algorithm deflate;
}

As you can see the pre-shared key is present at /etc/racoon/psk.txt

/etc/racoon/psk.txt

Code:

172.50.20.100  0x1234567890

That completes the phase 1.

Let us set-up some policies here and then go to the Solaris box.

Policies

Now we set up the required policies that needs to matched for the traffic. The command used is "setkey". Let us put all the configuration in /etc/racoon/setkey.conf

/etc/racoon/setkey.conf

Code:

spdadd 172.50.20.200 172.50.20.100 any -P out ipsec 
  esp/tunnel/198.168.1.200-198.168.1.100/require;

spdadd 172.50.20.100 172.50.20.200 any -P in ipsec
  esp/tunnel/198.168.1.100-198.168.1.200/require;

Use the setkey command to setup the policies as shown.

Code:

root > setkey -f /etc/racoon/setkey.conf

Verify if your policies are properly loaded using -PD option.

Code:

root > setkey -PD
172.50.20.100[any] 172.50.20.200[any] 255
 fwd prio def ipsec
 esp/tunnel/198.168.1.100-198.168.1.200/require
 created: Feb 26 17:26:14 2012  lastused:                     
 lifetime: 0(s) validtime: 0(s)
 spid=18 seq=1 pid=9246
 refcnt=1
172.50.20.100[any] 172.50.20.200[any] 255
 in prio def ipsec
 esp/tunnel/198.168.1.100-198.168.1.200/require
 created: Feb 26 17:26:14 2012  lastused:                     
 lifetime: 0(s) validtime: 0(s)
 spid=8 seq=2 pid=9246
 refcnt=1
172.50.20.200[any] 172.50.20.100[any] 255
 out prio def ipsec
 esp/tunnel/198.168.1.200-198.168.1.100/require
 created: Feb 26 17:26:14 2012  lastused:                     
 lifetime: 0(s) validtime: 0(s)
 spid=1 seq=0 pid=9246
 refcnt=1

As you can see the kernel adds a default forward policy. In older versions, we had to add it manually.

There is a hell lot of information I want to add along with this regarding the setkey command, the parameters in racoon configuration, the algorithms used etc. May be in another post!

Before going to the Solaris box, let us start the racoon daemon.

Code:

root > /etc/init.d/racoon start
Starting racoon (via systemctl):                           [  OK  ]

Just confirm it using pgrep

Code:

root > pgrep racoon
9278

So far everything looks fine! Let us move onto the Solaris Box.

Solaris

In Solaris, ike is the daemon responsible for what racoon does in Linux. ipseckey and ipsecconf is responsible for what setkey does.

So, like racoon.conf here we have the ike configuration which is present at 

/etc/inet/ike/config

Code:

#Phase 1
p1_lifetime_secs 21600
p1_nonce_len 20

p1_xform {auth_method preshared oakley_group 5 auth_alg md5 encr_alg 3des }

#Phase 2
p2_pfs 5

{
   label "172.50.20.100 - 172.50.20.200"
   local_id_type IP
   local_addr 172.50.20.100
   remote_addr 172.50.20.200
}

The preshared key needs to added at /etc/inet/secret/ike.preshared

/etc/inet/secret/ike.preshared 

Code:

172.50.20.200  1234567890

Now we need to set up the policy like we did in Linux. Let us have the policy in /etc/inet/ike/ipsecconf.conf

/etc/inet/ike/ipsecconf.conf

Code:

{ 
  tunnel tun0 
  negotiate tunnel laddr 198.168.1.100/32 raddr 198.168.1.200/32 dir out 
} ipsec { 
  encr_algs aes-cbc(128..256) encr_auth_algs hmac-md5(128) sa shared 
}

Unlike the Linux policy, we need only one in this case, basically because it supports "both" in direction.

Load the policy using the following command

Code:

root > ipsecconf -f /etc/inet/ike/ipsecconf.conf

Verify if your policies are properly loaded.

Code:

root > ipsecconf -ln
#INDEX tun0,1{ tunnel tun0   negotiate tunnel laddr 198.168.1.100/32 raddr 198.168.1.200/32 
dir out } ipsec {   encr_algs aes-cbc(128..256) encr_auth_algs hmac-md5(128) 
sa shared }

Now start the ike daemon and verify it.

Code:

root > /usr/lib/inet/in.iked

Try pinging the tunnels from either boxes.

Use the command "setkey -D" to see if the associations are created on the Linux box and "ipseckey dump" on Solaris box.

Will add more information later!

Have a good day!

Date command in Linux

Man page of date will give you more info than anything else, it is a very descriptive one.

I wanted to hightlight few of the options used and thier alternatives, mainly the option "-d". Why because it looks like solaris version of date does not have "-d" option.

