How to Create a Link Aggregation – Bonding on Solaris

A link aggregation consists of several interfaces on a system that are configured together as a single, logical unit. Link aggregation, also referred to as trunking

Basically link aggregation is like a bonding on server. Its work in active and passive mode. At a time one network device are up and other bonding device are in stand by mode.

Requirements for Link Aggregations:-

1. Your link aggregation configuration is bound by the following requirements:

2. You must use the dladm command to configure aggregations.

3. An interface that has been plumbed cannot become a member of an aggregation.

4. Interfaces must be of the GLDv3 type: xge, e1000g, and bge.

5. All interfaces in the aggregation must run at the same speed and in full-duplex mode.

If the solaris box matches these above requirement, after that we can only able to create a link aggregation.

How to create a link aggregation in Solaris operating system:-

1. As a root user please login on the solaris operating system so you have full administrative role to perform the action.

2. Determine which interfaces are currently installed on your system.

For this work you need to run the below command.

#dladm show-link

The above command show you the which interfaces are currently installed on your system.

#dladm show-link

ce0             type: legacy    mtu: 1500       device: ce0

ce1             type: legacy    mtu: 1500       device: ce1

bge0            type: non-vlan  mtu: 1500       device: bge0

bge1            type: non-vlan  mtu: 1500       device: bge1

As per link aggregation requirement, we can only use the interface which start from bge etc. here the device bge0, bge1 are currently installed on the server.

3. In this step now determine which interfaces have been plumbed. for this please run the below command.

#ifconfig -a

lo0: flags=2001000322 <UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1

        inet 127.0.0.1 netmask ff000000 

ce0: flags=1000420 <UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2

        inet 192.168.1.43 netmask ffffff00 broadcast 192.168.1.255

        ether 0:4:7c:8:92:7d 

4. Now create a link aggregation using below command

#dladm create-aggr -d bge0 -d bge1 1

here 1 is the key number which identify the link aggreation , and it is lowest number.

5. now configure & plumb the newly created aggregation. Please use the below command to do this.

#ifconfig aggr1 plumb 192.168.1.56 up

6. To check the status of the aggregation you just created, please run the below command.

#dladm show-aggr

key: 1 (0x0001) policy: L4      address: 0:4:7c:8:92:7d (auto)

device   address           speed         duplex  link    state

bge0     0:4:7c:9:87:4e    1000  Mbps    full    up      attached

bge1     0:4:7c:9:32:9e    0     Mbps    unknown down    standby

7. For link aggregations with IPv4 addresses, create an /etc/hostname.aggrkey file.

#vi /etc/hostname.aggr1

92.168.1.56

8. perform a reboot.

#reboot -- -r

How to Create VNIC and Assign a IP Address on Solaris 11

For creating Virtual NIC and assigning fix static IP address in Solaris 11 we need to understand the basic difference between older version of Solaris and Solaris 11.

In Solaris 10, according to the NIC manufacturer,physical network interfaces are named as (Ex:bge,e1000g,nxge).But in Solaris 11 onwards,the names are hidden from the view and all the interfaces will be named as net0,net1…netx.

Before forward to main work we need to know that using which command we can check which interface has been mapped to physical interface. Using below command you can check all the network interface details.

sun01# dladm show-phys

LINK              MEDIA                STATE      SPEED  DUPLEX    DEVICE

net0              Ethernet             up         1000   full      bge001

net1              Ethernet             up         1000   full      bge002

If you want to show all dladm level devices,including VNIC’s & aggregation links,use the below command-

sun01# dladm show-link

LINK                CLASS     MTU    STATE    OVER

net0                phys      1500   up       --

net1                phys      1500   up       --

In Solaris 11,you can give a meaning full description(net1/oracle_VIP) to all the IP address on the system unlike Solaris 10. (e1000g1:2)

How to assigning new IP address to NIC:-

1. We can see how we assigned IP address to the physical interface, so in the first step we will check all the physical interface using below command.

