Set ILOM SUN Server by serial CLI

Set ILOM SUN Server by serial CLI

To configure ILOM on a SUN server make a serial connection to the rear of the server.

Putty 9600 baud

Default credentials on a new system:
root
changeme or l3tm3in

First reset the ILOM to defaults:
Reset to defautl: set /SP reset_to_defaults=all

Now we can configure the network settings, these are for a static setup. Make sure to press enter between every line.

set /SP/network pendingipdiscovery=static
set /SP/network pendingipaddress=IPADRESS
set /SP/network pendingipgateway=IPADRESS
set /SP/network pendingipnetmask=IPADRESS
set /SP/network commitpending=true

Finaly check if the settings are applied with:

show sp/network

 

 

Use WDS to boot Linux live cd’s

In a previous article, Installing Linux via PXE using Windows Deployment Services (WDS), I talked about using PXELinux to enable deployment of WDS images, Linux distros and a multitude of tools. It got a bit heavy when trying to make this all work but the system is now up & running and we have already benefitted from it on many occasions. So here is my guide for Microsoft admins wanting to enhance their existing Windows Deployment Services server.
Read More

Calculate vCPU ESXi and Hyper-V

There is much confusion about creating number of Virtual machines on a single host, keeping memory and storage unlimited. In-order to clarify , I would say ,Yes we do have limitation in creating VMs and its explained below in brief.

If you are using Windows Hyper-V for virtualization, so the number of VMs is dependent on the number of Virtual CPU available. We can have a multiple formula based on the scenario however the best on the basis of my personal experience I can put it in as

  Total vCPU = (Number of processors) * (Number of cores) * (Number of threads per core) * 8

Number of Processor: In desktop environment we usually uses one physical processor however in server environment it exceeds the number.
Number of Cores: Usually now a days 2 core, 4 core, 6 core, 8 core and even 12 core and 16 core processors are also available.
Number of threads per core: if hyper-threading is supported and enabled through BIOS, you can consider the number as 2 else 1.

If you need to get into the details you can also follow the below mentioned link:-

http://blogs.technet.com/b/virtualization/archive/2011/04/25/hyper-v-vm-density-vp-lp-ratio-cores-and-threads.aspx

So if you have 1 dual core physical processor with hyper threading enable ,then based on the formula you have  Total vCPU = 1 (Physical processor) x 2 (dual core) x 2 (hyperthreading) x 8 = 32 vCPU

Since you have 32vCPU so you can create 32 VM with 1 vCPU each or 8VM with 4 vCPU each.

The maximum number of vCPU you can assign to a virtual machine in Hyper-V is 4.

Where as in VMWare , ESXi host enabled for hyper-threading should behave similarly to a host without hyper-threading . So the number of CPU allocation is not restricted or dependent on the basis of hyper-threading enabled or disabled.

In ESXi you doesn’t calculated based on vCPU available however it is totally dependent on the processing capacity. As you can restrict the usage of CPU capacity. So processing capacity can be calculated as :-

Total Processing Capacity = Processing capacity, rated on processor * No. Of Physical Socket * No. of Cores per socket

For example :-
If we have 2 physical processor of 2.599GHz capacity and each processor has 6 Cores then as per formula,Total Processing Capacity = [2.599 (Processor rating)*2(physical processor)*6(cores per processor) ]GHz = 31.188GHz

Since we are now aware of the total processing capacity, so we can create a virtual machine by allocating a CPU and restricting its maximum CPU utilization to be 512 MHz ,the number of VM so similar configuration can be upto 609 VMs or you can create 12 VMs of 2.599GHz capacity its scalable or you can calculate based on your requirement or even if you wish you can leave its default configuration and VMWare will smartly manage the CPU utilization of all the CPU processing capacity ,only condition is that all the VMs collectively must not exceed the processing capacity at the same time which may result in bottleneck causing latency or slowness.