Change the World

The Way Things Are in the Philippines

2012-01-11T11:29:00.000+08:00

Consider these facts and figures from the Doing Business Project and Transparency International. The Doing Business Project provides objective measures of business regulations and their enforcement across 183 economies and selected cities at the subnational and regional level.

The image below shows how the Philippines fared against the rest of the world, it’s in freaking 136^th place out of the 183 countries listed herein! Singapore topped the list.

Next image shows how the Philippines fared against fellow ASEAN neighbors. It’s in 7^th place out of the ten member countries, just ahead of Cambodia, Lao PDR, and the unlisted Myanmar.

We're not even close to Malaysia and Thailand, we were second to Japan back in the days, look where we are now!

To add insult to injury, the Philippines is also at the bottom of Transparency International's corruption perception index 2011 list, we're in 129^thplace out of 183 countries/territories to be exact! The Corruption Perceptions Index ranks countries/territories based on how corrupt their public sector is perceived to be.

Why is it we're always in it for all the wrong reasons? Is there still hope for us? Why can't we get our acts together? When will we get out acts together? Am I going to see that in my lifetime?

You tell me!

Installing OTRS on a 64-bit CentOS 6.2 using PostgreSQL as database

2012-01-09T16:34:00.002+08:00

OTRS is the leading open-source Help Desk and IT Service Management (ITSM) solution used by thousands of organizations worldwide, enabling transparency and collaboration for service desk and customer support teams, including those implementing ITIL Best Practices. OTRS Group offers consulting, support, customization and hosting services.

OTRS is compatible with a number of RDBMS that includes MySQL, PostgreSQL, Oracle, Ingres, DB2, and MSSQL.

I'll be installing OTRS on a 64-bit CentOS 6.2 using PostgreSQL as database in my VirtualBox machine.

First, download the OTRS rpm, I'll be using version 3.1 beta which can be downloaded using wget.

# wget http://ftp.otrs.org/pub/otrs/RPMS/fedora/4/otrs-3.1.0.beta2-02.noarch.rpm

Next is to determine the dependencies required for OTRS to work.

# rpm -qpR otrs-3.1.0.beta2-02.noarch.rpm
perl
perl(DBI)
perl(URI)
mod_perl
httpd
procmail
perl(Date::Format)
perl(LWP::UserAgent)
perl(Net::DNS)
perl(IO::Socket::SSL)
perl(XML::Parser)
/bin/sh
/bin/sh
rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <= 3.0.4-1
rpmlib(PayloadIsBzip2) <= 3.0.5-1

Obviously, we need to install the aforementioned required packages for OTRS to work. Use yum to determine the exact Perl package name, e.g.

# yum provides "perl(Date::Format)"

Loaded plugins: fastestmirror, security
Loading mirror speeds from cached hostfile
* base: centos.mirror.aussiehq.net.au
* extras: mirror.aarnet.edu.au
* updates: centos.mirror.aussiehq.net.au
1:perl-TimeDate-1.16-11.1.el6.noarch : A Perl module for time and date
: manipulation
Repo : base
Matched from:
Other : perl(Date::Format)

Do above for the rest of the Perl packages, install them once done.

# yum install procmail perl-TimeDate perl-libwww-perl perl-Net-DNS perl-IO-Socket-SSL perl-XML-Parser

Install the OTRS package.

# rpm -ivh otrs-3.1.0.beta2-02.noarch.rpm
Preparing... ########################################### [100%]
Check OTRS user ... otrs added.
1:otrs ########################################### [100%]
hostname: Unknown host

Next steps:

[httpd services]
Restart httpd 'service httpd restart'

[install the OTRS database]
Make sure your database server is running.
Use a web browser and open this link:
http:///otrs/installer.pl

[OTRS services]
Start OTRS 'service otrs start' (service otrs {start|stop|status|restart).

((enjoy))

Your OTRS Team

The output message is a bit misleading, we still have to set up PostgreSQL before accessing OTRS.

Initialize PostgreSQL.

# service postgresql initdb

Initializing database: [ OK ]

Start PostgreSQL.

service postgresql start

Starting postgresql service: [ OK ]

postgres=# CREATE USER otrs PASSWORD 'otrs' LOGIN NOSUPERUSER;

CREATE ROLE

postgres=# CREATE DATABASE otrs owner otrs;

CREATE DATABASE

Create tables and insert initial data.

# cd /opt/otrs/scripts/database

Enter the 'otrs' user password when prompted.

# psql -U otrs -W -f otrs-schema.postgresql.sql otrs

# psql -U otrs -W -f otrs-initial_insert.postgresql.sql otrs

# psql -U otrs -W -f otrs-schema-post.postgresql.sql otrs

It's now time to configure OTRS connection parameters to PostgreSQL, comment the default MySQL parameter and uncomment PostgreSQL's, as well as insert the otrs user password used earlier.

# vim /opt/otrs/Kernel/Config.pm

# DatabasePw

# (The password of database user. You also can use bin/otrs.CryptPassword.pl

# for crypted passwords.)

$Self->{DatabasePw} = 'otrs';

# DatabaseDSN

# (The database DSN for MySQL ==> more: "man DBD::mysql")

# $Self->{DatabaseDSN} = "DBI:mysql:database=$Self->{Database};host=$Self->{DatabaseHost};";

# (The database DSN for PostgreSQL ==> more: "man DBD::Pg")

# if you want to use a local socket connection

$Self->{DatabaseDSN} = "DBI:Pg:dbname=$Self->{Database};";

Save file and exit, then start Apache.

# service httpd start

Starting httpd: httpd: apr_sockaddr_info_get() failed for otrs.lan

httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName

[FAILED]

I had to edit /etc/httpd/conf/httpd.conf file and add the following line:

ServerName otrs.lan

You can use another server name, once done, restart Apache

# service httpd start

I checked httpd's error logs and found the following error:

[Mon Jan 09 13:14:56 2012] [error] Can't load Perl file: /opt/otrs/scripts/apache2-perl-startup.pl for server otrs.lan:0, exiting...

[Mon Jan 09 13:15:23 2012] [error] Can't locate DBD/Pg.pm in @INC (@INC contains: /opt/otrs/Custom /opt/otrs/Kernel/cpan-lib /opt/otrs/ /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 . /etc/httpd) at /opt/otrs/scripts/apache2-perl-startup.pl line 58.\nBEGIN failed--compilation aborted at /opt/otrs/scripts/apache2-perl-startup.pl line 58.\nCompilation failed in require at (eval 2) line 1.\n

Based on the above logs, it says that perl file named DBD/Pg is missing, to resolve, install missing Perl package.

# yum install perl-DBD-Pg

Restarted Apache again, and tried to access OTRS. If I try to access OTRS without specifying the full URL i.e. http://ip.add.re.ss/otrs instead of http://ip.add.re.ss/otrs/index.pl, Apache spewed this error:

[Mon Jan 09 13:17:31 2012] [error] slurp_filename('/var/www/html/otrs') / opening: (2) No such file or directory at /usr/lib64/perl5/ModPerl/RegistryCooker.pm line 541

I tried to Google the above mentioned error but to no avail.

BTW, I can login using the username 'root@localhost' and password 'root', and as suggested by OTRS don't use the root user, instead create a new user to administer OTRS.

See login page and root dashboard screenshot below.

I would appreciate any suggestion on how to resolve the last error shown.

That's all there is to it.

Cheers!

References:

http://wiki.otrs.org/index.php?title=Installation_on_Debian_6_with_Postgres
http://gkoenig.wordpress.com/2010/01/14/install-latest-otrs-ticketing-system-with-postgres-on-centos-5-4/

Enabling KVM Bridge Networking in CentOS 6

2011-12-07T07:30:00.001+08:00

The KVM installation in CentOS/RHEL does not have bridge networking enabled by default, hence the need, besides, it's also a requirement that everything should be connected to the physical network. Here's what I did to get it working, this assumes that you have KVM installed, if not you can install it using:

# yum groupinstall KVM

I'm quite sure that bridge-utils is included, if not we have to install it.

# yum install bridge-utils

Next is to disable NetworkManager and use network instead (I have X Windows installed on this server, hence), according to my references, the former does not play well with bridge networking.

# chkconfig NetworkManager off

# chkconfig network on

# service NetworkManager stop

# service network start

Afterwards, edit /etc/sysconfig/network-scripts/ifcfg-eth0 file and append the following:

BRIDGE=br0

Save and exit.

This is what mine looks like after appending:

DEVICE=eth0
HWADDR=e4:1f:13:c2:0f:2a
NM_CONTROLLED=yes
ONBOOT=yes
IPADDR=192.168.100.101
BOOTPROTO=none
NETMASK=255.255.255.0
DNS2=8.8.8.8
TYPE=Ethernet
GATEWAY=192.168.100.254
DNS1=192.168.100.254
IPV6INIT=no
USERCTL=no
BRIDGE=br0

Next is to create the /etc/sysconfig/network-scripts/ifcfg-br0 file, and add the following:

DEVICE="br0"
BOOTPROTO="static"
IPV6INIT="yes"
IPV6_AUTOCONF="yes"
NM_CONTROLLED="no"
ONBOOT="yes"
TYPE="Bridge"
DELAY="0"
IPADDR=192.168.100.101
NETMASK=255.255.255.0
GATEWAY=192.168.100.254
DNS1=8.8.8.8
DNS2=192.168.100.254

Save and exit.

Since I've disabled iptables and SELinux, I don't have to worry about creating rules to allow the changes made, I'm not recommending you do this in your setup.

Restart network to make sure.

# service network restart

To check if bridge networking is enabled, use ifconfig command, result should be similar to the one below:

# ifconfig

br0       Link encap:Ethernet HWaddr E4:1F:13:C2:0F:2A
          inet addr:192.168.100.101 Bcast:192.168.100.255 Mask:255.255.255.0
          inet6 addr: fe80::e61f:13ff:fec2:f2a/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:6711 errors:0 dropped:0 overruns:0 frame:0
          TX packets:3480 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:696422 (680.0 KiB) TX bytes:255890 (249.8 KiB)

eth0      Link encap:Ethernet HWaddr E4:1F:13:C2:0F:2A
          inet6 addr: fe80::e61f:13ff:fec2:f2a/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:5435 errors:0 dropped:0 overruns:0 frame:0
          TX packets:5294 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:724893 (707.9 KiB) TX bytes:556572 (543.5 KiB)
          Interrupt:28 Memory:92000000-92012800

eth1      Link encap:Ethernet HWaddr E4:1F:13:C2:0F:2C
          UP BROADCAST MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
          Interrupt:40 Memory:94000000-94012800

lo        Link encap:Local Loopback
          inet addr:127.0.0.1 Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING MTU:16436 Metric:1
          RX packets:16237 errors:0 dropped:0 overruns:0 frame:0
          TX packets:16237 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:28106982 (26.8 MiB) TX bytes:28106982 (26.8 MiB)

virbr0    Link encap:Ethernet HWaddr BE:E4:C5:B8:DD:30
          inet addr:192.168.122.1 Bcast:192.168.122.255 Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:752 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 b) TX bytes:38001 (37.1 KiB)

vnet0     Link encap:Ethernet HWaddr FE:54:00:29:F6:07
          inet6 addr: fe80::fc54:ff:fe29:f607/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:1547 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2549 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:500
          RX bytes:230891 (225.4 KiB) TX bytes:204319 (199.5 KiB)

You should now be able to see the newly added br0 when you create your virtual machine during advanced guest network configuration.