sun01# dladm show-phys

LINK              MEDIA                STATE      SPEED  DUPLEX    DEVICE

net0              Ethernet             up         1000   full      bge001

net1              Ethernet             up         1000   full      bge002

2. In this step we would know how we will create a new interface so you can use the below command for this.

sun01# ipadm create-ip net1

3. This is the main step to assign the static ip address to the newly created interface net1. You can use below command or syntax to assign the fixed IP.

sun01#ipadm create-addr -T static -a local=10.135.0.2/24 net1

You can change your IP address accordingly. 

4. Now we will Verify whether IP address is configured or not which we assigned on above step.

sun01# ifconfig  net1

net1: flags=1000843 mtu 1500 index 7

        inet 10.135.0.2 netmask ffffff00 broadcast 10.135.0.255

        ether 0a:cB:12:8e:15:e2

If you see the above output, the new IP address is shown on net1 interface successfully. Using above all 4 steps you can know how we will assinged the static ip address to the sun solaris 11 operating system.

Now we will go for Virtual Network Interface creation steps.I can create N number of VNIC’s using single physical interface.These VNIC are treated as actual physical interface and possible to assign to local zones with  full access to it.

How to create a new VNIC using interface net2:-

1. In the initial steps I will run the same command to check out the all physical interface which is available on the Solaris 11 Server.

sun01# dladm show-phys

LINK              MEDIA                STATE      SPEED  DUPLEX    DEVICE

net0              Ethernet             up         1000   full      bge001

net1              Ethernet             up         1000   full      bge002

net2              Ethernet             unknown    1000   full      bge003

If you see the above command output network interface "net2" is in unknow state, So in the next step I will create a new VNIC using net2.

2. In this step we will create a new VNIC using net2. In my case i will suppose VNIC name “vnic01”, so for creating the new VNIC we will run the below command which describe below.

sun01#dladm create-vnic  -l net2 vnic01

3.Now, We will plumb the virtual interface and create a default IP to new VNIC.

sun01# dladm create-vnic  -l net2 vnic01

sun01# ipadm create-ip vnic01

sun01# ifconfig vnic01

vnic01: flags=1000842 mtu 1500 index 8

        inet 0.0.0.0 netmask 0

        ether 2c:18:10:Ce:1a:12

4. Now we will assign the new IP address to VNIC and verfif that new IP for vnic01 is configured or not.

sun01# ipadm create-addr -T static -a local=10.135.0.3/24 vnic01

sun01# ifconfig vnic01

vnic01: flags=1000843 mtu 1500 index 8

        inet 10.135.0.3 netmask ffffff00 broadcast 10.135.0.255

        ether 2c:18:10:Ce:1a:12

5. In the final step we will try to run snoop on VNIC which we have create now. You can verify through snooping is that VNIC01 is working fine or not.

sun01# snoop -d vnic01

The same way you can create a multiple VNIC and assign a new IP address to it. Hope you got some idea about Solaris 11 networking part after reading this post. Please let me know if you have any doubt regarding this post.

How to Set a Solaris 11 Machine's NIS Domain Name

If domainname command not showing any output then you need to check either defaultdomain or NIS domain is configured or not.

Steps which we need to follow to set the NIS domain name as describe below.

1. Please run  "cat /etc/defaultdomain" command and check the output that domain is set or not.

    #domainname

The above command shows you the current domain name on the server, if no output is display you need to set the domain by below command.

   #domainname UNIX

Run the below command to check the output again.

   #domainname

   UNIX

You can see the output. All the above command is used for checking and setting the domain name on the Solaris operating server.

2. If you want to set the domain using NIS, then first please check the NIS/DOMAIN service is running or not

root@sun:~# svcs status nis/domain

STATE          STIME    FMRI

disabled       18:20:04 svc:/network/nis/domain:default

In my Solaris machine, currently NIS domain services is stop, we will start it later once we will configured the NIS domain name on the Solaris server.