The man page says

       -d, --date=STRING

              display time described by STRING, not `now'

Lets take few examples

Firing the date command would give you the date

Code:

root@bt > date
Wed Jan  4 07:52:17 IST 2012

Do you want to customize this? No problem. Use the format control. There are like more than 20 format control options.

Say, you want the date in DD/MM/YYYY format

Code:

root@bt > date '+%d/%m/%Y'
04/01/2012

       %d     day of month (e.g, 01)

       %m     month (01..12)

       %Y     year

Want the current date in seconds since the epoch? Use "+%s" option

Code:

root@bt > date '+%s'
1325644010

       %s     seconds since 1970-01-01 00:00:00 UTC

As you can see, "+" is mandatory which is followed by the format specifier.

To set/change the system date, use the "-s" option.

       -s, --set=STRING

              set time described by STRING

Code:

root@bt > date
Wed Jan  4 08:21:47 IST 2012

root@bt > date -s "Mon Dec 31 00:00:00 IST 1990"
Mon Dec 31 00:00:00 IST 1990

root@bt > date
Mon Dec 31 00:00:01 IST 1990

Now, what does "-d" option do? Hmmm, if you see we were dealing with current system date till now. What if you want to do the same with some other date? Well, that is where "-d" option comes to the rescue!

Say, you want to find the seconds since epoch till December 31 1990

Code:

root@bt > date -d "Dec 31 1990" '+%s'
662581800

You want to format a date given to you in "31 Dec 1990" to MM-DD-YY

Code:

root@bt > date -d "31 Dec 1990" '+%m-%d-%y'
12-31-90

Cool right? Now, the "-d" option is not available in Solaris (not sure about 10 though).

Here is an alternative in awk for achieving the above

To get the time in seconds

Code:

root@bt > echo "Dec 31 1990" | nawk 'BEGIN{month="JanFebMarAprMayJunJulAugSepOctNovDec"}
{ print mktime($3" "int((index(month,$1)/3))+1" "$2" 00 00 00") }'
662581800

Code:

root@bt > echo "31 Dec 1990" | nawk 'BEGIN{month="JanFebMarAprMayJunJulAugSepOctNovDec"}
{ print int((index(month,$2)/3))+1"-"$1"-"substr($3,3) }'
12-31-90

We can practically do anything in awk! Check the right side link for AWK tutorial.

One more thing before we go, how do we convert the seconds to date?

Say, I have 662581800 (from our previous example), I want to convert it to actual date format

Code:

root@bt > date -d @662581800
Mon Dec 31 00:00:00 IST 1990

Note the "@" symbol before the seconds.

Also note that I have used nawk instead of awk, because in solaris the awk version is pretty old and it is kind of useless. Either use "nawk" or awk from this path "/usr/xpg4/bin/awk" in solaris.

Good Day!

Command Substitution in Linux

In Linux, we do a lot of onliners on the shell. Something like,

Code:

root@bt > date
Tue Jan  3 01:19:19 IST 2012

Now if we want to store this date into a variable, what do we do?

There are 2 ways in which we can achieve this i.e. command substitution

One way, using backticks i.e. `

Code:

root@bt > var=`date`
root@bt > echo $var
Tue Jan 3 01:20:47 IST 2012

Other way is using $(...)

Code:

root@bt > var=$(date)
root@bt > echo $var
Tue Jan 3 01:21:18 IST 2012

Command line or script, you can use it anywhere you want. I prefer the second way!

Note, there is no space between var = and the command. Why so? Because that is how it should be.
That is the syntax. If you have a space in between you will see strange errors.
Let us try it once with a space inbetween,

Code:

root@bt > var =$(date)
No command 'var' found, did you mean:
 Command 'jar' from package 'openjdk-6-jdk' (main)
 ...
 Command 'par' from package 'par' (universe)
var: command not found

Oops, we don't want that now, do we?

Good Day!

File Creation Time in Linux

Actually, the creation time of any inode is(was) not stored anywhere. I wonder why they did that. Normally we get to see only the following :

ctime - change time
mtime - modification time
atime - access time

What about crtime - creation time?

But things have changed now. We can get the file creation time if your filesystem type is ext4.

First lets see how we can get the details of a file. There is a command called "stat" - display file or filesystem status.

Code:

root@bt > stat somefile
File: `somefile'
Size: 149        Blocks: 8          IO Block: 4096   regular file
Device: 804h/2052d Inode: 211582      Links: 1
Access: (0644/-rw-r--r--)  Uid: (    0/    root)   Gid: (    0/    root)
Access: 2012-01-01 20:21:15.276575811 +0530
Modify: 2012-01-01 20:21:14.256570755 +0530
Change: 2012-01-01 20:21:14.256570755 +0530

"somefile" is a normal ascii file. A "stat" on the file will give you all these details. You can also customize the output. Let us say you just want the size of the file.