That's all there is to it!

References:

http://linux.dell.com/files/whitepapers/KVM_Virtualization_in_RHEL_6_made_easy.pdf

http://www.techotopia.com/index.php/Creating_a_CentOS_KVM_Networked_Bridge_Interface

Installing NginX + PHP + PostgreSQL on Fedora 16

2011-11-12T17:56:00.001+08:00

Installing PHP 5.3.8 + POSTGRESQL 9.1 NGINX 1.0.5 on FEDORA 16

Update your machine to get the latest packages:

$ sudo yum update

We'll install PostgreSQL first, we don't need to edit the repository because Fedora 16 comes with PostgreSQL 9.1.

$ sudo yum install postgresql postgresql-server postgresql-libs postgresql-devel postgresql-contrib

Install NginX and PHP next.

$ sudo yum install nginx php php-fpm php-common php-pear php-pdo php-mysql php-pgsql php-pecl-apc php-gd php-mbstring php-mcrypt php-xml

Enable PostgreSQL at boot:

$ sudo systemctl enable postgresql.service

Initialize PostgreSQL:

$ sudo postgresql-setup initdb

Start PostgreSQL:

$ sudo systemctl start postgresql.service

Enable NginX at boot:

$ sudo chkconfig nginx on

Start NginX

$ sudo systemctl start nginx.service

Afterwards, edit /etc/php-fpm.d/www.conf file and change the user and group value from apache to nginx.

$ sudo vim /etc/php-fpm.d/www.conf

---snip---

user = nginx
group = nginx

---snip---

Save the file and exit.

Next, change ownership of /var/lib/php/session/ directory.

$ sudo chown root:nginx /var/lib/php/session/

Change /etc/php.ini settings as well.

$ sudo vim /etc/php.ini

date_timezone = "Asia/Manila"
cgi.fix_pathinfo=0

Save and exit.

Time to start PHP-FPM service, first enable it during boot.

$ sudo chkconfig php-fpm on

Next, start php-fpm.

$ sudo systemctl start php-fpm.service

Test if PHP is working, we need to edit /etc/nginx/conf.d/default.conf and make the following changes, mine looks like this:

$ sudo vim /etc/nginx/conf.d/default.conf

server {
    listen       80;
    server_name _;

    location / {
        root   /usr/share/nginx/html;
        index index.php index.html index.htm;
    }

    error_page 404              /404.html;
    location = /404.html {
        root   /usr/share/nginx/html;
    }

error_page   500 502 503 504 /50x.html;
    location = /50x.html {
        root   /usr/share/nginx/html;
    }

       location ~ \.php$ {
            fastcgi_pass   127.0.0.1:9000;
            fastcgi_index index.php;
            fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
            include        fastcgi_params;
    }

location ~ /\.ht {
        deny all;
    }
}

Save and exit as usual.

Next is to create your index.php test page as well as phpinfo page in /usr/share/nginx/html directory (default location), my index.php just prints "Hello, World!" on the browser.

Restart NginX.

$ sudo systemctl restart nginx.service

Launch your favorite browser and try to access http://localhost/index.php and http://locahost/info.php respectively.

PHP info page

That's all there is to it.

Cheers!

---

Next update will include virtual hosting.

Steve Jobs: One of the few Level 5 and Authentic Leaders

2011-10-06T19:01:00.000+08:00

I first saw this video in my Leadership class in the Ateneo Graduate School of Business, in the video, Steve Jobs is the guest speaker at a commencement exercise in Stanford University in 2005, the speech is so moving, he spoke about life and death, how he viewed life, and how we should live it, I thought of sharing it. The text copy (entire) of that speech can be read here --> http://news.stanford.edu/news/2005/june15/jobs-061505.html .

Steve built enduring greatness through professional will and fostered the hedgehog concept. He got the right people in the bus and confronted the brutal facts. He may not made it in Jim Collins' "Good to Great" Level 5 Leader list, he is however in my book, truly an authentic leader.

Steve, you will be missed.

First Glance at Windows 8 Developer's Preview Release Desktop

2011-10-04T11:59:00.000+08:00

I first got a glimpse of this new OS from Microsoft using VirtualBox, and I thought that the only way I can really experience this is to install it on a real machine, and so off to the races. My PC (Intel Core 2 Duo 1.86Ghz, 3GB RAM) has an existing Windows 7 installed, and at first I thought of upgrading it rather than doing a clean install. Upgrading by the way, only works if I'm logged on to Windows 7, however despite repeated tries, it won't allow me to upgrade, maybe because of the remaining hard drive space I have left, 7GB to be exact. I have no recourse but to do a clean install, I have the data backed up anyways. I booted via the installer disc, and installed it on drive C: (drive 1 in Windows 7 speak), the installer gave a warning that I won't be able to boot to the previous OS but I can still access the contents, and it would back up the contents of the OS by naming it Windows.old. The first thing I noticed is that when the installation was about to get done, it was asking me to login using a Windows Live ID, since I don't want to provide one, I opted for a local login (traditional way).

When I logged in to the newly installed Windows 8, I immediately noticed the radical change that Microsoft did, instead of the normal desktop when the user logs in, I was presented with the screen below.

In my opinion Microsoft designed it with the tablet (the current fad right now) in mind, hoping to catch up with Apple's Ipad and others' Android based tablets. To access the desktop, I clicked on the desktop thumbnail, see screenshot below.

As you can see, I installed Chrome, Firefox, Adobe Acrobat Reader, and other applications, especially MS Office 2007 which works just fine with this release of Windows.

When I clicked on the logo, I was redirected to the Start Page (see the above start page image). When I hovered the mouse pointer over the Windows logo, see screenshot below.

Microsoft has an entirely new concept of doing things, the start menu is completely revised. Gone are the days of Program Files, My Document, My Computer....etc..What will I do then if I want to open a particular program if there's no more Program Files if you may ask, Microsoft renamed Program Files to "Apps" and it is accessible when you hover over the Windows Logo and clicked on Search, see screenshot below.

Even Windows Explorer is different, previously when you want to open a file, you just have to right click on the file and choose the program from the menu, now instead of right clicking, Microsoft incorporated the menus right inside Windows Explorer. When you highlight a file, you will notice that the "Open" menu is enabled, see screenshot below.

You can then open the file using the suggested program or click the drop down arrow to choose a different apps.

Another thing worth noticing is that Windows 8 has combined Microsoft Security Essentials and Windows Defender into one, and comes with Windows 8 by default, so you'll have protection of sort from malware.

These changes made by Microsoft are indeed radical, a totally whole new user experience. If you ask me, will previous version users (Windows 7, Vista, and Xp) will welcome this with open arms, your guess is as good as mine.

Feed a Child Build a Nation Project

2011-05-08T21:34:00.001+08:00

The Ateneo Graduate of School of Business is launching a feeding program called "Feed a Child and Build a Nation" for the School Year 2011-2012. This school year's objective is to feed at least 4,000 students in four elementary schools in Quezon City. It only cost Php11.50 to feed a child per day or Php1,500.00 for one year which consists of 130 school days.

Please watch the video for more of the program and details where to send your donations.

Aside from this, our SPLEADER class at AGSB pledged to raise at least Php100,000.00 for the program by selling pouch bags for only Php100.00 each. All proceeds will go directly to the program.

For those interested, just email me your details.

Ubuntu Narty Narwhal Impressions

2011-05-05T20:25:00.002+08:00

I immediately downloaded Ubuntu 11.04 the moment it was released and get my hands and feet wet, replacing Meego that is currently dual booting with Windows 7 on my Lenovo S10-3S. Burned a copy, and then made a bootable USB disk (this is a netbook, no built-in DVD drive). Tried it live first and see how it performs before permanently installing it, I noticed that it was able to detect Windows' wireless adapter setting, unfortunately, it's just that, wireless is not working. I guess the proprietary Broadcom Linux driver is not loaded when using LiveCD. I decided then to install it permanently using the partitions where Meego is installed. This is the part that I thought it was going to be a piece of cake but I thought wrong, I can't install Ubuntu whether I choose "Try Ubuntu" or "Install Ubuntu", I keep on getting absurd error messages e.g. "Only 4MB of disk space remains" that is preventing the installation. Tried it over and over again but ended up just the same, also verified if the md5sum is correct just to make sure.

Gave it a rest, slept over it tried to re-install the moment I wake up, and this time I told myself, why not delete all Linux partitions and make the new partitions for Ubuntu. I created 4 partitions i.e. /boot (ext3), /(btrfs), /home(btrfs) and swap, and voila! the installation proceeded and I was able to complete it. Now comes another issue, as previously mentioned the wireless is not working, it's still not working even if it said it loaded the proprietary driver, in fact I made sure that the wireless switch is turned on, to no avail, worst, every time I boot to Windows, I had to turn on the wireless function using the function key + F5. This happens every time I boot to Ubuntu, using the same key combination does not make the wireless work in Ubuntu.

The new Unity desktop looks fine to me, just to need to familiarize myself, I still prefer the Meego UI anytime though. Another issue for me is boot time, i.e. from the time choosing the OS in the bootloader menu up to the time the desktop is fully loaded, Ubuntu took 1 minute and 22 seconds to fully load in my Intel Atom N550 2GB RAM Lenovo S10-3S netbook while Windows 7 only took 1 minute and 10 seconds to do the same.

Overall, I'm disappointed because if the WiFi is not working then I can't do my usual stuff, I guess I have to wait for Meego 1.2 stable to get released probably last week of May.

FreeSwitch + Bluebox High Availability (HA) Cluster using Heartbeat and DRBD

2011-01-07T23:00:00.027+08:00

I need to test FreeSwitch first in a virtual environment before going into production, the objective is to come up with a highly available FreeSwitch cluster using Heartbeat 2.1 and DRBD 8.2. I'm using CentOS 5.5 for this demonstration, the real production servers will be using RHEL5 and has the following hardware specifications as stated in the systems requirement:

IBM X3650 M2
Intel® Xeon® Processor 2.5 Ghz
12GB DDR3 RAM
8 - 146GB SFF SCSI HDD RAID10
2 Integrated Gigabit Ethernet

Installing FreeSwitch

I'm going to install FreeSwitch using the latest stable version 1.0.6 which you can download from http://files.freeswitch.org, it would be nice if I could have used git unfortunately something is blocking it here in the office hence I couldn't get the files using git.

Off to the races, update the OS first.

# yum -y update

Next, make sure you installed the necessary tools to compile FreeSwitch.

# yum install autoconf automake gcc-c++ git-core libjpeg-devel libtool make ncurses-devel openssl-devel perl cpio expat-devel gettext-devel curl-devel curl

Keep things tidy, cd to /usr/src.

# cd /usr/src

Fetch file using wget

# wget http://files.freeswitch.org/freeswitch-1.0.6.tar.gz

Untar file.