3. Now you need to configure the NIS domain for your machine ( in my case - UNIX is my domain name)

root@sun:~# svccfg -s svc:/network/nis/domain setprop config/domainname = UNIX

If you see the above command, "svccfg" command is used to configure the domain and set the proper properties.

4. In the final step once the configuration has been completed you need to refresh & enable the NIS domain service again.

root@sun:~# svccfg -s nis/domain:default refresh

root@sun:~# svcadm enable nis/domain

You can check the NIS domain service status as describe in step 2. 

5. Now you can verify the domain name using below command so we can ensure that the NIS domain name has been set successfully. 

root@sun:~# domainname

UNIX

The above command output show "UNIX" domain name which we set during the configuration, so using these above method we can easily set the NIS domain name on the Solaris operating server.

Solaris Server process Monitoring tool- prstat

We have different type of tools and command which are used in Solaris or other Unix system to monitor the system process. But if we are talking about only Sun Solaris server then we have very good process tool which is called "prstat".

In this post, we will find that how prstat is work on the Solaris platform.

⇾   !-[solaris]# prstat

When you run the above command on the command line you will get the below output on the CLI screen which are refreshing in every few seconds and sorting all the information regarding the system resource.

  PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP

 21322 root      11M 3236K cpu0    59    0   0:00:00 0.0% prstat/1

 21323 root      18M 4788K sleep   59    0   0:00:00 0.0% sshd/1

 22345 root      10M 2188K sleep   59    0   0:00:00 0.0% bash/1

   584 root       13M 3832K sleep   59    0   0:01:59 0.0% nscd/51

   154 root       13M 2068K sleep   59    0   0:00:00 0.0% syseventd/18

   183 root     1772K  776K sleep   59    0   0:00:13 0.0% utmpd/1

   538 root       11M 2572K sleep   59    0   0:00:00 0.0% picld/4

Total: 12 processes, 31 lwps, load averages: 0.00, 0.00, 0.00


This is a quick view of the prstat command but if you wanted to get a different view of the same info, like a summary of what users own these CPU consuming processes

⇾   !-[solaris]# prstat -a

If you run prstat with the -a option (prstat -a) you will get an output similar to the default one, but the last few lines of it will be used for providing a really useful report of the users consuming top system resources.


  PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP

 21322 root      11M 3236K cpu0    59    0   0:00:00 0.0% prstat/1

 21323 root      18M 4788K sleep   59    0   0:00:00 0.0% sshd/1

 22345 root      10M 2188K sleep   59    0   0:00:00 0.0% bash/1


 NPROC USERNAME  SWAP   RSS MEMORY      TIME  CPU

     5 root      52M   13M   1.3%   0:00:00 0.0%

    50 root      841M  571M    56%   0:22:22 0.0%

     2 daemon     17M 4520K   0.4%   0:00:04 0.0%

Total: 12 processes, 31 lwps, load averages: 0.00, 0.00, 0.00

We have different type of syntax which we can used to monitor the Solaris server process which are listed below.

!-[solaris]# prstat -L  -> This shows thread per line instead of one process per line
!-[solaris]# prstat -s -> prstat output can be sorted using set of sub-options .sub options are                  cpu,pri,rss,size,time 

!-[solaris]# prstat -t  -> It provides complete users resource utilization.
!-[solaris]# prstat -Z  -> It provides summary per local zone.

Solaris zone: error: net0: failed to create VNIC: operation not supported

In this post, I will discuss with you one of the most interesting error which I am facing when I boot the local zone on Solaris 11.3. The description of this interesting issue as describe below.

Description of error:

!-[solaris]# zoneadm -z sun01 boot

zone 'sun01': error: net0: failed to create VNIC: operation not supported

zoneadm: zone sun01: call to zoneadmd(1M) failed: zoneadmd(1M

I have try to create and configure the  VNIC on Solaris 11.3 operating server but it get failed with the same error.