Code:

root@bt > ls -lrt somefile
-rw-r--r-- 1 root root 149 2012-01-01 20:21 somefile
root@bt > stat -c %s somefile
149

-f will give you information about the file system.

Code:

root@bt > stat -f /dev/sda4
File: "/dev/sda4"
  ID: 0        Namelen: 255     Type: tmpfs
Block size: 4096       Fundamental block size: 4096
Blocks: Total: 255610     Free: 255524     Available: 255524
Inodes: Total: 214563     Free: 213660

Now, the -t option is interesting. If you want to get all this info in one line, use this option. This will be very helpful when it comes to scripts where we use stat to get various information into a variable and do some calculation etc.

Code:

root@bt > stat -t -f /dev/sda4
/dev/sda4 0 255 1021994 4096 4096 255610 255524 255524 214563 213660

So, we didn't get the file creation time yet! What do we do for that?
Ok, here it is
1. You will need a ext4 filesystem.
2. You will also need the debugfs -ext2/ext3/ext4 file system debugger utility which is normally available with any distros. If you don't have it, then simply install it.

Check List
Lets get some info on / and /dev/sda4.

Code:

root@bt > df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda4              14G   10G  2.6G  80% /
none                  999M  344K  999M   1% /dev

Code:

root@bt > mount | grep sda4
/dev/sda4 on / type ext4 (rw,errors=remount-ro)

Cool! It is on ext4.

Once you have all that, then all you have to do is run the following command

Code:

root@bt >  debugfs -R 'stat /root/Desktop/somefile' /dev/sda4
debugfs 1.41.11 (14-Mar-2010)
Inode: 211582   Type: regular    Mode:  0644   Flags: 0x80000
Generation: 1690641204    Version: 0x00000000:00000001
User:     0   Group:     0   Size: 149
File ACL: 0    Directory ACL: 0
Links: 1   Blockcount: 8
Fragment:  Address: 0    Number: 0    Size: 0
ctime: 0x4f0072e2:3d2bd60c -- Sun Jan  1 20:21:14 2012
atime: 0x4f0072e3:41f0d90c -- Sun Jan  1 20:21:15 2012
mtime: 0x4f0072e2:3d2bd60c -- Sun Jan  1 20:21:14 2012
crtime: 0x4f0072e2:3d2bd60c -- Sun Jan  1 20:21:14 2012
Size of extra inode fields: 28
EXTENTS:
(0): 772301
(END)

Do you see the crtime? Wow!

bc - a powerful calculator

"An arbitrary precision calculator language' - this is what the man page says. This is fairly a simple tool in term of usage.
For instance :- Just type in "bc" in your shell prompt and you have the tool running.

Code:

root@bt > bc
bc 1.06.95
Copyright 1991-1994, 1997, 1998, 2000, 2004, 2006 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'.

22/7
3

See the "22/7", that is what I have typed in and there is the result. You can also use this like an one liner.

Code:

root@bt > echo " 22/7 " | bc
3

Hmmm I know, the decimals are missing right? No problem, we just need to tell the "bc" to print the decimals. Use "scale" for that purpose.

Code:

root@bt > echo " scale=5; 22/7 " | bc
3.14285

Wow! happy now? ;)

If you go thru the man page, you can see that whatever you do in C using the math lib, you can do it using "bc" as well. What got my attention is that, we can actually have our own functions written and call it.

Code:

root@bt > echo "define foo(a,b){ return a+b }; print foo(10,20)" | bc
30

Cool right? I mean look at the flexibility this small tool is providing you. Well, can't call this a small tool now, can we?

Just for information, I found another program "calc".

Code:

root@bt > calc
C-style arbitrary precision calculator (version 2.12.3.3)
Calc is open software. For license details type:  help copyright
[Type "exit" to exit, or "help" for help.]

; 22/7
~3.14285714285714285714
;

Check this out!

Code:

root@bt > bc
bc 1.06.95
Copyright 1991-1994, 1997, 1998, 2000, 2004, 2006 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'.

scale=100
22/7
3.142857142857142857142857142857142857142857142857142857142857142857\
1428571428571428571428571428571428

I did try with a scale of 1000 and it worked!

Excellent Tutorial on "bc"

Good Day!

match() funtion in awk has a 3rd argument

Just came to know that the function match() in awk has a third argument!
The third argument is an array where you can capture the search results.

For eg:-

Code:

root@bt > cat infile
ajith & ajeesh
root@bt> awk '{match($0,"([a-z]*) & ([a-z]*)",arr); print arr[1]" : "arr[2] }' infile
ajith : ajeesh

([a-z]*) => will capture the first element of the array
&
([a-z]*) => will capture the second element of the array

And the array name here is "arr"