# tar xzvf freeswitch-1.0.6.tar.gz

Change directory name to freeswitch (optional)

# mv freeswitch-1.0.6 freeswitch

Go to freeswitch directory and compile. This will take a while so go ahead, take a break, drink a soda or something, I prefer lemon iced tea :D

# cd freeswitch
# ./configure
# make

+---------- FreeSWITCH Build Complete ----------+
+ FreeSWITCH has been successfully built. +
+ Install by running: +
+ +
+ make install +
+-----------------------------------------------+

# make install

+---------- FreeSWITCH install Complete ----------+
+ FreeSWITCH has been successfully installed. +
+ +
+ Install sounds: +
+ (uhd-sounds includes hd-sounds, sounds) +
+ (hd-sounds includes sounds) +
+ ------------------------------------ +
+ make cd-sounds-install +
+ make cd-moh-install +
+ +
+ make uhd-sounds-install +
+ make uhd-moh-install +
+ +
+ make hd-sounds-install +
+ make hd-moh-install +
+ +
+ make sounds-install +
+ make moh-install +
+ +
+ Install non english sounds: +
+ replace XX with language +
+ (ru : Russian) +
+ ------------------------------------ +
+ make cd-sounds-XX-install +
+ make uhd-sounds-XX-install +
+ make hd-sounds-XX-install +
+ make sounds-XX-install +
+ +
+ Upgrade to latest: +
+ ---------------------------------- +
+ make current +
+ +
+ Rebuild all: +
+ ---------------------------------- +
+ make sure +
+ +
+ Install/Re-install default config: +
+ ---------------------------------- +
+ make samples +
+ +
+ +
+ Additional resources: +
+ ---------------------------------- +
+ http://www.freeswitch.org +
+ http://wiki.freeswitch.org +
+ http://jira.freeswitch.org +
+ http://lists.freeswitch.org +
+ +
+ irc.freenode.net / #freeswitch +
+ +
+ Register For ClueCon: +
+ ---------------------------------- +
+ http://www.cluecon.com +
+ +
+-------------------------------------------------+
# make cd-sounds-install && make cd-moh-install
# make samples

Once done, test to make sure that FreeSwitch is working! The compilation process including fetching sound files took almost an hour to complete, told you to take a break, didn't I? :D

# cd /usr/local/freeswitch/bin

# ./freeswitch

_____ ______ _____ _____ ____ _ _

| ___| __ ___ ___/ ___\ \ / /_ _|_ _/ ___| | | |

| |_ | '__/ _ \/ _ \___ \\ \ /\ / / | | | || | | |_| |

| _|| | | __/ __/___) |\ V V / | | | || |___| _ |

|_| |_| \___|\___|____/ \_/\_/ |___| |_| \____|_| |_|

************************************************************

* Anthony Minessale II, Michael Jerris, Brian West, Others *

* FreeSWITCH (http://www.freeswitch.org) *

* Paypal Donations Appreciated: paypal@freeswitch.org *

* Brought to you by ClueCon http://www.cluecon.com/ *

************************************************************

2010-12-29 00:36:26.546866 [CONSOLE] switch_core.c:1578

FreeSWITCH Version 1.0.6 (svn-exported) Started.

Max Sessions[1000]

Session Rate[30]

SQL [Enabled]

freeswitch@pbx1.cluster.local> sofia status

Name Type Data State

=================================================================================================

internal profile sip:mod_sofia@192.168.136.136:5060 RUNNING (0)

external profile sip:mod_sofia@192.168.136.136:5080 RUNNING (0)

external::example.com gateway sip:joeuser@example.com NOREG

internal-ipv6 profile sip:mod_sofia@[::1]:5060 RUNNING (0)

192.168.136.136 alias internal ALIASED

=================================================================================================

3 profiles 1 alias

To shutdown FreeSwitch type "shutdown" or "..." inside the FreeSwitch shell to turn off FreeSwtich. Afterwards, edit PATH to include path to Freeswitch binaries. Create /etc/profile.d/freeswitch.sh, and add the line "export PATH=$PATH:/usr/local/freeswitch/bin".

# vim /etc/profile.d/freeswitch.sh

Add this line --> export PATH=$PATH:/usr/local/freeswitch/bin

Save and exit.

Make /etc/profile.d/freeswitch.sh executable by:

# chmod 755 /etc/profile.d/freeswitch.sh

Be sure to log off and log on back to see changes in effect.

# echo $PATH

Add FreeSwitch user:

# useradd freeswitch

Change freeswitch user passwd setting:

# vim /etc/passwd

freeswitch:x:500:500:Freeswitch:/home/freeswitch:/sbin/nologin

Change ownership of /usr/local/freeswitch.

# chown -R freeswitch:freeswitch /usr/local/freeswitch

To automatically start FreeSwitch at boot, copy /usr/src/freeswitch/build/freeswitch.init.redhat to /etc/init.d/ and make the following changes to PID_FILE, FS_FILE, and FS_HOME parameters:

# vim /etc/init.d/freeswitch

PID_FILE=${PID_FILE-/usr/local/freeswitch/log/freeswitch.pid}

FS_FILE=${FS_FILE-/usr/local/freeswitch/bin/freeswitch}

FS_HOME=${FS_HOME-/usr/local/freeswitch}

Save and exit.

Next, do the following:

# chmod 755 /etc/init.d/freeswitch

# chkconfig freeswitch on

# chkconfig --list freeswitch

Start FreeSwitch using the rc.d script:

# /etc/init.d/freeswitch start

Wait about 30 seconds for Freeswitch to be up and running, and connect to the server through the console:

# fs_cli

Check SIP status:

freeswitch@internal> sofia status

Check a specific SIP profile:

freeswitch@internal>sofia status profile internal

To exit the console, type "/exit" or "..."; To stop the Freeswitch server and exit the console, type "fsctl shutdown elegant".

Next step would be to clone this machine's current state, this is in order for me to have the second node in a jiffy, I'm using VMWare Workstation by the way.

Configuring DRBD

Make sure you have extra partition for DRBD, mine is on /dev/sda3 on both nodes, it's recommended that they're both have the same disk size.

Install DRBD on both nodes:

# yum install drbd82 kmod-drbd82

Configure DRBD.

# vim /etc/drbd.conf

resource "r0" {
   protocol C;
   disk { on-io-error pass_on; }
   startup { wfc-timeout 30; degr-wfc-timeout 20; }
   syncer { rate 10M; }
   on pbx1.cluster.local {
   device /dev/drbd0;
   disk /dev/sda3;
   address 192.168.136.136:7789;
   meta-disk internal;
   }
   on pbx2.cluster.local {
   device /dev/drbd0;
   disk /dev/sda3;
   address 192.168.136.137:7789;
   meta-disk internal;
   }
}

Save file and exit.

Copy /etc/drbd.conf to node2.

# scp /etc/drbd.conf root@node2:/etc/

Bringing up DRBD.

With DRBD off on both nodes, reboot node1 and watch DRDB during start up, type 'yes' when prompted.

Login back to node1 and do the following to:

# drbdadm create-md r0

Node1 is now in secondary inconsistent state. Next, reboot node2 to bring up DRBD, once startup is complete login back to node2.

# drbdadm create-md r0

We now have to secondaries in incosistent state, it's time to synchronize them, login back to node1 and promote it to primary.

# drbdadm -- --overwrite-data-of-peer primary r0

Watch DRBD sync.

# watch cat /proc/drbd

Once syncing is done, login to node1 and create filesystem on /dev/drbd0 and test DRBD.

# mkfs.ext3 -L r0 /dev/drbd0

We don't have to repeat this on node2, everything we do on node1 is replicated on node2, that's DRBD at work.

Create /replication partition on both nodes, it will be used as mount point.

Installing and Configuring Heartbeat

Install Heartbeat on both nodes.

# yum install heartbeat

This will install heartbeat together with heartbeat-stonith and heartbeat-pils.

Create the following files under /etc/ha.d/ directory:

authkeys

ha.cf

haresources

Next, you can copy sample configuration files from the /usr/share/doc/heartbeat-version.no. directory, or create one from scratch, we'll do the latter.

# cd /etc/ha.d/
# vim authkeys

auth 1

1 sha1 secretkey!

Save and exit.

Change authkeys permission to 600 or else it will not work.

# chmod 600 authkeys

Next, create ha.cf file.

# vim ha.cf

debugfile /var/log/ha-debug

logfile /var/log/ha-log

logfacility local0

keepalive 2

deadtime 30

warntime 10

initdead 120

udpport 694

bcast eth0

auto_failback off

node pbx1.cluster.local

node pbx2.cluster.local

Third, create haresources file.

# vim haresources

pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255

pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch

Save and exit.

Read sample configuration file which explains each parameter for more details. While in the/etc/ha.d/ directory, copy ha.cf, authkeys, and haresources to node2.

# scp authkeys ha.cf haresources root@pbx:/etc/ha.d/

Hold your horses for now, don't start Heartbeat yet, we still have to configure FreeSwitch to use the DRBD, go to primary node and cd to mounted drbd partition i.e. /replication

# cd /replication

1. Backup FreeSwitch installation first! or use VMWare snapshot functionality in case we messed things up.

# tar czvf usr-local-freeswitch.tar.gz /usr/local/freeswitch

2. Untar usr-local-freeswitch.tar.gz in current drbd directory /replication

# tar xzvf usr-local-freeswitch.tar.gz

3. Remove /usr/local/freeswitch.

# rm -rf /usr/local/freeswitch

4. Create symbolic link for freeswitch to /usr/local/freeswitch from drbd directory.

# ln -s /replication/usr/local/freeswitch /usr/local/freeswitch

Repeat steps 1 to 4 on node2.

Now for the finale, it's time to start heartbeat on both nodes, I'll show my heartbeat logs to prove that failover is working.

Check if our FreeSwitch HA cluster is working.

We need to have multiple terminal open to properly monitor the failover i.e. 2 terminals for each node. Monitor ha-debug logs in real time on both nodes by:

# tail -f /var/log/ha-debug

Go to node1, stop heartbeat and see what happens on terminal 2's ha-debug log.

