Each version of vSphere has some improvements and one of important improvements are configuration maximums that allows administrators to have much bigger virtual machines, hosting more virtual machines, use faster network and storage connections. This is very important that you should aware about your current configuration maximums because you can prepare your forecast plans for increasing virtual machine or ESXi host resources or even changes on network or SAN environments based on these configuration maximums. Lets review latest vSphere configuration maximums and compare them with earlier versions. Virtual Machine Item vSphere 5.5 vSphere 6.0 vSphere 6.5 vCPU 64 128 128 Memory 1TB 4TB 6128GB Swap File 1TB 4TB 6128GB Virtual SCSI adapters per virtual machine 4 4 4 Virtual SCSI targets per virtual SCSI adapter 15 15 15 Virtual SCSI targets per virtual machine 60 60 60 Virtual disk size 62TB 62TB 62TB IDE controllers per virtual machine 1 1 1 IDE devices per virtual machine 4 4 4 Floppy controllers per virtual machine 1 1 1 Floppy devices per virtual machine 2 2 2 Virtual SATA adapters per virtual machine 4 4 4 Virtual SATA devices per virtual SATA adapter 30 30 30 Virtual NICs per virtual machine 10...
I have read a KB on VMware Knowledge Base and it says that an Intel Virtualization Technology can be cause of PSOD. This is little funny because ESXi will be affected by wide range of Intel Xeon processor family: Intel® Xeon® Processor 55xx Series Intel® Xeon® Processor 56xx Series Intel® Xeon® Processor 65xx Series Intel® Xeon® Processor 75xx Series Intel® Xeon® Processor E5-1400 v2 Product Family Intel® Xeon® Processor E5-1600 v2 Product Family Intel® Xeon® Processor E5-1600 v3 Product Family Intel® Xeon® Processor E5-2400 Product Family Intel® Xeon® Processor E5-2400 v2 Product Family Intel® Xeon® Processor E5-2600 Product Family Intel® Xeon® Processor E5-2600 v2 Product Family Intel® Xeon® Processor E5-2600 v3 Product Family Intel® Xeon® Processor E5-2600 v4 Product Family Intel® Xeon® Processor E5-4600 Product Family Intel® Xeon® Processor E5-4600 v2 Product Family Intel® Xeon® Processor E5-4600 v3 Product Family Intel® Xeon® Processor E5-4600 v4 Product Family Intel® Xeon® Processor E7-2800 Product Family Intel® Xeon® Processor E7-4800 Product Family Intel® Xeon® Processor E7-8800 Product Family Intel® Xeon® Processor E7-8800/4800/2800 v2 Product Families Intel® Xeon® Processor E7-8800/4800 v3 Product Families Intel® Xeon® Processor E7-8800/4800 v4 Product Families There is a workaround for preventing the problem and PSOD on server but VMware...
Three flings are updated by VMware engineers and published during this week and past week. HCIBench This fling is a “Hyper-converged Infrastructure Benchmark” tools and if you have vSAN in your environment, this fling is very useful for you. Changelog Increased Timeout value of client VM disk from 30 seconds to 180 seconds. Disabled client VM password expiration. Disabled client VM OS disk fsck. Set Observer interval to 60 seconds to shrink the size of observer data. Fixed PCPU calculation. Created link directory of /opt/automation/logs, user will be able to review the testing logs in http://HCIBENCH/hcibench_logs/ Increased the RAM of HCIBench from 4GB to 8GB to avoid running out-of-resource issue.
Last week, we went limit some our machines IOPS but we saw the limitation is not working on our machines, after searching the issue on VMware KB, we found an KB (2059192) that explain an known issue on ESXi 5.5 and ESXi 6. It seems, disk IO scheduling model has been changed on the platforms and it’s cause of the issue. But the solution is so simple, just you need to change an ESXi’s parameter: Revert the disk I/O scheduler to an earlier version by using the vSphere Web Client In the vSphere Web Client, edit the Disk.SchedulerWithReservation parameter in the Advanced System Settings list for the host. Navigate to the host. On the Manage tab, click Settings and click Advanced System Settings. Locate the Disk.SchedulerWithReservation parameter.Note: You can use the Filter or Find text boxes to find the parameter easily. Click Edit and set the parameter to 0. Click OK. Revert the disk I/O scheduler to an earlier version by using an ESXCLI command In the ESXi Shell to the host, run this console command: esxcli system settings advanced set -o /Disk/SchedulerWithReservation -i=0 There is no need to reboot or anything else. The configuration will be applied immediately.
