Redhat network

17/05/2014

**Internet Protocol is a set of technical rules that defines how computers communicate over a network. There are currently two versions: IP version 4 (IPv4) and IP version 6 (IPv6).

19/02/2014

****vmware***
VMware, Inc. is an American software company that provides cloud and virtualization software and services, and was the first to successfully virtualize the x86 architecture. Founded in 1998, VMware is based in Palo Alto, California, USA. In 2004 it was acquired by and became a subsidiary of EMC Corporation.
VMware's desktop software runs on Microsoft Windows, Linux, and Mac OS X, while its enterprise software hypervisors for servers, VMware ESX and VMware ESXi, are bare-metal hypervisors that run directly on server hardware without requiring an additional underlying operating system

19/02/2014

The Cisco 3750 range has been around for many years now, and has a vast following. The 3750-X is the new kid on the Cisco block, and it combines plenty of stuff that will be familiar to users of its predecessors with some funky new features that are clearly a step forward.

The device comes in a number of flavours – between 12 and 48 ports, with or without Power over Ethernet. We had the 48P, which is the PoE variant of the 48-port device. Now, the traditional 3750 had four 1Gbit/s SFP ports in addition to the 48 10/100/1000 copper ports; the 3750-X instead has a slot into which you can slot either a four-port 1Gbit/s SFP daughter-board or a two-port 10Gbit/s alternative. In these days when 10Gbit/s Ethernet is almost a commodity, I can certainly see myself making use of the latter both for uplinking to core switches and for hooking into the 10Gbit/s ports that are becoming prevalent on blade servers.

Alongside the port combinations, there are three software installs. The LAN Base software is a layer-2 only software image, and quite frankly I wouldn't ever expect to buy one of these if I only wanted layer-2 functionality. More sensible is the IP Base image which makes the device a proper Layer-3 routing switch, albeit with a limited selection of routing protocols. At the top is the IP Services image, which makes the unit a full-blown router (just like its ancestors – two of my BGP-shouting WAN routers are actually 3750Gs, in fact). The main market will of course be for the IP Base version.

19/02/2014

Millions of pieces of malware and thousands of malicious hacker gangs roam today's online world preying on easy dupes. Reusing the same tactics that have worked for years, if not decades, they do nothing new or interesting in exploiting our laziness, lapses in judgment, or plain idiocy.

But each year antimalware researchers come across a few techniques that raise eyebrows. Used by malware or hackers, these inspired techniques stretch the boundaries of malicious hacking. Think of them as innovations in deviance. Like anything innovative, many are a measure of simplicity.

04/11/2013

***Network Classes:** Very important Topic

The concept of network classes is a little obsolete as subnets are now used to define smaller networks using CIDR (Classless Inter-Domain Routing) as detailed above. These subnets may be part of a class A, B, C, etc network. For historical reference the network classes are defined as follows:

=> Class A: Defined by the first 8 bits with a range of 0 - 127.
First number (8 bits) is defined by Internic i.e. 77.###.###.###
One class A network can define 16,777,214 hosts.
Range: 0.0.0.0 - 127.255.255.255
=> Class B: Defined by the first 8 bits with a range from
128 -191
First two numbers (16 bits) are defined by Internic i.e. 182.56.###.###
One class B network can define 65,534 hosts.
Range: 128.0.0.0 - 191.255.255.255
=> Class C: Defined by the first 8 bits with a range from
192 - 223
First three numbers (24 bits) are defined by Internic i.e. 220.56.222.###
One class B network can define 254 hosts.
Range: 192.0.0.0 - 223.255.255.255
=> Class D: Defined by the first 8 bits with a range from
224 - 239
This is reserved for multicast networks (RFC988)
Range: 224.0.0.0 - 239.255.255.255
=> Class E: Defined by the first 8 bits with a range from
240 - 255
This is reserved for experimental use.
Range: 240.0.0.0 - 247.255.255.255

04/11/2013

Linux TCP/IP Network Configuration Files:-

/etc/resolve.conf :--> List DNS servers for internet domain name resolution.
/etc/hosts :--> Lists hosts to be resolved locally (not by DNS).
/etc/nsswitch.conf :--> List order of host name search. Typically look at local files, then NIS server, then DNS server.
/etc/sysconfig/network :--> Specify network configuration. eg. Static IP, DHCP, NIS, etc.
/etc/sysconfig/network-scripts/ifcfg-device :--> Specify TCP network information.

