PROCEDURE OF MIRRORING

PROCEDURE OF MIRRORING

e.g. root disk=c0t0d0 mirror disk=c0t1d0

1) Create 7th slice (e.g c0t0d0s7 100mb)

2) prtvtoc /dev/rdsk/c0t0d0s2 | fmthard -s - /dev/rdsk/c0t1d0s2

3) metadb -afc 3 c0t0d0s7 c0t1d0s7

4) metadb (to check whether metadb is created or not)

5) metainit -f d11 1 1 c0t0d0s0
metainit -f d12 1 1 c0t1d0s0

metainit -f d21 1 1 c0t0d0s1
metainit -f d22 1 1 c0t1d0s1

metainit -f d31 1 1 c0t0d0s3
metainit -f d32 1 1 c0t1d0s3

metainit -f d41 1 1 c0t0d0s4
metainit -f d42 1 1 c0t1d0s4

metainit -f d51 1 1 c0t0d0s5
metainit -f d52 1 1 c0t1d0s5

metainit -f d61 1 1 c0t0d0s6
metainit -f d62 1 1 c0t1d0s6

6) metastat -p ( to check)

7) metainit d10 -m d11
metainit d20 -m d21
metainit d30 -m d31
metainit d40 -m d41
metainit d50 -m d51
metainit d60 -m d61

8) Take Backup of system and vfstab files

cp -p /etc/system /etc/system.org
cp -p /etc/vfstab /etc/vfstab.org

9) metaroot d10 (Modifies /"root" entry in vfstab)
lockfs -af

It will do changes in /etc/system file & /etc/vfstab file

In vfstab file /dev/md/dsk/d10 will appear & make manual changes to
replace /dev/dsk/c0t0d0s1 by /dev/md/dsk/d20.

10) Enter the following line in system file

set md:mirrored_root_flag=1


11) init 6

12) dumpadm -d swap

13) metattach d10 d12
metattach d20 d22
metattach d30 d32
metattach d40 d42
metattach d50 d52
metattach d60 d62

14) To check mirroring progress give

# metastat -i | grep %


15) eeprom "use-nvramrc?=true"

eeprom boot-device="rootdisk rootmirror net"

eeprom "nvramrc=devalias rootdisk /pci@0/pci@0/pci@2/scsi@0/disk@0,0 devalias rootmirror /pci@0/pci@0/pci@2/scsi@0/disk@1,0"

Solaris IP Multipathing

I've setup lots of machines for IPMP (IP Multipathing), so I thought I would share a simpler step by step approach.

• Get 2 NIC's in your machine (most of sun servers have multiple NIC's built-in). Not compulsory to have same type of NIC, but it's important that they have the same speed capability.

• Obtain 3 IP addresses in the same local LAN or (VLAN) segment. In IPMP there are 2 fixed (or private) addresses and 1 floating (or public) address. The 2 fixed addresses I refer to as internal. The 1 floating addresses is the external one. If one of the NICs detects link failure, the address tied to that NIC fails over to the working NIC. When the NIC comes back up, the address fails back to its original home. Determine right now which will be your internal IPs and which will be your external. I recommend keeping the same convention for all Multipath machines, no matter what convention you choose. Here are two typical conventions

1. The 1st IP is floating
2. Next 2 IPs are fixed
   

• Edit /etc/hosts with your 3IPs. Example:

192.168.200.11 host loghost #External IP
192.168.200.12 host-e1000g0 # Internal IP for NIC 1
192.168.200.13 host-nxge0 # Internal IP for NIC 2

• Configure network interfaces.

You need to configure the secondary interface and make it have a unique ether address that is persistent across reboots.
# eeprom 'local-mac-address?=true'

• Setup hostname.* files.

You can pretty much copy these two files as is and just modify them slightly to fit your naming conventions in the same way that you setup the /etc/hosts file above.

/etc/hostname.e1000g0
host-e1000g0 netmask + broadcast + group ipmp0 deprecated –failover up \
addif host netmask + broadcast + failover up

/etc/hostname.nxge0
host-nxge0 netmask + broadcast + group ipmp0 deprecated –failover up

• make it active

# ifconfig –a unplumb; ifconfig lo0 plumb; svcadm restart physical

Solaris Zones

Introduction

Zones are virtual process environments in Solaris 10 that allows system administrator to isolate process in one zone from processes in another zone. Zones have their own (chroot) storage environment which again separates data between zones.
This post is an introduction into Solaris zones. Kindly read the Solaris zones commands quick reference for actual commands involved in managing zones.

Zone Types

Type	Abbr.	Definition
Container		Zone with resource management controls
Global zone	GZ	The base OS installation. Is always present.
Local zone	LZ	Any zone created manually by the administrator. Aka. non-global zone
Non-global zone	NGZ	Any zone created manually by the administrator. Aka. local zone

Zone Storage Models

• Sparse root model :- Any zone that shares any of the /lib, /platform, /sbin or /usr directories with the GZ is called a sparse root model zone.
• Whole root model :- Any zone not sharing directories with the GZ is called a whole root model zone.