[root@pbx1 ~]# /etc/init.d/heartbeat stop

Stopping High-Availability services:

[ OK ]

[root@pbx2 ~]# tail -f /var/log/ha-debug

heartbeat[9455]: 2011/01/04_11:36:45 info: Received shutdown notice from 'pbx1.cluster.local'.

heartbeat[9455]: 2011/01/04_11:36:45 info: Resources being acquired from pbx1.cluster.local.

heartbeat[9455]: 2011/01/04_11:36:45 debug: StartNextRemoteRscReq(): child count 1

heartbeat[9726]: 2011/01/04_11:36:45 info: acquire all HA resources (standby).

heartbeat[9727]: 2011/01/04_11:36:46 info: No local resources [/usr/share/heartbeat/ResourceManager listkeys pbx2.cluster.local] to acquire.

heartbeat[9455]: 2011/01/04_11:36:46 debug: StartNextRemoteRscReq(): child count 1

ResourceManager[9752]: 2011/01/04_11:36:46 info: Acquiring resource group: pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255

IPaddr[9779]: 2011/01/04_11:36:46 INFO: Resource is stopped

ResourceManager[9752]: 2011/01/04_11:36:46 info: Running /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start

ResourceManager[9752]: 2011/01/04_11:36:46 debug: Starting /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start

IPaddr[9877]: 2011/01/04_11:36:46 INFO: Using calculated netmask for 192.168.136.138: 255.255.255.0

IPaddr[9877]: 2011/01/04_11:36:46 INFO: eval ifconfig eth1:0 192.168.136.138 netmask 255.255.255.0 broadcast 192.168.136.255

IPaddr[9877]: 2011/01/04_11:36:46 DEBUG: Sending Gratuitous Arp for 192.168.136.138 on eth1:0 [eth1]

IPaddr[9848]: 2011/01/04_11:36:46 INFO: Success

INFO: Success

ResourceManager[9752]: 2011/01/04_11:36:46 debug: /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start done. RC=0

ResourceManager[9970]: 2011/01/04_11:36:46 info: Acquiring resource group: pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch

ResourceManager[9970]: 2011/01/04_11:36:46 info: Running /etc/ha.d/resource.d/drbddisk r0 start

ResourceManager[9970]: 2011/01/04_11:36:46 debug: Starting /etc/ha.d/resource.d/drbddisk r0 start

ResourceManager[9970]: 2011/01/04_11:36:46 debug: /etc/ha.d/resource.d/drbddisk r0 start done. RC=0

Filesystem[10038]: 2011/01/04_11:36:46 INFO: Resource is stopped

ResourceManager[9970]: 2011/01/04_11:36:46 info: Running /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start

ResourceManager[9970]: 2011/01/04_11:36:46 debug: Starting /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start

Filesystem[10119]: 2011/01/04_11:36:46 INFO: Running start for /dev/drbd0 on /replication

Filesystem[10108]: 2011/01/04_11:36:46 INFO: Success

INFO: Success

ResourceManager[9970]: 2011/01/04_11:36:46 debug: /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start done. RC=0

ResourceManager[9970]: 2011/01/04_11:36:46 info: Running /etc/init.d/freeswitch start

ResourceManager[9970]: 2011/01/04_11:36:46 debug: Starting /etc/init.d/freeswitch start

Starting freeswitch: [ OK ]

ResourceManager[9970]: 2011/01/04_11:36:47 debug: /etc/init.d/freeswitch start done. RC=0

heartbeat[9726]: 2011/01/04_11:36:47 info: all HA resource acquisition completed (standby).

heartbeat[9455]: 2011/01/04_11:36:47 info: Standby resource acquisition done [all].

heartbeat[10243]: 2011/01/04_11:36:47 debug: notify_world: setting SIGCHLD Handler to SIG_DFL

harc[10243]: 2011/01/04_11:36:47 info: Running /etc/ha.d/rc.d/status status

mach_down[10261]: 2011/01/04_11:36:47 info: Taking over resource group IPaddr::192.168.136.138/24/eth1/192.168.136.255

ResourceManager[10287]: 2011/01/04_11:36:47 info: Acquiring resource group: pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255

IPaddr[10314]: 2011/01/04_11:36:47 INFO: Running OK

mach_down[10261]: 2011/01/04_11:36:47 info: Taking over resource group drbddisk::r0

ResourceManager[10373]: 2011/01/04_11:36:47 info: Acquiring resource group: pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch

Filesystem[10415]: 2011/01/04_11:36:47 INFO: Running OK

mach_down[10261]: 2011/01/04_11:36:47 info: /usr/share/heartbeat/mach_down: nice_failback: foreign resources acquired

mach_down[10261]: 2011/01/04_11:36:47 info: mach_down takeover complete for node pbx1.cluster.local.

heartbeat[9455]: 2011/01/04_11:36:47 info: mach_down takeover complete.

As you can see node2 took over from node1 and is now primary, FreeSwitch is now running on node2.

Double check!

[root@pbx2 ~]# mount

/dev/sda1 on / type ext3 (rw)

proc on /proc type proc (rw)

sysfs on /sys type sysfs (rw)

devpts on /dev/pts type devpts (rw,gid=5,mode=620)

tmpfs on /dev/shm type tmpfs (rw)

none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)

/dev/drbd0 on /replication type ext3 (rw)

[root@pbx2 ~]# /etc/init.d/freeswitch status

freeswitch (pid 10253) is running...

As you can see /dev/drbd0 is mounted and FreeSwitch is running.

Start Heartbeat again on node1 and perform the same test, stop Heartbeat on node2 and see what happens on ha-debug log on node1.

[root@pbx2 ~]# /etc/init.d/heartbeat stop

Stopping High-Availability services:

[ OK ]

[root@pbx1 ~]# tail -f /var/log/ha-debug

heartbeat[10484]: 2011/01/04_11:50:17 info: Received shutdown notice from 'pbx2.cluster.local'.

heartbeat[10484]: 2011/01/04_11:50:17 info: Resources being acquired from pbx2.cluster.local.

heartbeat[10484]: 2011/01/04_11:50:17 debug: StartNextRemoteRscReq(): child count 1

heartbeat[10527]: 2011/01/04_11:50:17 info: acquire all HA resources (standby).

ResourceManager[10551]: 2011/01/04_11:50:18 info: Acquiring resource group: pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255

IPaddr[10590]: 2011/01/04_11:50:19 INFO: Resource is stopped

IPaddr[10615]: 2011/01/04_11:50:19 INFO: Resource is stopped

heartbeat[10528]: 2011/01/04_11:50:19 info: Local Resource acquisition completed.

heartbeat[10484]: 2011/01/04_11:50:19 debug: StartNextRemoteRscReq(): child count 2

heartbeat[10484]: 2011/01/04_11:50:19 debug: StartNextRemoteRscReq(): child count 1

ResourceManager[10551]: 2011/01/04_11:50:19 info: Running /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start

ResourceManager[10551]: 2011/01/04_11:50:19 debug: Starting /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start

IPaddr[10781]: 2011/01/04_11:50:19 INFO: Using calculated netmask for 192.168.136.138: 255.255.255.0

IPaddr[10781]: 2011/01/04_11:50:19 INFO: eval ifconfig eth1:0 192.168.136.138 netmask 255.255.255.0 broadcast 192.168.136.255

IPaddr[10781]: 2011/01/04_11:50:19 DEBUG: Sending Gratuitous Arp for 192.168.136.138 on eth1:0 [eth1]

IPaddr[10752]: 2011/01/04_11:50:20 INFO: Success

INFO: Success

ResourceManager[10551]: 2011/01/04_11:50:20 debug: /etc/ha.d/resource.d/IPaddr 192.168.136.138/24/eth1/192.168.136.255 start done. RC=0

ResourceManager[10874]: 2011/01/04_11:50:20 info: Acquiring resource group: pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch

ResourceManager[10874]: 2011/01/04_11:50:20 info: Running /etc/ha.d/resource.d/drbddisk r0 start

ResourceManager[10874]: 2011/01/04_11:50:20 debug: Starting /etc/ha.d/resource.d/drbddisk r0 start

ResourceManager[10874]: 2011/01/04_11:50:20 debug: /etc/ha.d/resource.d/drbddisk r0 start done. RC=0

Filesystem[10942]: 2011/01/04_11:50:20 INFO: Resource is stopped

ResourceManager[10874]: 2011/01/04_11:50:20 info: Running /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start

ResourceManager[10874]: 2011/01/04_11:50:20 debug: Starting /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start

Filesystem[11023]: 2011/01/04_11:50:20 INFO: Running start for /dev/drbd0 on /replication

Filesystem[11012]: 2011/01/04_11:50:20 INFO: Success

INFO: Success

ResourceManager[10874]: 2011/01/04_11:50:20 debug: /etc/ha.d/resource.d/Filesystem /dev/drbd0 /replication ext3 start done. RC=0

ResourceManager[10874]: 2011/01/04_11:50:20 info: Running /etc/init.d/freeswitch start

ResourceManager[10874]: 2011/01/04_11:50:20 debug: Starting /etc/init.d/freeswitch start

Starting freeswitch: [ OK ]

ResourceManager[10874]: 2011/01/04_11:50:21 debug: /etc/init.d/freeswitch start done. RC=0

heartbeat[10527]: 2011/01/04_11:50:21 info: all HA resource acquisition completed (standby).

heartbeat[10484]: 2011/01/04_11:50:21 info: Standby resource acquisition done [all].

heartbeat[11149]: 2011/01/04_11:50:21 debug: notify_world: setting SIGCHLD Handler to SIG_DFL

harc[11149]: 2011/01/04_11:50:21 info: Running /etc/ha.d/rc.d/status status

mach_down[11165]: 2011/01/04_11:50:21 info: /usr/share/heartbeat/mach_down: nice_failback: foreign resources acquired

mach_down[11165]: 2011/01/04_11:50:21 info: mach_down takeover complete for node pbx2.cluster.local.

heartbeat[10484]: 2011/01/04_11:50:21 info: mach_down takeover complete.

heartbeat[11200]: 2011/01/04_11:50:21 debug: notify_world: setting SIGCHLD Handler to SIG_DFL

harc[11200]: 2011/01/04_11:50:21 info: Running /etc/ha.d/rc.d/ip-request-resp ip-request-resp

ip-request-resp[11200]: 2011/01/04_11:50:21 received ip-request-resp IPaddr::192.168.136.138/24/eth1/192.168.136.255 OK yes

ResourceManager[11221]: 2011/01/04_11:50:21 info: Acquiring resource group: pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255

IPaddr[11248]: 2011/01/04_11:50:23 INFO: Running OK

heartbeat[11306]: 2011/01/04_11:50:23 debug: notify_world: setting SIGCHLD Handler to SIG_DFL

harc[11306]: 2011/01/04_11:50:23 info: Running /etc/ha.d/rc.d/ip-request-resp ip-request-resp

ip-request-resp[11306]: 2011/01/04_11:50:23 received ip-request-resp drbddisk::r0 OK yes

ResourceManager[11327]: 2011/01/04_11:50:23 info: Acquiring resource group: pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch

Filesystem[11368]: 2011/01/04_11:50:24 INFO: Running OK

Check!

[root@pbx1 ~]# mount

/dev/sda1 on / type ext3 (rw)

proc on /proc type proc (rw)

sysfs on /sys type sysfs (rw)

devpts on /dev/pts type devpts (rw,gid=5,mode=620)

tmpfs on /dev/shm type tmpfs (rw)

none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)

/dev/drbd0 on /replication type ext3 (rw)

[root@pbx1 ~]# /etc/init.d/freeswitch status

freeswitch (pid 11145) is running...

Installing Bluebox aka FreePBX3

Since our HA Cluster is already working, installing Bluebox aka FreePBX3 would be a breeze. Bluebox requires MySQL, Apache, and PHP >=5.2. CentOS 5.5 comes with PHP 5.1.6 hence we need to install the EPEL and REMI repository on both nodes.

# rpm -Uvh http://download.fedora.redhat.com/pub/epel/5/i386/epel-release-5-4.noarch.rpm
# rpm -Uvh http://rpms.famillecollet.com/enterprise/5/remi/i386/remi-release-5-7.el5.remi.noarch.rpm

Install MySQL, Apache, and PHP on both nodes.