04/11/2013

***Market and future of Linux and Unix***

According to International Data Corp. (IDC).Linux has grown faster than any other server OS over the past few years. Linux user base is estimated to be about more than 25 million machines, compared to 5.5 million for combined UNIX installations.
Linux is gaining popularity because of its application in embedded technologies, free and easily availability. To compete with Linux, vendors such as HP, IBM, Sun are making customized UNIX with graphical user interface and user friendly interface which is also compatible with Linux. The main UNIX vendors--IBM, Sun, and Hewlett-Packard are already putting Linux interoperability features into future releases of AIX, Solaris, and HP-UX.

04/11/2013

***Solaris vs GNU/Linux – A functional comparison***

With Solaris 11 being pretty much “Free” (well, for non-commercial use anyhow) I have been sitting on the fence wondering which I should use for my home file server. While there isn’t really a need to have any sort of “end-all be-all” operating system (each has a valid use) when choosing an os for a home server you dont really want to run a ton of expensive hardware so you really want something which can do it all.

As far as GNU/Linux and Solaris are concerned each has its pro’s and cons, and so I was hoping to put together a comparison list, and perhaps through a collaborative effort help the community and myself put together a really great pro/con list.

The Killer features for each OS would be:

Solaris: ZFS, Zones, and dtrace

GNU/LINUX: Virtualization, firewall, applications available.

Below you will find a breakdown of each feature/functionality and who has the stronger footing.

Service management (+ to Solaris for its SMF)

*Linux

Relies on sys-V style init scripts
Can be cumbersome to manage any sort of interdependence as all you can do is change the start order (each file is numbered).
sys-V init scripts can become complex shell commands and there is no real “standard” on how to write them.
No automatic “restarter” if the service dies unexpectedly.
No automatic service log files (if you want the service to log you need to configure this in your startup script or the service it is starting.

*Solaris

SMF makes it incredibly simply to say that A depends on B by simply adding a dependency to the XML definition.
Almost anything can become a Solaris service as long as you code up the XML.
Standardized XML interface (you include the necessary sections in the XML document, SMF does the rest).
Automatic logs for any SMF service without you doing anything.

Fault Management (+ to Solaris for FMA)
Solaris

Fault Management Architecture (FMA)
Provides automated diagnosis of faulty hardware, and can take proactive measures to correct.

**Linux

Some individual components can do FMA functions such as mdadm emailing issues but at a global system level this is a lot more difficult to implement then under Solaris with FMA

File system/disk management (+ to Solaris for ZFS)

This was actually a tough call because Linux offers the ability to reshape a raid 5 array which is some pretty cool technology. Flip side, who among us has not had some sort of linux mdadm related issues (kernel upgrade breaks array, etc).

**Linux

Overly complicated (LVM, mdadm, dmraid….)
You need to create the storage (mdadm), then the file system on top (mkfs), usually using LVM to give you flexibility in the future.
numerous commands, mdadm is responsible for the array, while lvm is responsible for the storage layer, and the file system sits on top of this.
fsck
mdadm can reshape raid 5 arrays making it possible to add a single drive to an existing array.
While you can run ZFS as a kernel module now (not FUSE) any kernel upgrade would break your ZFS module. Depending on your distribution you would then have to rebuild it from sources.

**Solaris

Two commands (zpool to manage the pool, and zfs to manage the file systems).
Existing pools can be expanded, but only by adding similar redundant units. Basically if you have a raidz pool built on a single vdev it can expanded but only by adding a new vdev with similar redundancy (raidz in this case). You can not add a single drive to ‘reshape’ the array, and you should avoid mixing redundancy levels.
The limits of ZFS are sufficiently high you will never reach them.
Any subsequent Solaris upgrade always leaves you in a “SAFE” position meaning Solaris upgrades will never update your zpool/zfs versions automatically.
No fsck, the system is always in sync since its based on COW (copy on write).