Introduction Oracle VM introduced two main modes or domain types: Paravirtualized (PVM):A virtual machine with a kernel that is recompiled to be made aware of the virtual environment. Runs at near native speed, with memory, disk and network access optimized for maximum performance.Paravirtualized guests use generic, idealized device drivers, which are part of the guest’s OS. The I/O operations using these generic device drivers are mapped to the real device drivers in dom0. The generic, abstracted drivers in the guest seldom change and provide excellent guest stability. The dom0 domain, alternatively, can use the native hardware vendor drivers, and the guests can safely migrate to another dom0 with slightly different drivers. For other resources such as CPU and memory, paravirtualized kernels make special “hypercalls” to the Xen hypervisor. These hypercalls provide better performance by reducing the number of instructions and context switches required to handle an incoming request. By contrast, on an emulated (hardware virtualized) guest, driver requests engage the guest’s interrupt handler, increasing the I/O operation overhead. Hardware Virtualized Machine (HVM):A hardware virtualized guest runs on the virtualization platform as it would on a physical host. Because the device drivers of the hardware virtualized guest are emulated, dom0 must...
Today, virtualization is under development by many companies and big companies trying to have their virtualization solution. Oracle VM is Oracle virtualization platform that it’s based on Xen hypervisor. You may know that any hypervisor has some limitations and configuration maximums and you should consider the limitation when you are deploying infrastructure or creating virtual machines. Oracle VM Manager Maximums em x86 Maximum SPARC Maximum Oracle VM Servers in a server pool (unclustered) 64 64 Oracle VM Servers in a server pool (clustered) 32 32 Number of servers 256 (16 servers * 16 server pools) 256 (16 servers * 16 server pools) Number of server pools 16 16 Number of configured virtual machines 5120 (20 virtual machines * 256 servers) 5120 (20 virtual machines * 256 servers) Number of running virtual machines 2,560 (10 virtual machines per server * 256 servers) 32768 (128 virtual machines per server * 256 servers) Fujitsu M10-4S, this limitation is 65536 (256 virtual machines per server * 256 servers) Oracle SPARC M-series servers, this limitation is 128 virtual machines per server * the number of physical domains (PDoms). Each physical domain acts as its own server with its own set of logical domains.
What’s OverCommitment? OverCommitment means virtual machines can use more resources than physical resources. CPU OverCommitment means you can create virtual machines with vCPU more than your server physical CPU, for example: Your server has two socket and each socket has 12 cores and also hyper-threading is enable, so you have 24 physical cores and 48 logical cores totally. But you can create and power on some virtual machines that those virtual machines have more than 48 cores totally. Memory OverCommitment means your virtual machines can use more memory than the physical machine (the host) has available. For example, you can have a host with 2GB memory and run four virtual machines with 1GB memory each. In that case, the memory is overcommitted. Advantages vs Disadvantages With OverCommitment, you can run more virtual machines but if you don’t have sensitive machine. If you have critical services on virtual machines, you need take care about OverCommitment because it can decrease your machines performance. How Take Care? You know, there is many native and third-party applications for monitoring vSphere environment but if you don’t have budget for buy monitoring software, you can use PowerCLI! There is a free PowerCLI module that you can...
Today, many companies have virtualized farms for their server infrastructure or desktop infrastructure and cloud services. The companies have critical information on their virtualized farms and keeping safe them is one of big concerns. Big companies or even small companies have security teams and the teams tries to keeping secure the environments in different layers. Most of the security products are working on physical layer or network and application layer but what about Hypervisor layer? vSphere Hardening VMware publishing a hardening guide for each vSphere version to help administrator to keep their environments more secure. vSphere hardening guides are available in the below link as Excel files: Download – Hardening Guides Previously, VMware had published an application to analyzing your vSphere environment and report you any security issue according to hardening guides. VMware Sphere Compliance Checker was available up to vSphere 5.5 and that’s not available for vSphere 6.x but you can use “VMware vRealize Configuration Manager” on this regard. Anyway, you can check and change security configurations accordion to hardening guides on your servers manually.
VMware has introduced new feature for kernel protection against memory error in ESXi. VMware called the new feature: Reliable Memory Technology or RTM. The feature is one of new features in ESXi 5.5! ESXi use a zone of memory that it’s more reliable than other offsets of memory, so risk of PSOD will be reduced. Also when part of memory has error, ESXi will stop to using the part of memory. There is some other technique against memory corruption or memory health error such as memory mirroring but Reliable Memory Technology can help you on this regard without loosing half of your memory capacity. Because memory mirroring is just like to RAID 1 on hard disks. Dell has introduced another feature on its server by using Reliable Memory Technology and called the new feature: Fault Resilient Memory or FRM. Fault Resilient Memory will provide “Fault Resilient Zone” and ESXi will put its kernel to the zone. The features can protect ESXi kernel and VMs as well. So if you have critical service on a VM, you can force ESXi to keep its memory on RTM or FRM zone to avoid memory error and down time for the machine. You can configure...