!-[solaris]#dladm create-vnic -l net0 vnic01

dladm: vnic creation failed: operation not supported

If you are also facing a such error while booting the local zone on solaris 11 server, then please use the below solution to resolve such issues.

Solution of error:

➥ This error "failed to create VNIC: operation not supported" would normally come when there are not enough mac addresses to assign to the zone. So now we need to add alternate mac addresses to the network interface.So before adding the new mac address we will stop LDM.

!-[solaris]#ldm list-domain

NAME            STATE     FLAGS  CONS   VCPU MEMORY  UTIL NORM UPTIME

primary         active    -n-cv- UART   8    8G      2.0% 2.0% 41d 20h 14m

0004fb0000060000ff1d3d8336112f6f active    -n---- 5001   50   64G     0.1% 0.1% 18h 23m

➥ Now log in to the Solaris global zone and check if net0 have additional MAC addresses or not. Please use the below command to check the status.

!-[solaris]# dladm show-phys -m

LINK               SLOT    ADDRESS           INUSE CLIENT

net0               primary 0:21:f6:d6:d3:e5  yes  net0

                   1       0:14:4f:f9:6d:8d  no   --

                   2       0:14:4f:fb:10:2b  no   --

                   3       0:14:4f:f9:41:d6  no   --

                   4       0:14:4f:f8:dd:c8  no   --

net1               primary 0:21:f6:51:be:4d  yes  net1

➥ Now zone will start without any issue, as we have assigned the new mac address to this zone.

!-[solaris]# zoneadm -z sun01 start

Please let me know if you are facing any issue regarding this error.

How to configure NTP server and client on Solaris 10 and Solaris 11

As you know NTP ( Network time Protocol) is one of the oldest internet protocol still in use and it allows the synchronization of computer clocks distributing UTC (Coordinated Universal Time) over the network.

NTP Service on Solaris 10 and Solaris 11:

Solaris 10 used the SMF utility (Service Management Facility) and the NTP service is now managed by SMF (Service Management Facility). NTP daemon configured using Service management facility (svc:/network/ntp:default) and a bunch of sample ntp.conf files to quickly configure a machine as a client or as a server. On Solaris 11 only ships with NTP v. 4, the NTP v. 4 service is identified by the name ntp4. You can check the ntp status using below command.

!-[solaris]# svcs status ntp

STATE  STIME   FMRI

online 10:14:23 svc:/network/ntp:default

If you see the above command output it is shown that network time protocol services is enable and online on the server.

Steps for Configuring a NTP client:

Suppose your machine is just a client machine, then you can just take the /etc/inet/ntp.client file and copy it to /etc/inet/ntp.conf.

multicastclient 127.0.0.1

If you see the configuration it's a passive configuration for a Server host which listens for NTP server putting packets on the NTP multicast network, 127.0.0.1. If your machine is on LAN without NTP server then in that case we are not recveied any packet and for this we need to use Public NTP server for host.

In my case, I'm using the Indian pool in.pool.ntp.org and my configuration file contains:

server 2.in.pool.ntp.org

server 1.asia.pool.ntp.org

server 3.asia.pool.ntp.org

Normally NTP requires a poll period to elapse before starting synchronizing your clock. If you want NTP to start immediately, which you most probably will if you're configuring a desktop environment, you can take advantage of iburst keyword, introduced in NTP v. 4: it instructs NTP to start the synchronization almost right away.

server 2.in.pool.ntp.org ibrust

server 1.asia.pool.ntp.org ibrust

server 3.asia.pool.ntp.org ibrust

You must make sure you're configuring NTP implementation corresponding to the syntax you're using.