Actual implementation details to hopefully show the differences:
Linux

mdadm to create the array (/sbin/mdadm –create –verbose /dev/md0 –level=5 –raid-devices=3 /dev/hdb1 /dev/hda1 /dev/hdf1).
pvcreate to create the physical layout on LVM (/sbin/pvcreate /dev/md0).
vgcreate to create the volume group (/sbin/vgcreate lvm-raid /dev/md0).
mkfs to create the file system (/sbin/mkfs.ext4 /dev/lvm-raid/lvm0).
Now you have your file system, you need to mount it (or automount via fstab)

**Solaris

zpool to create the array (zpool create -f tank raidz c7t1d0 c7t2d0 c7t3d0 c7t4d0 c7t5d0 c8t0d0 c8t1d0) Note how Solaris wants DISKS not partitions
zfs to create the file system (zfs create tank/myfilesystem)
That is it, you are done, /tank/myfilesystem is auto-mounted and will auto-mount on the next boot up.

Community support (+ Linux).

Virtualization (+Linux for KVM)

**Linux

Linux has KVM and XEN, both very powerful and incredibly easy to use.
Ability to “remote connect” to a KVM/XEN server via Virtual Machine Manager (http://virt-manager.org/)

**Solaris

Solaris use to have Linux Zones, but this was discontinued in Solaris 11 leaving Virtualbox as the only opensource VM solution.
VirtualBox isn’t easy to have things like having a VM start once the system is loaded.
Ability to run Virtual Box in a zone opens up some interesting doors such as enforcing the Solaris ”Fair scheduler” on your VM’s.
VirtualBox isn’t as robust as the Linux offering. You need to install additional software, and use odd management tools (VboxManage).

System Upgrades (+ to Solaris for boot environments)

**Solaris

Any upgrade to the CORE OS results in a ZFS snapshot being made before the upgrade. if you need to revert to your pre-upgrade system state you simply select it on the boot menu.

**Linux

Upgrading Linux has gotten cleaner over the years but is no way near the the level of boot-able ZFS snapshots

Licensing (+ to GNU/Linux for using GPL)

**Linux

GNU/Linux follows a very liberal “fair use” license (GPL). This license is so generous it allows things like centos and Scientific Linux to rebuild Redhat legally.

**Solaris

CDDL – incompatible which the GPL which prevents some of the Solaris technologies from being incorporated into LINUX directly. Case in point zfs on Linux which has the SPL (Solaris Porting Layer) to separate the GPL/CDDL code.

Package Management (+ to Linux )

Most Linux distro’s have better repositories and a better system then IPS.
Security (+ to Linux)

The only way to get updates for Solaris 11 is to wait for a full release from Oracle, or have a support contract.

Under Linux, security updates are always available as they are released.

04/11/2013

Q. Why the name UNIX?

Ans --> Bell Labs in the period 1968-1969 was involved in the Multics project which was supposed to provide a convenient interactive computing service to the entire community. This OS failed because Multics could not support many users at an exorbitant cost.

During 1969 an alternative was being worked on, and in 1970 this alternative was given the name of UNIX, a pun on the name Multics since UNIX can support multiple users and multiple processes. The name was coined by Brian Kernighan.

28/10/2013

Practical room

28/10/2013

19/10/2013

How to setup Boot Password (Grub)

Even before the operating system is booted, GRUB enables access to file systems. Users without root permissions can access files in your Linux system to which they have no access once the system is booted. To block this kind of access or prevent users from booting certain operating systems, set a boot password.
As the user root, proceed as follows to set a boot password:

At the root prompt, encrypt the password using grub-md5-crypt:

# grub-md5-crypt
Password: ****
Retype password: ****
Encrypted: $1$lS2dv/$JOcHkLosIn7CJk9xShzzJVw/

Paste the encrypted string into the global section of the file menu.lst:

gfxmenu (hd0,4)/message
color white/blue black/light-gray
default 0
timeout 8
password --md5 $1$lS2dv/$JOcHkLosIn7CJk9xShzzJVw/

Now GRUB commands can only be executed at the boot prompt after pressing P and entering the password. However, users can still boot all operating systems from the boot menu.

To prevent one or several operating systems from being booted from the boot menu, add the entry lock to every section in menu.lst that should not be bootable without entering a password. For example:

title linux
kernel (hd0,4)/vmlinuz root=/dev/sda7 vga=791
initrd (hd0,4)/initrd
lock

After rebooting the system and selecting the Linux entry from the boot menu, the following error message is displayed:

must be Authenticated

Press Enter to enter the menu. Then press P to get a password prompt. After entering the password and pressing Enter, the selected operating system (Linux in this case) should boot.