# yum install mysql-server php php-mysql php-gd php-xml php-mbstring php-pdo php-soap php-common httpd

Once your done installing on both nodes, it's now time to configure MySQL to use DRBD.

1. Go to node1 (primary node) and cd to mounted drbd directory and backup MySQL.

# cd /replication
# tar czvf var-lib-mysql.tar.gz /var/lib/mysql

2. Untar var-lib-mysql.tar.gz in current drbd directory /replication

# tar xzvf var-lib-mysql.tar.gz

3. Remove /var/lib/mysql

# rm -rf /var/lib/mysql

4. Create symbolic link for MySQL to /var/lib/mysql from drbd directory.

# ln -s /replication/var/lib/mysql /var/lib/mysql

Repeat steps 1 to 4 on node2.

Configuring Apache.

Create /replication/web directory on both nodes, this directory will be used by Apache.

Next, login to node1 and edit /etc/httpd/conf/httpd.conf and make the following changes.

# vim /etc/httpd/conf/httpd.conf

Replace all instances of "/var/www/html" with "/replication/web".

Save and exit.

Copy to node2.

# scp /etc/httpd/conf/httpd.conf root@pbx2:/etc/httpd/conf/

It's now time to update /etc/ha.d/haresources on both nodes and add MySQL and Apache to the mix, login to node1 and edit haresources.

# vim /etc/ha.d/haresources

pbx1.cluster.local IPaddr::192.168.136.138/24/eth1/192.168.136.255
pbx1.cluster.local drbddisk::r0 Filesystem::/dev/drbd0::/replication::ext3 freeswitch mysqld httpd

Save and exit, and copy to node2.

# scp /etc/ha.d/haresources root@pbx2:/etc/ha.d/

Start Apache and MySQL on the primary node and test!

Finally, install Bluebox on primary node.

# cd /replication/web
# git clone git://source.2600hz.org/bluebox.git
# cd bluebox
# ./preinstall.sh

While on primary node, add bluebox database then login to mysql prompt and do the following:

mysql> CREATE USER 'bluebox'@'localhost' IDENTIFIED BY 'password';
mysql> GRANT ALL ON bluebox.* TO 'bluebox'@'localhost';
mysql> FLUSH PRIVILEGES;

Open web browser and go to http://server_ip/bluebox and follow on-screen instructions to install Bluebox.

We now have a highly available FreeSwitch Cluster with Bluebox using DRDB and Heartbeat.

Cheers!

Many thanks to Vener who provided his valuable insights.

References:

http://www.fredshack.com/docs/freeswitch.html

http://www.voipkb.com/wiki/index.php/NEW_Install:Linux#CentOS2

UK trip and welcomed by Ondoy on our way back

2011-01-01T15:49:00.004+08:00

This piece is long overdue, we went to UK last 2009, stayed at my brother Jing's flat in Newport, Wales which also sponsored our vacation, many thanks to him. Newport, Wales by the way is a 3 hour bus trip from Heathrow Airport. Guess what was my first British dinner, McDonalds of course but this time it's not good old Big Mac, rather a cold cut sandwich, yes you read that right, McDonalds in UK serves cold cuts :D. We only get to see the sights on our third day, jet lag and all. We went to Cardiff Bay, got to see for the first time an Albatross up close, these birds had a thing for my BigMac, hehe. We also got to see the church where Michael Douglas tied the knot with Catherine Zeta Jones.

I think on our 5th day we get to see London, went to see Buckingham palace first, strolled past St. James Park to see the Queen's guard next, afterwards to Downing Street, yes we got to see where the British Prime Minister lives.

Big Ben was next, then Westminster Abbey, unfortunately we forgot to go where the Fab Four (The Beatles) recorded most of their albums i.e. Abbey Road Studios.

All that walking made us hungry so we had a pretty decent lunch in a Malaysian restaurant, it costs us 8 pounds per head, the food is good by the way. After that off we go to Trafalgar Square then to St. Paul's Cathedral, too bad though because we're not able to go inside the church because it's already past 4PM at that time. Lastly, our London trip will not be complete without seeing the London Bridge, afterwards we called it a day and went back to Newport, Wales.

On our second week, my brother booked us a 5-day vacation to Scotland, we stayed in Tinto Hotel located in Biggar. We get to see Scotland's capital Edinburgh on our second day, Glasgow on our third and our fourth day we went to see the Scottish Borders (whatever that means). I toured Scotland wearing just my Havaianas because of an accident I had in a bathtub (I slipped while taking a shower where I cut my toes).

We went to my other brother Eric's place who lives in Hull, England on our third week, where we stayed for ten days, we got to see my nephew Klien and niece Tintin, last time we saw them was way back 2003 in Manila before they left for UK. I also got the chance to play golf for the first time at the driving range. We went back to Newport after that to spend our remaining days in UK, did some shopping too of course along the way.

We went back to Manila on September 26, 2009, the very same day Ondoy wreaked havoc in Metro Manila, the entire NAIA airport was without electricity, a lot of people were stranded at the airport, I think our flight was the only one allowed to land on that day. We were not able to go home immediately because the entire Metro Manila was flooded, we stayed in a hotel for two days and waited for the flood to subside. When we finally got home, I didn't have an idea as to the extent of the damage done, the first floor aircon was submerged, most of the electronics were floating including two luggages full of electronic gadgets. We spent a week cleaning the house repeatedly to get rid of the terrible smell that comes with flood water. Except for the Ondoy horror, the UK vacation was a blast, I hope to go back there in the near future, everything was nice, people were polite and courteous, they always give way to one another.

MeeGo experience

2010-12-31T16:23:00.009+08:00

I bumped into this new Linux distro sometime last September 2010, I wanted to install an operating system (OS) tailor fit for my Netbook, a Lenovo S10-3s (Intel dual core Atom processor N550 with 2GB RAM) that comes preloaded with Windows 7 Starter. I don't want to wipe out Windows, since I paid for it, so I might as well use it. The Lenovo S10-3s does not have any bare model i.e. there's no OS installed hence I installed MeeGo to dual boot with Windows 7. Before permanently installing MeeGo on the hard drive, I had the option to try it live via USB drive, and next is history. I was impressed at the GUI, it's clearly above the rest, see screenshot below.

Need I say more? The GUI is so simple to use even my wife who's no technology savvy loves it, she gets to do Facebook, watch movies, and play her favorite game called "Frozen Bubble" :D, she's on level 71 already and having a hard time getting passed that level, hehe. Others who see my netbook uttered "that looked like a cellphone UI", they all love the GUI which clearly resemble that of a cellphone.

Of course, this would not be possible without installing the necessary audio/video codecs, and wireless ethernet driver by downloading the source and compiling it, they don't work out of the box by default. This is major turn off for first time MeeGo users, other Linux distros such as Ubuntu already works out of the box. MeeGo says it does not install proprietary softwares/drivers by default, come on guys you can do better than that, if Ubuntu can so could MeeGo, it's purely politics if you ask me, after all they both have the same engine (kernel). I also installed OpenOffice to make my netbook productive.

Aside from the occasional crashes i.e. Flash plugin in Chrome, OpenOffice, I rate my MeeGo experience an 8.5 in a scale of 1 to 10 with 10 being the highest. In my opinion MeeGo should improve on the following in order to gain wider adoption especially for first time users:

Easy upgrade path to newer version! I'm still using version 1 even though version 1.1 was released last October because there was no easy upgrade path, not that I can't do it manually, I simply don't have time to reinstall the newer release and reinstall the softwares, codecs, and drivers again, besides there's no reason for me to upgrade to a newer version. There was no mention of the newer release's advantages if any, heck they did not even make a comparison.
Multi-user support, MeeGo doesn't come with this feature, imagine an OS with no multi-user support. This limitation is fine if MeeGo is only intended for mobile phones, but since MeeGo is supporting so many platforms like In-Vehicle, Netbook, Tablet, and Mobile phones just to name a few, they should have thought otherwise, there should be a distinct feature list between platforms. Perhaps Intel and Nokia is overdoing it, trying to catch up with the competition who's already way ahead.

MeeGo still has plenty of opportunity to catch up with the likes of Android and Apple' iOS if they do it right.

Just my two cents worth!

LG Optimus One promo BROUHAHA

2010-12-31T14:34:00.006+08:00

Last November 13, 2010 LG Philippines held the LG Optimus One 50% off promo (SM Megamall, SM North Edsa, SM Cebu, and Gaisano Davao), sounds exciting isn't it? It's a steal at 50% off even if it's just an entry level Android Froyo smartphone, unfortunately the promo is for one hour only i.e. 2-3PM. How many people do you think can they serve in one hour? Do the math. Like many of SALE event aficionados, I still went to SM Megamall despite knowing it will just be futile. I arrived at 10:30AM, went straight to LG booth located at the 4th level (Cyberzone), the place is already cramped, the queue is already all the way down to the basement, imagine that! See my apprehension, anyways I talked to one of the event organizers cum saleslady, I just wanted to get an assurance in case I decide to fall in line, unfortunately it's either she was that dumb or a freaking liar. She keeps on bragging that there are several cashiers that will simultaneously cater to all the buyers, how can that be, there are only two cash registers at that time or was it one? Moreover she can't give me a straight answer if everybody will be served, she even tried to persuade me to join their Facebook promo, I told her I did not waste my time so that I could join their Facebook promo, I went there to buy the phone at 50% off, nothing more!

This is purely a marketing gimmick, they got the attention they were looking for, unfortunately they also got the ire of everyone who were not able to avail of the promo, the only ones not angry are those who were lucky enough (or should I say UNSCRUPULOUSLY!) to get the phone, I wish LG were more truthful.

I would still buy phone if and only if at a discount! I just wanted to try an Android smartphone, I wish LG would put the phone on sale again, this time with no catch (it's already 2011), newer models are released almost every quarter.

I wonder if Samsung Google Nexus S would make its way here, I want to have one. It's a bare Android Gingerbread phone, meaning there's no third party UI so I can upgrade the OS (rooting) on my own whenever there's a new version release.

I'll keep my fingers crossed! :D

When will Oracle release OpenSolaris 2010.03?

2010-05-18T23:22:00.000+08:00

It's been a long time coming, when will Oracle release OpenSolaris 2010.03, it's already four months behind the original release schedule. There's no news about it, are they going to scrap it?

Ingres High Availability Cluster using DRBD

2009-12-22T16:29:00.000+08:00

I've been painstakingly trying to setup an Ingres database High Availability Cluster using Distributed Replicated Block Device (DRBD) for some time already using CentOS 5.4 as OS, however the experience has been a PITA if you ask me because installing an Ingres HA Cluster is totally different from MySQL's or PostgreSQL's. I've tried so many combinations and most of the time insisting on installing all (system and data files) on the shared drive to make it simple however I can't make it work until I decided to install Ingres system files separately on each member node and just use DRBD shared drive for the rest (data, work, journal, checkpoint).