Zone Features

1. Granularity :- Zones can run on any number of available CPUs and amount of available memory
2. Isolate :- Run the same application with different configurations
3. Security :- Hacking one zone does not compromise applications running in other zones
4. Transparency :- Applications do not need to be recompiled to run in zones (except for some privileged operations)
5. Virtualization :- Hide configuration information from applications
6. Memory capping :- Manage the memory usage of zones
7. Dynamic resource pools :- Assign CPUs to zones
8. Fair share scheduler :- Grant a zone minimal CPU usage

Zone states

• A running zone cannot be brought back to the ready state. A halt on a running zone will bring the zone back to the installed state.
• A zone doesn't have to be booted into the ready state before it can enter the running state. Issuing the zoneadm -z zone boot command on a zone in the installed state will have the zone be booted into the running state

Directories
/etc/zones - Zone configuration files

Zone daemons and software package parameters
Daemons

1. zoneadmd : Responsible for booting and shutting down zones
2. zsched : Keeps track of kernel threads belonging to zones

Software package parameters

1. SUNW_PKG_ALLZONES - Determines the type of zone in which a package can be installed
2. SUNW_PKG_HOLLOW - Determines the visibility of the package in a zone
3. SUNW_PKG_THISZONE - Determines if the package must be installed in the current zone only

Commands

/usr/sbin/zlogin : Log in to zones
/usr/sbin/zoneadm : Administer zones
/usr/sbin/zonecfg : (re)Configure zones
/usr/bin/zonename : Print current zonename

Zone Configuration

• Basic zone configuration
  root@srv# zonecfg -z zone
  zonecfg:zone>create
  zonecfg:zone>set zonepath=/zones/zone
  zonecfg:zone>set autoboot=true
  zonecfg:zone>set pool=zone_pool
  
• Storage Configuration
  UFS
  zonecfg:zone>add fs
  zonecfg:zone:fs>set dir=/ufs
  zonecfg:zone:fs>set special=/dev/dsk/c1t1d0s5
  zonecfg:zone:fs>set raw=/dev/rdsk/c1t1d0s5
  zonecfg:zone:fs>set type=ufs
  zonecfg:zone:fs>set options=rw
  zonecfg:zone:fs>end
  LOFS
  zonecfg:zone>add fs
  zonecfg:zone:fs>set dir=/lofs
  zonecfg:zone:fs>set special=/global/dir
  zonecfg:zone:fs>set type=lofs
  zonecfg:zone:fs>end
  ZFS file system
  zonecfg:zone>add fs
  zonecfg:zone:fs>set dir=/zfs
  zonecfg:zone:fs>set special=pool/data
  zonecfg:zone:fs>set type=zfs
  zonecfg:zone:fs>end
  ZFS dataset
  zonecfg:zone>add dataset
  zonecfg:zone:dataset>set name=pool/dataset
  zonecfg:zone:dataset>end
  ZFS volume
  zonecfg:zone>add device
  zonecfg:zone:device>set match=/dev/zvol/dsk/pool/volume
  zonecfg:zone:device>end
  zonecfg:zone>add device
  zonecfg:zone:device>set match=/dev/zvol/rdsk/pool/volume
  zonecfg:zone:device>end
  Inherit-pkg-dir
  zonecfg:zone>add inherit-pkg-dir
  zonecfg:zone:inherit-pkg-dir>set dir=/opt
  zonecfg:zone:inherit-pkg-dir>end
  
• Network configuration
  zonecfg:zone>add net
  zonecfg:zone:net>set physical=e1000g0
  zonecfg:zone:net>set address=172.16.20.12/24
  zonecfg:zone:net>end
  
• Device configuration
  zonecfg:zone>add device
  zonecfg:zone:device>set match=/dev/rmt/*
  zonecfg:zone:device>end
  
• Resource configuration
  Fair Share Scheduler (FSS)
  zonecfg:zone>add rctl
  zonecfg:zone:rctl>set name=zone.cpu-shares
  zonecfg:zone:rctl>add value (priv=privileged,limit=20,action=none)
  zonecfg:zone:rctl>end
• Extra attributes
  zonecfg:zone>add attr
  zonecfg:zone:attr>set name=creator
  zonecfg:zone:attr>set type=string
  zonecfg:zone:attr>set value="Sys Admin"
  zonecfg:zone:attr>end
  
• Save configuration
  zonecfg:zone>info
  zonecfg:zone>verify
  zonecfg:zone>commit
  zonecfg:zone>exit

After this check directory or filesystem with /zones/zone is available with 700 permissions and check and start installation
#zoneadm -z zone verify
#zoneadm -z zone install
#zoneadm -z zone boot
#zlogin -C zone --> For Console of Zone