Setting up the drift file:

The last thing which is remaining for NTP server setup in the client machine is to set up  drift file location. On my machine it is 

driftfile /var/ntp/ntp.drift

After setup the drift file configuration we will start the NTP servivce again 

!-[solaris]# svcadm restart svc:/network/ntp:default

!-[solaris]# svcs status svc:/network/ntp:default

STATE  STIME   FMRI

online 12:20:12 svc:/network/ntp:default 

Once the service is running, you can check which server you're using with ntpq, Please run the below command to check the ntpq.

!-[solaris]# ntpq -p

Setting up an NTP server:

Now in above step you see the NTP service has been started successfully, so now, you'll probably want to setup all of your machines.

If you're in a LAN, you can setup an internal NTP server which will provide data to other clients on your LAN. As before, you can take inspiration from the server configuration file shipped with Solaris 10 or Solaris 11, /etc/inet/ntp.server.

After setting up the drift file and the clients you're going to use, you can examine the other options and fine-tune them at your taste. Let's give a quick look at it.

server 127.127.XType.0

Now you have configured the NTP server properly. Please comment on the post if you have any suggestion.

NFS mount on Solaris 11 Non-Global zones systems

In this article, we would learn how we mount folder from one Non-Global zone to another zone on Solaris 11 operating system using network file sharing.

For an example, we can use take a two local zone "sunz01" & "sunz02". We will mount one folder named "/export/backup" from "sunz01" local zone to another zone "sunz02" on "/project/export/data" location.

Step by Step method of NFS mount on Solaris 11:
 

➦ In the first step we will create the directory on "sunz02" zone where we want to mount the folder.

sunz02#mkdir /project/export/data
 

➦ In second step, we will make a configuration for this process. So for this work you need to login on global zone with root access and make an entry on the dfstb configuration file.

sun#vi /etc/dfs/dfstab

share -F nfs -o rw=sunz02 /zones/sunz01/root/export

If you see the above entry, we have provided the read/write access to directory on sunz02 server where we mount the folder from sunz01 local zone.
 

➦ In next step you need to login on sunz02 server and mount the shared folder using below command.

sunz02#mount sun:/zones/sunz01/root/export/backup /project/export/data
 

➦ Once you run the above command the folder is mount from one local zone to another zone temporarily. You can go to the directory and verify that the data which is listed on /export/backup folder is show on sunz02 directory.
 

➦ In the last step you need to restart the NFS service on the global zone so the configuration files and other changes makes affect. But these configuration are available until we are not taking reboot of the zone.

If you have any query regarding this topic, please post your comment here, we will get back to you shortly.

How to create whole root zone on Sun Solaris 11 Operating System

In Solaris 10, we can’t install it without configuring repo on Solaris while in Solaris 11 ,first we need to create a  Solaris 11 repo , then we can install Solaris local zone. In Solaris 11, all the local zones by default uses exclusive ip address. You can’t set the IP address while configuring the zone. After the installation of zone, you can able to configure the IP from the local zone itself.

On my Solaris 11 machine I have already installed one local zone which is used for my R&D work so for this post work I need to install second local zone. So I will clone it from the first local zone.

Step by step method to create a zone on Solaris 11:-

➤ In the first step we will create a new local zone. For the zone creation we will used "zonecfg" command and configure the zones as whole root zone without any options supplied.

!-[solaris]#zonecfg -z sun02

sun02: No such zone configured

Use 'create' to begin configuring a new zone.

zonecfg:sun> create

zonecfg:sun> info

zonename: sun02

zonepath:

brand: native

autoboot: false

bootargs:

pool:

limitpriv:

scheduling-class:

ip-type: shared

hostid:

zonecfg:sun> set zonepath=/zones/sun02

zonecfg:sun > add anet

zonecfg:sun:anet> set linkname=net0

zonecfg:sun:anet> end

zonecfg:sun > verfiy

zonecfg:sun > commit

zonecfg:sun > exit

In the above step, we will create a zne and set the zonepath. In my case my new solaris zone path is "/zones/sun02" in your case you can change your installation zone path. Here, I assign the network interface "net0" for this newly zone.