Here's my test environment setup:

Node1:

FQDN=node1.cluster.example

eth0=192.168.1.2

eth1=172.16.0.2

/dev/drbd0=/dev/sda3=5GB

Node2:

FQDN=node2.cluster.example

eth0=192.168.1.3

eth1=172.16.0.3

/dev/drbd0=/dev/sda3=5GB

Both nodes need to have two nics, take note that eth1 on both nodes will be used as dedicated link between nodes (heartbeat or pacemaker/openais automatic failover but this will not yet be covered here, perhaps in the future).

I'm using CentOS 5.4 with server GUI for this setup, don't forget to include the Development Tools and Development Libraries in your installtion, in case you forgot install them using yum groupinstall "Installation Group Name".

Install DRBD 8.3:

# yum install drbd83 kmod-drbd83

By default /etc/drbd.conf is a blank file, I suggest you read more about DRBD from www.drbd.org to fully appreciate the things it can and can't do.

Here's my drbd.conf:

# please have a a look at the example configuration file in

# /usr/share/doc/drbd83/drbd.conf

global { usage-count no; }

common {

syncer { rate 10M; }

}

resource ingres {

protocol C;

startup { wfc-timeout 0; degr-wfc-timeout 120; }

disk { on-io-error detach; } # or panic, ...

on node1.cluster.example {

device /dev/drbd0;

disk /dev/hda3;

address 192.168.1.2:7788;

meta-disk internal;

}

on node2.cluster.example {

device /dev/drbd0;

disk /dev/hda3;

address 192.168.1.3:7788;

meta-disk internal;

}

We have to test DRBD first and make sure that failover works!

[root@node1]# drbdadm create-md ingres

Where ingres is my resource name;

Then start DRBD ;

[root@node1]# /etc/init.d/drbd start

Type 'yes' when prompted and hit return key.

Check drbd;

[root@node1]# cat /proc/drbd

The output will be similar to this one:

[root@node1 ~]# cat /proc/drbd

version: 8.3.2 (api:88/proto:86-90)

GIT-hash: dd7985327f146f33b86d4bff5ca8c94234ce840e build by mockbuild@v20z-x86-64.home.local, 2009-08-29 14:02:24

0: cs:Connected ro:Secondary/Unknown ds:Inconsistent/Unknown C r----

ns:412 nr:0 dw:412 dr:2093 al:8 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:0

As you can see it's in a secondary and unknown state, we'll fix this later by promoting it as primary but first we need to copy /etc/drbd.conf to node2;

[root@node1]# scp /etc/drbd.conf root@node2:/etc/

Enter password:

Then login to node2 as root and do the configure drbd;

[root@node2]# drbdadm create-md ingres

Then start drbd;

[root@node2]# /etc/init.d/drbd start

[root@node2]# cat /proc/drbd

We now have two secondaries, login to node1 and promote it as primary

[root@node1]# drbdadm -- --overwrite-data-of-peer primary ingres

Watch /proc/drbd in real time and see it being sync'd with node2;

Once syncing is done you now have a two-node manual failover cluster, login to node1 and check;

[root@node1]# cat /proc/drbd

version: 8.3.2 (api:88/proto:86-90)

GIT-hash: dd7985327f146f33b86d4bff5ca8c94234ce840e build by mockbuild@v20z-x86-64.home.local, 2009-08-29 14:02:24

0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r----

ns:412 nr:0 dw:412 dr:2093 al:8 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:0

Now it's time to check if DRBD is really working as it should, our mount point for the DRBD drive for both nodes would be /ha/database, create the directory on both nodes:

# mkdir -p /ha/database

And we need to change the mount point's ownership from root to ingres installation owner which in this case is ingres, do the following on both nodes:

# chown -R /ha/database

To have a working system, first we need to create a filesystem on /dev/drbd0 before mounting it, and ext3 is an excellent choice:

[root@node1]# mkfs.ext3 -L ingres /dev/drbd0

Now try to do the same for node2:

[root@node2]# mkfs.ext3 /dev/drbd0

mke2fs 1.35 (28-Feb-2004)

mkfs.ext3: Wrong medium type while trying to determine filesystem size

The error message occurred because of two things, one because node2 is secondary and two since it's secondary it is read-only!

Now login to node1 then mount /dev/drbd0 to /ha/database and perform some simple test to see if DRBD really works;

[root@node1]# mount /dev/drbd0 /ha/database

Create files and directories on /ha/database e.g.

[root@node1]# touch /ha/database/node1testfile

[root@node1]# mkdir /ha/database/node1testdir

Unmount /ha/database and downgrade node1 to secondary using the following commands:

[root@node1]# umount /dev/drbd0

[root@node1]# drbdadm secondary ingres

[root@node2]# mount /dev/drbd0 /ha/database

[root@node2]# drbdadm primary ingres

List the contents of /ha/database, the file node1testfile and the directory node1testdir should be there and it proves that our DRBD failover setup is working!

Login to node1 and promote it as primary node, it's now time to install Ingres! I prefer to install Ingres on the physical machine and use the install wizard than to remotely install because if you do the latter you need to forward X or configure a response file. A response file is a text file which tells Ingres where to install what, e.g.

II_SYSTEM=/dev/sda3

II_DATABASE=/dev/drbd0

II_TIMEZONE_NAME=GMT8

II_CHARSET=UTF8

II_CHECKPOINT=/dev/drbd0

II_DUMP=/dev/drbd0

II_WORK=/dev/drbd0

II_LOG_FILE=/dev/drbd0

II_LOG_FILE_SIZE_MB=256000

II_ENABLE_SQL92=ON

The response file indicates that I would be installing Ingres system files (binaries and configuration) on /dev/sda3 partition and the rest on /dev/drbd0 shared drive.

Download Ingres database from http://esd.ingres.com (as of this writing it's version 9.3 stable), uncompressed the file using tar;

[root@node1]# tar xzvf ingres-9.3.0-159-NPTL-com-pc-linux-i386.tgz.gz

Change directory:

[root@node1]# cd ingres-9.3.0-159-NPTL-com-pc-linux-i386

Begin the installation:

[root@node1]# ./ingres_install

Follow the on-screen wizard and respond to the dialogs, remember to use /dev/sda3 for the system files and /dev/drbd0 i.e. /ha/database for the rest. Take note of the instance name of the ingres installation, this same instance name will also be used on node2 to make failover work.

Once the installation is done, login as the ingres installation owner i.e. ingres and start the service:

[root@node1]# su - ingres

Then to access the instance you must first source the environment:

[ingres@node1 ~]$ . .ingIIbash

where II is the ingres instance.

[ingres@node1 ~]$ ingstart

Ingres/ingstart

Checking host "node1.example.cluster" for system resources required to run Ingres...

Your system has sufficient resources to run Ingres.

Starting your Ingres installation...

Starting the Name Server...

Allocating shared memory for Logging and Locking Systems...

Starting the Recovery Server...

II_DBMS_SERVER = 44442

Starting the Archiver Process...

Starting DBMS Server (default)...

II_DBMS_SERVER = 46170

Starting Net Server (default)...

GCC Server = 36678

TCP_IP port = I2 (21392)

Starting Data Access Server (default)...

DAS Server = 35011

TCP_IP port = I27 (21399)

Starting the Visual DBA Remote Command Server...

Ingres installation successfully started.

Now it's time to stop Ingres on node1, then unmount /ha/database, and then demote node1 as secondary node:

[ingres@node1 ~]$ ingstop

[root@node1]# umount /dev/drbd0

[root@node1]# drbdadm secondary ingres

Login to node2 then promote it as primary, mount /ha/database and then install Ingres using the same install path used on node1.

[root@node2]# drbdadm primary ingres

[root@node2]# mount /dev/drbd0 /ha/database

Copy the entire ingres-9.3.0-159-NPTL-com-pc-linux-i386 from node1 to node2 then start the installation process:

[root@node2]# cd /path/to/ingres-9.3.0-159-NPTL-com-pc-linux-i386

Again, use the same procedure done on node1 and used the same partitions!

[root@node2]# ./ingres_install

Once the installation is finished, login as the ingres user, source the environment and start ingres:

[root@node2]# su - ingres

[ingres@node2 ~]$ . .ingIIbash

[ingres@node2 ~]$ ingstart

You will see a similar standard output when you started ingres on node1;

Try to access the Ingres database and create a sample database:

[ingres@node2 ~]$ createdb node2testdb

Stop Ingres then unmount /ha/database and demote node2 as secondary.

Login to node1, promote it as primary, mount /dev/drbd0 to /ha/database then login as user ingres and start ingres, check if the database node2testdb created on node2 earlier is on the failover drive i.e. drbd0:

[ingres@node1 ~]$ infodb node2testdb

==================Mon Dec 28 15:19:59 2009 Database Information=================

Database : (node2testdb,ingres) ID : 0x4B31A544 Default collation : default

Unicode enabled : Yes

Default unicode collation : udefault Unicode normalization : NFC

Extents : 5 Last Table Id : 238

Config File Version Id : 0x00060001 Database Version Id : 8

Mode : DDL ALLOWED, ONLINE CHECKPOINT ENABLED

Status : VALID

The Database is not Journaled.

Journals are not valid from any checkpoint.

----Journal information---------------------------------------------------------

Checkpoint sequence : 0 Journal sequence : 0

Current journal block : 0 Journal block size : 16384

Initial journal size : 4 Target journal size : 512

Last Log Address Journaled : <0:0:0>

----Dump information------------------------------------------------------------

Checkpoint sequence : 0 Dump sequence : 0

Current dump block : 0 Dump block size : 16384

Initial dump size : 4 Target dump size : 512

Last Log Address Dumped : <0:0:0>

----Checkpoint History for Journal----------------------------------------------

Date Ckp_sequence First_jnl Last_jnl valid mode

----------------------------------------------------------------------------

None.

----Checkpoint History for Dump-------------------------------------------------

Date Ckp_sequence First_dmp Last_dmp valid mode

----------------------------------------------------------------------------

None.

----Cluster Journal History-----------------------------------------------------

Node ID Current Journal Current Block Last Log Address

------------------------------------------------------------

None.

----Cluster Dump History-----------------------------------------------------

Node ID Current Dump Current Block Last Log Address

------------------------------------------------------------

None.

----Extent directory------------------------------------------------------------

Location Flags Physical_path

------------------------------------------------------------------

ii_database ROOT,DATA /ha/database/ingres/data/default/node2testdb

ii_journal JOURNAL /ha/database/ingres/jnl/default/node2testdb

ii_checkpoint CHECKPOINT /ha/database/ingres/ckp/default/node2testdb

ii_dump DUMP /ha/database/ingres/dmp/default/node2testdb

ii_work WORK /ha/database/ingres/work/default/node2testdb

================================================================================

That's all there is to it!

I don't make any guarantee that this will work for you!

The reason I've done this is because there's no documentation publicly available and we have an upcoming project wherein the aforementioned components will be used i.e. Ingres HA Cluster using DRBD.

If I have more spare time in the future hopefully I can make the failover automatic using Pacemaker/OpenAIS or Heartbeat with preference to the former.