As you may know, you can install hypervisors on virtual machines for testing purpose. When you had installed ESXi on virtual machine, you did nesting virtualization. But think about installation process, any installation needs around 20 minutes and if you cloned a ESXi machine, it will be not worked properly on other virtual machine and you have to change your configuration manually. Now there is good solution and it’s ESXi virtual appliance. You can download ESXi OVA file and import it to your host or vCenter and also you can configure it during import process. ESXi virtual appliance is available at the below links: ESXi 6.0 Virtual Appliance download link ESXi 5.5 Virtual Appliance download link ESXi virtual appliance maybe updated after releasing new version. The last versions are: ESXi 6.0 U2 ESXi 5.5 U3 The ESXi virtual appliance including the below configurations: ESXi 6.0 U2: ESXi 6.0 Update 2 GuestType: ESXi 5.x (backwards compat) vHW 10 2 vCPU 6GB vMEM 2 x vmxnet3 vNIC 1 x 2GB HDD (ESXi Installation) 1 x 4GB SSD (for use w/VSAN, empty by default) 1 x 8GB SSD (for use w/VSAN, empty by default) VHV added dvFilter Mac Learn VMX params added disk.enableUUID VMX param added...
You may know, we have two types of VM, Full-Clone and Linked-Clone. Now, you have two questions, what are Full-Clone and Linked-Clone? Full-Clone: The Full Clone — A full clone is an independent copy of a virtual machine that shares nothing with the parent virtual machine after the cloning operation. Ongoing operation of a full clone is entirely separate from the parent virtual machine. Linked-Clone: The Linked Clone — A linked clone is a copy of a virtual machine that shares virtual disks with the parent virtual machine in an ongoing manner. This conserves disk space, and allows multiple virtual machines to use the same software installation. You can find more information on “VMware Documentation Center”. VMware offers VMware Horizon View Composer for creating and delivering linked-clone machines to your user but is it possible without View Composer? Answer is yes. You can create linked-clone VM without using VMware Horizon View on VMware ESXi or VMware Workstation. You can create a linked-clone VM on VMware Workstation easily by its GUI, you should follow the below steps: Step 1: Create new virtual machine, install OS, install VMware Tools and other applications. Step 2: Power off VM and create a snapshot from...
When you enabled HA on a clster, some definition alarm will be activated. “vSphere HA virtual machine monitoring action” is one of them, this alarm has two triggers: vSphere HA enabled VM reset with screenshot. vSphere HA is resetting VM. These are useful alarms for troubleshooting and know, what was happened for a virtual machines during fail-over. Also you can find the screenshot on the VM’s folder as a PNG file. [quotes_and_tips]
As I know and I have read it in many blogs that DRS is not sensitive to CPU ready time and when your cluster has enough CPU frequency and memory, cluster is balanced. There is a simple solution to resolve the issue or reduce the issue in your environment. Of course, this is not true solution and you need to analyze your platform, calculate your requirements especially CPU cores for your VMs and add more physical cores to by adding more physical servers. Create Alarm At first step, you need to define an alarm in your vCenter for virtual machine CPU ready: Create a user-defined alarm and enter a name as you wish. In Triggers tab, select “VM CPU Ready Time (ms) and define waning and error values and condition length. Default values are fine for most platforms. Just that, click on OK and our job is done. If you have virtual machines with high CPU ready, it will be detected by vCenter and alarm will be generated. Note: Please copy your alarm name and paste it in the below script. Also you can change exception time, if you need to run it after working hours. The Script Further Reading...
Achieving best performance on ESXi by HP ProLiant servers needs to change some default configuration on HP RBSU (ROM-Based Setup Utility) or UEFI (Unified Extensible Firmware Interface). Some of the configurations have been mentioned on “Performance Best Practices for. VMware vSphere” but some them not mentioned. You can change the below configuration to achieve best performance on ESXi: Setting Default Recommended Reason No-Execute Page Protection (AMD)System Options -> Processor Options -> No-Execute Page Protection No-Execute Memory Protection (Intel) System Options -> Processor Options -> No-Execute Memory Protection Enabled Enabled It’s recommended by HP and the features protects systems against malicious code and viruses. Intel Virtualization TechnologySystem Options -> Processor Options -> Intel Virtualization Technology AMD V (AMD Virtualization) System Options -> Processor Options -> AMD V (AMD Virtualization) Enabled Enabled When enabled, a hypervisor supporting this feature can use extra hardware capabilities provided by AMD/Intel. Intel Hyperthreading OptionsSystem Options -> Processor Options -> Intel Hyperthreading Options Enabled Enabled Intel Hyperthreading Options is a toggle setting that allows Intel Hyperthreading Technology to be enabled or disabled. Intel Hyperthreading delivers two logical processors that can execute multiple tasks simultaneously using the shared hardware resources of a single processor core. Enhanced...
If you have read HP or VMware documents about power management on ProLiant servers, both have suggested that configure power management to Static High Performance. I did it on all my ESXi servers and everything was good. But I have performance problem on some of G7 servers, I have checked all factors but I couldn’t find any reason about it. Then I found an article from HP about power management in ESXi and iLO3 (G7 servers). Seems, Static High Performance is root cause of performance issue on the servers with iLO3. This is related to power capping problem on iLO3. Based on the article suggestion, you should configure power management to OS control on the servers to preventing high CPU ready on virtual machines. You can see the instruction on the below link: https://h20566.www2.hp.com/hpsc/doc/public/display?sp4ts.oid=5178763&docId=mmr_kc-0105395&docLocale=en_US