➤ After successfully zone creation in above step 1, we will go for start the Solaris zone installation in this step. As I explained you for solaris 11 local zone installation we required solaris 11 repo which is used for installation part.

!-[solaris]#zoneadm -z sun02 install

The following ZFS file system(s) have been created:

    rpool/zones/sun02

Progress being logged to /var/log/zones/zoneadm.30220110Z233232Z.sun02.install

       Image: Preparing at /zones/sun02/root.

 AI Manifest: /tmp/manifest.xml.F_ayqq

  SC Profile: /usr/share/auto_install/sc_profiles/enable_sci.xml

    Zonename: sun02

Installation: Starting ...

              Creating IPS image

Startup linked: 1/1 done

        Installing packages from:

solaris  origin:  http://localhost:1008/solaris/ce43f14c4791b5320596e2023cde1ec08709a3af/

DOWNLOAD                                PKGS         FILES    XFER (MB)   SPEED

Completed                            183/183   33556/33556  222.2/222.2  139k/s

PHASE                                          ITEMS

Installing new actions                   46825/46825

Updating package state database               Done

Updating image state                          Done

Creating fast lookup database                 Done

Installation: Succeeded

Note:Man pages can be obtained by installing pkg:/system/manual done. Done: Installation completed in 2392.837 seconds.

Now our newly zone "sun02" has been installed successfully. All the configuration files are keep in /zones/sun02/root directory.

➤ In this step we will boot the newly installed zone "sun02".

!-[solaris]#zoneadm -z sun02 boot

You can check the status of zone that it is running or not after boot.

!-[solaris]#zoneadm list -icv

ID NAME      STATUS     PATH          BRAND    IP

   0 global   running    /            solaris  shared

   1 sun01    running    /zones/sun01 solaris  excl

   3 sun02    running    /zones/sun02 solaris  excl

if you see the above output newly zone "sun02" is running working fine on the solaris 11 operating system.

➤ Now in the next step, we will login on the local zone console to complete the configuration process.

!-[solaris]# zlogin -C sun02

[Connected to zone 'sun02' console]

You can press enter when you will get this message. Now it is asking for some configuration step we need to give a details one by one.

Time Zone: Regions

select the region that contains your time zone.

Regions

UTC/GMT

Africa

Americas

Antarctica

Arctic Ocean

Asia

Atlantic Ocean

Australia

Europe

Indian Ocean

Pacific Ocean

F2_Continue  F3_Back  F6_Help  F9_Quit

Time Zone: Locations

Select the location that contains your time zone.

Locations

x Afghanistan

x Armenia

x Azerbaijan

x Bahrain

x Bangladesh

x Bhutan

x Brunei

x Cambodia

x China

x Cyprus

x East Timor

x Georgia

x Hong Kong

v India

F2_Continue  F3_Back  F6_Help  F9_Quit

Time Zone

Select your time zone.

Time Zones

Asia/Kolkata

F2_Continue  F3_Back  F6_Help  F9_Quit

System Configuration Summary

Review the settings below before continuing. Go back (F3) to make changes.

Time Zone: Asia/Kolkata

Language: *The following can be changed when logging in.

Default language: C/POSIX

Terminal type: vt100

Users:

No user account

Network:

Computer name: sun02

Network Configuration: Automatic

Support configuration:

Not generating a Support profile as OCM and ASR services are not installed.

Hostname: sun02

So now your zone is fully configured ans installed successfully. You can login in zone very easily. So in next step we will see the post configuration settings which is required on local solaris zone.

➤ In the final step you need login to local zone sun02 and configured the ip address on the zone sun02.

!-[solaris]# zlogin -z sun02

You can successfully login in your newly created solaris 11 zone. Please leave a comment if you have any doubt , I will get back to you as soon as possible.