Pagudpud Weekend Adventure

2009-08-18T22:15:00.000+08:00

Last time I was in Pagudpud was way back 1995 when I was still working for a Government Owned Controlled Corporation (GOCC), back then the Saud beach front have only two resorts where guest can stay and if I remember it correctly they are both owned by a prominent politician which as of this writing is still the incumbent lawmaker for Ilocos Norte's 1st district. The beach during that time was full of see weeds which makes it unattractive for us to swim however I still enjoyed the experience because it was my first time to snorkle.

I always wanted to go back there ever since to see what it's like nowadays, so last weekend that wish came true, one of my balikbayan brother in law whom I've not seen in a while is in town and would be leaving again next month for the US and as a treat to him my wife and I decided to take him there and so off we went last Friday evening 14th of August 2009 together with my other brother in-law. We rode the Florida Bus to Pagudpud because it's the only bus line that goes there. The bus left Manila at around 9:30PM and arrived in Pagudpud eleven hours later, the bus did too many stops that is why it took that long. I have no idea where to stay, good thing there's Google, haha, in short we stayed in Polaris, room for two costs P800 for an overnight stay which to my standard is reasonable enough. We already have in mind the places we're going to visit once we get there; we're so lucky that we don't have to look for a rent-a-car because all the tricycles here also doubles as tour guides. First destination was the Patapat Viaduct, it's an elevated road right beside the shoreline going to Cagayan, amazing isn't it?

We also went to Bantay Abot Cave which does not look like a cave but instead resembles more like a hill with a hole in the middle, we also went to the Blue Lagoon cove which the local folks call the Boracay of the north which IMO is overrated.

We visited the farthest places first and the nearest as last and that decision proved wise because it allowed us to complete the tour, if we've chosen otherwise we would not be able to finish because we're dead tired. Last on our itenerary was Kabigan Falls whose picture is shown below, a sight to behold right?

We had to endure a total distance of three kilometers (back and forth and a total of two hours including the time we spent swimming and bathing), walking on uneven ground, climbing hills and crossing water ways, but all of our effort did not go to waste, once we reached the place and bathed in it, all body aches brought about by the long trek were somehow relieved and our energy replenished.

After the tour we had lunch along the way and tried local dishes as our tour guide suggested before heading back to the resort. We chatted for a while and then went to the beach (Saud) and swam our hearts out. The beach looked different from the last time I saw it, no more sea weeds this time, I wonder where they went :D

On our way back to Manila we decided to go to Vigan first to see the Heritage Village (Houses and paved roads built during the Spanisn Colonization) and buy pasalubongs like their famous longanisa and bibingka.

What an eventful weekend! We planned to go back again soon if time permits.

Tita Cory's Funeral

2009-08-05T15:03:00.000+08:00

These are the videos I took when the funeral convoy passed by near our home, as you can see the crowd are noisy chanting Tita Cory's name, the rain is also pouring hard but the crowd did not mind even if majority of them are soaking wet. Also there are several firetrucks lined up along Osmeña Highway which I have no idea at first the reason for their presence but when the truck that is carrying the casket of Tita Cory was about to pass by, they suddenly aimed their water canon about forty five degrees up and then releases the water, I think that is their way of saluting her.

I almost dropped my camera phone because of people are shoving one another to get a better glimpse of Tita Cory one last time before she's laid to rest.

The remaining videos below are the end result of me not being ready to shoot this rare moment.

The last time I saw something like this was during the funeral of her late husband Ninoy, I was still in college back then.

Leaders come and go but only a few like her makes an indelible mark on peoples mind.

Best Pizzas I've ever tasted!

2009-07-28T13:10:00.000+08:00

I was in Davao City for the nth time last week of June 2009 and during my one week stay, me and my colleagues scoured this bustling City for restaurants we have not been to yet. I remember during our flight I was reading Cebu Pacific's in-flight magazine and I chanced upon the article that is highly recommending Picobello Italian Restaurant as places to go to if you love Italian food. We had a hard time finding this place because of wrong assumptions, we assumed that there is only one Gaisano Mall so when we got there we were kind of frustrated because all the people we've asked doesn't know about the restaurant so we decided to proceed to Information to get better results and what do you know there is another branch of Gaisano located in Ilustre, we decided to ride a cab because we're already starving. Again it's disappointing that the aforementioned mall has no floor directory so it took us some time locating the restaurant, it's located at the top floor of Gaisano Ilustre branch, the mall isn't much to look at, it's old and there's not much to see but once you get inside the restaurant the ambiance totally changed. We got there late so there's not much patrons around, we were hoping to avail of the afternoon Buy One Take One Pizza promo but since it's already late evening it's not available anymore but for the heck of it we still tried to ask the waitress if the promo is still available in a pleading/funny manner but I guess she didn't buy our act, haha. We ordered the Quattro Formaggi (four cheese) pizza for all,

I ordered lasagna for myself,

and one of my colleague ordered ribs with rice, the other ordered spaghetti and the last ordered if I remember it correctly, calzone.

All I can say is their pizza is truly mouth watering, it's baked using a brick oven with wood as fuel and I guess that's the secret to its flavor compared to other who are using conventional LPG ovens.

My other favorite pizza is the one my former supervisor at the Bureau of Local Employment - DOLE used to bake, his home made pizza is also out this world!

Open Source Migration Success Story

2009-07-08T23:47:00.000+08:00

Diliman Preparatory School have been using Free Open Source Software (FOSS) for almost three years already for its back office applications however all workstations still uses Windows XP as the primary OS and only this year that Management decided to migrate all workstations to Ubuntu (except the laboratory computers and other workstations that needs to access MS based applications) based on careful deliberations and recommendation from the IT Department as well. The IT Department cited the pros and benefits of using an FOSS based OS like Ubuntu, foremost of which are as follows:

Perennial malware/virus infestations would be minimized/eradicated hence there's no need to procure expensive Antivirus software licenses which are ineffective/useless against malware anyway.
Technical Support Personnel can focus more on other aspect of support such as guiding users on the use of in-house applications instead of almost weekly reformatting the PC's.
Access to the latest and best technologies Open Source has to offer.

We encountered difficulties at first during the implementation because instead of first informing them that there will be user training for the new OS, we proceeded right away with the migration. There are numerous complaints e.g. who gave us the authority to format their machines, who gave the order, where's the memo..stuff like that, but after telling them that the order came directly from the School President and showing them the actual memo, it was like ah okay, if you know what I mean.

After migration, we've conducted several user trainings, introducing the GNOME desktop that came with Ubuntu, their replacement for MS Office applications that is OpenOffice and other things they can do with the new OS. It took us about a week of continuous user training to have them accustomed to the new Desktop.

So to make the long story short, the migration was a success despite the minor glitches encountered along the way, but this would not be possible without the help of the school's top executives who are very supportive of the undertaking.

Mini RSYNC Howto

2009-06-05T09:05:00.000+08:00

RSYNC is a versatile command line tool in Unix/Linux for copying files either remotely or locally.

To copy files from remote web root directory called 'apps' to local web root directory located in 'var/www/html/dps' :

# rsync -avz root@192.168.1.2:/var/www/html/apps/ /var/www/html/dps

To copy files from remote web root directory called 'apps' to local web root directory located in 'var/www/html/dps' using a custom ssh port :

# rsync -avz -e "ssh -p 12345" root@192.168.1.2:/var/www/html/apps/ /var/www/html/dps

If you want to exclude files or directory especially if there are several of them, we need to create to an exclude file and list down the files and directories that you don't want to copy.

# vim exclude.txt

cashier_20090427

grading-20080*

accounting/configurations/*

Save the file and exit vim.

Let me explain the contents of exclude.txt:

The directory cashier_20090427 along with its contents will not be copied.

The '*' in grading-20080* means that there are several directories whose names begins with grading-20080 and along with its contents will not be copied.

The '*' signifies that all the contents of the accounting/configurations/ directory will be omitted during copying.

Important Note:

The --exclude-from switch of rsync uses relative path.

To copy files from remote web root directory called 'apps' to local web root directory located in 'var/www/html/dps' and exclude files and directories defined in 'exclude.txt'.

# rsync -avz --exclude-from 'exclude.txt' root@192.168.1.2:/var/www/html/apps/ /var/www/html/dps

That's all there's it to it.

Cheers!

Install Zenoss on CentOS 5

2009-06-02T17:17:00.000+08:00

This installation of Zenoss was done on a CentOS 5 machine. Zenoss is an Enterprise Monitoring software used to monitor CPU, memory and swap usage, imagine installing mrtg with rrdtool without any hassles.

So let's begin:

1. Prerequisites: You must have mysql5-server running, net-snmp and net-snmp-utils installed in order for Zenoss to run properly.

If those packages are not present, install them using yum.

# yum install mysql mysql-server net-snmp net-snmp-utils

2. Download and install the latest Zenoss rpm package from Sourceforge http://downloads.sourceforge.net/zenoss/zenoss-1.1.2-0.rhel5.i386.rpm

# rpm -ivh zenoss-1.1.2-0.rhel5.i386.rpm

3. Start Zenoss by restarting mysql daemon and zenoss itself

# /etc/init.d/mysqld restart

# /etc/init.d/zenoss start

4. I don't have any iptables rules so I don't have to open any port, so if you're using a firewall you need to open the ports needed by Zenoss to operate. Modify /etc/sysconfig/iptables and add the following:

-A INPUT -i eth0 -p tcp -m tcp --dport 8080 -j ACCEPT

-A INPUT -i eth0 -p tcp -m udp --dport 514 -j ACCEPT

-A INPUT -i eth0 -p tcp -m udp --dport 162 -j ACCEPT

Make sure you restart iptables after saving the changes made.

# service iptables restart

To launch Zenoss, open your browser and type the following in the address box --> http://ip.add.ress.of.zenoss.box:8080

The default username and password is admin and zenoss respectively.

That's all there's to it!

Cheers!

Setting up pfSense as main firewall and Untangle in bridge mode as UTM

2009-05-25T10:31:00.000+08:00

Overview:

pfSense is one of the best if not the best Open Source based firewall out there today and I've been using it since 2005. It's based on FreeBSD and uses OpenBSD's pf (packet filtering), not only that it's lighweight (you can install it even in an old 486 based PC, in fact my first install was using a Pentium 266 MHz MMX with just 256MB of RAM and a 10GB HDD) it's also very easy to use.

Untangle is a Linux based firewall/router and Unified Threat Management Machine (UTM) in one( a jack of all trade box) and is also a new kid on the block. It's also very easy to use caveat though is it's heavy on resources (you need at least 2GB of RAM for a decent performance) but it works best as a filtering device (anti SPAM, web filter, anti Phising and Layer 7 protocol control).

My network is setup such that pfSense is the main perimeter firewall facing the internet and Untangle as UTM as shown in the figure:

As you can see the network has several servers behind pfSense and Untangle e.g., FTP server, Web Server, Mail Server and File Server.

If you don't have yet the pfSense and Untangle iso's you can download it from the following sources:

Download pfSense (version 1.2.2 as of this writing) :

http://www.pfsense.com/index.php?id=22

Download Untangle (version 6.1 as of this writing):

http://www.untangle.com/Downloads/download1

Burn the downloaded ISO's using your favorite burner.

Installing pfSense.

Installing pfSense is pretty much straight forward, all you need to have is a little understanding about slices, also don't be intimidated by its curses based install.

Set your bios to boot from cdrom and load your bootable CD so that we can boot into pfSense on your PC. You will need to have at least two network cards installed in the PC, one for the WAN interface and the other for the LAN interface.

Once the PC booted to pfSense, you will be presented with the FreeBSD bootloader screen, there are 7 options that are listed, just wait for the default option in this case Option (1) to boot up. Take a sheet of paper and write down the initials for the Valid interfaces, you will need them in a moment. Mine are em0 and em1. You will be asked if you want to configure your virtual LAN, just select no if you don't want to setup VLAN's

“Do you want to set up VLAN’s now (y|n)?” Enter “n” to select No.

Then you are asked to

“Enter your LAN interface name”

Enter one from the notes I’ve told you to create earlier, in this case ‘em0’, next you’re asked to:

“Enter your WAN interface name”

Enter ‘em1’ then press the Return key

Since we only have two interfaces just hit the Return key when the system asked about the optional interface.

You will then be asked

“Do you want to proceed? y|n”

Make sure you enter ‘y’ here to proceed or else you have to repeat the whole process.

pfSense is now running in RAM and almost fully functional. If you wish you may plug your LAN interface into a hub or switch and connect via the web interface. pfSense is by default assigned an ip of 192.168.1.1. Next we're going to install pfSense to the hard drive permanently instead of just running it on RAM.

Hard Drive Install

Select option (99) to install pfSense to the hard drive.

This is a curses based install. The install works best if you use an entire hard disk. If there is any data on the disk make sure that you have copied it to another location. Now you can as a rule of thumb accept the default settings that are presented during the curses based install.

Remember to remove the cdrom from the drive when you reboot.

Now we have rebooted and are presented with the “pfsense console setup” for a second time. At this moment you can unplug your monitor cable and manage this firewall via a browser or you could select option 8 and explore via a Shell.

Make sure your computers interface is in the 192.168.1.0 subnet, because 'pfSenses' LAN interface is by default 192.168.1.1.

The defualt username password for the web GUI is 'admin' 'pfsense'.

Now we are going to select System > Setup Wizard > The wizard will guide you through the initial configuration of pfSense.

Enter the following information:

Hostname: pfsense

Domain: local

Primary DNS: 208.67.222.222

Secondary DNS: 208.67.220.220

Notice that I used OpenDNS server's ip addresses as my Primary and Secondary DNS servers respectively and highly recommended instead of using your unpredictable ISP's DNS.

Click Next and enter your correct timeZone then click Next to configure the WAN interface.

I chose static address for the WAN interface since I've have my own Public Static IP address provided by my ISP otherwise it means you have dynamically assigned IP address so just use the default DHCP assigned and then click Next to configure the LAN interface.

For the LAN IP enter 192.168.1.1 and subnet mask /24. then click Next to change the default password.

To effect the changes you will be asked for your new username and password and will be asked to reload your browser.

For this tutorial we'll just use the default firewall rule created by pfSense, for more elaborate rules just visit http://doc.pfsense.org for more information.

Installing Untangle:

Installing Untangle is a walk in the park because of its Graphical User Interface (GUI), just follow the on screen instructions of the wizard, remember though that we're going to use Untangle in bridge mode! so don't configure it as Router! When the installation is done you'll be asked to reboot, remove the Untangle CD when you reboot.

When the system is done rebooting it will automatically start the wizard to help you configure Untangle for use in your network, follow the steps in the wizard. Download and install the following applications to extend the functionality of your Untangle box:

Web Filter
Anti SPAM
Anti Phising
Virus Control
Protocol Control

For more information regarding Untangle just visit this URL -> http://wiki.untangle.com/index.php/Untangle_Server_User's_Guide

Network Configuration:

As you can see in the network diagram shown above, the Untangle box is behind pfSense which is our main firewall, Untangle must be in bridge mode (Untangle will not perform any routing function). The LAN interface of pfSense should be connected to Untangle's WAN (external) interface ( you can just use a cross-over cable for this or an extra switch if you have one) and the second interface of Untangle will then be bridged to its external interface hence Untangle will only have one ip address i.e. in my case 192.168.1.9.

Now we're ready to connect clients behind Untangle to see if everything works. Try to configure Untangle's web filter to block porn sites then try to surf a porn site from behind Untangle, if you see the default stop sign HTML page then it means you've configured and connected the cables correctly, if you just see a plain text block message then you have to interchange the cable connection of the Untangle box.

That's all there's to it!

Cheers!

Build a Chat Server using OpenFire (XMPP/Jabber based messaging Server) on CentOS 5

2009-05-21T09:16:00.000+08:00

Installing OpenFire on CentOS 5.3

Download and install the openfire rpm 3.6.4 (as of this writing).

# rpm -ivh openfire-3.6.4.i386.rpm

Openfire has an embedded database that you can use in case you don’t have MySQL installed, I prefer using MySQL though, I also tried using the embedded database, works just fine too.

Create the database using mysql command line.

a. database name = openfire
b. database user = openfire
c. dbpassword = password

$ mysql -u root -p

Enter password:

mysql -> create database openfire;

mysql -> GRANT ALL ON openfire.* TO openfire@localhost IDENTIFIED BY "password";

\q <-- exit mysql

Go to openfire’s directory

# cd /opt/openfire/resources/database

Import the sql file to openfire database

# cat openfire_mysql.sql | mysql -u root -p openfire

Enter password:

Reload the database:

# mysqladmin -u root -p reload

Enter password:

We’re almost done, now start openfire.

# cd /opt/openfire/bin

# ./openfire.sh

Then launch your browser (Firefox) and enter this at the url box -> https://localhost:9091 then follow the procedure on screen and complete the installation process.

Verify English is selected and click Continue
Click Continue for the Server Settings
Verify Standard Database Connection is chosen and click Continue
Set the following Database Settings (then click Continue):
Database Driver Presets: MySQL
Database URL: jdbc:mysql://localhost:3306/openfire
Username: openfire
Password: password

Launch Pidgin or Spark IM clients and create/register an account with the server, to send a message to user1, don’t forget to include the server hostname e.g. user@example.org

That’s all there is to it!

Cheers!

Upgrading to PHP 5.2.9 and MySQL 5.1 on CentOS 5.3

2009-05-16T11:13:00.000+08:00

One of our client has several CentOS 5.3 servers wherein the default PHP and MySQL packages version are installed and our developers requested me to upgrade to the latest versions for the aforementioned packages because apparently some of the functions they are using on PHP is only available on PHP 5.2.x series. However PHP 5.2.9 is not officially supported as well as MySQL Server 5.1 so use it at your own risk, but so far we have not yet encountered any unexpected issues with the aforementioned respective versions.

After much Googling I've found out that PHP 5.2.9 and MySQL Server 5.1 is available on remi's repository so let's start by installing the repo files, we need to download two repo files to use remi's repo.

# wget http://download.fedora.redhat.com/pub/epel/5/x86_64/epel-release-5-3.noarch.rpm

# wget http://rpms.famillecollet.com/enterprise/remi-release-5.rpm

Then install the files.

# rpm -Uvh epel-release-5-3.noarch.rpm remi-release-5.rpm

Take note that I used the x86_64 base because I'm using a 64 bit Operating System (OS), if you're using an i386 OS then use the corresponding platform.

First I've removed php and others like php-mysql, php-mbstring and php-mcrypt which are all based on version 5.1.6.

# yum remove php php-mysql php-mbstring php-mcrypt

Second, I've updated MySQL using the remi repo and backup your existing databases in case something goes awry.

# yum update mysql*

Notice that I've used an asterisk so that all mysql related applications will be updated/upgraded.

Then finally I've reinstalled php including eaccelerator and the suhosin patch to harden php.

# yum install php php-mysql php-mbstring php-mcrypt php-eaccelerator php-suhosin

There are several ways to check the installed version of php and mysql, i.e. either use rpm or yum, but my preferred method is the following:

To check PHP's version:

# php --version

PHP 5.2.9 (cli) (built: Feb 27 2009 14:42:58)

As you can see php version is 5.2.9 with eaccelerator and suhosin patch.

To check MySQL Server's version, login to mysql shell:

# mysql -p

Enter Password:

Welcome to the MySQL monitor. Commands end with ; or \g.

Your MySQL connection id is 3

Server version: 5.1.34 MySQL Community Server (GPL) by Remi

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql>

That's all there's to it.

Cheers!

Will Oracle GPL ZFS now that they own Sun Microsystem?

2009-05-10T20:49:00.000+08:00

Will Oracle GPL ZFS now that they own Sun Microsystem? My gut tells me they will, but if I'm wrong they will eventually drop Oracle Unbreakable Linux and instead focus on Solaris but by doing so will make the Linux Community frown.

What do you think?

My Davao Experience

2009-05-07T13:10:00.002+08:00

I've been to Davao City several times already, not to sightsee but to work. The duration of my stay was brief; the latest if I’m not mistaken will last only for five (5) days. One of the pleasant things you can associate this bustling city is that there are no traffic jams and the air is clean unlike Metro Manila where you can see the thick smog covering the metropolis.

With it comes to food, they are not that far behind, you can have a hearty meal with your favorite softdrink to boot for as low as P50.00, Imagine eating pork barbeque as low as P5.00 per stick, your P50.00 will go a long way. The third time I was here we went straight to a McDonald outlet direct from the airport because we’re so starved because our flight was delayed for 3 hours and all the airline can provide is of course their on-board games, imagine that! I ordered two sets of chicken fillet value meal only to find out later that their version of the chicken fillet value meal was a little different or should I say big different, haha, the chicken fillet was so big that I wasn’t able to gobble up the second order. Can you imagine the different serving size of the chicken fillet, only in Davao! I was so used to seeing the tiny winy chicken fillet when you order here in Manila. Another great value for your money eating place I’ve been to is Banok’s, their charcoal grilled chicken (lechon manok) is one of the best!

One of the unique places here I've been to is a place called "Kasagingan", it is Davao’s local version of Starbucks. The term "Kasagingan" translated to English literally means The Banana Place, guess what restaurant's theme is, the Chimpanzee of course! My officemates and I usually go this place just to make use of their free internet access and of course savor the taste of their unique menu most of which are based on what else, Banana. The place is decorated by lots of frame mounted posters and there’s this one poster which caught my attention the most, the satire’s original content is written in Bisaya (local dialect here in Davao), the translation goes like this:

“You’ve been staying here for some time now and have been using our free internet service without ordering any food but you still have the guts to whine because you suddenly lost internet connection! And that is not even enough you called up the waiter and asked to bring you a cup of hot water and then brought out your own Tea Bag and make your own tea, HUMANS!”

I’m not sure if my translation was good enough to reflect the author’s message and for that I apologize. Anyways I think they would be an instant hit in case the owners decide to venture or put up a franchise in Metro Manila.

There are also very distinct establishment names only found in Davao like Kini Rogers (remember American Country Singer Kenny Rogers?) and Mallengke (combining Mall and Palengke which is the Tagalog word for market) just to name a few.

Other places of interests that someday I hope I can visit are Samal Island and Mount Apo and that is if I have time.