Vmware vsphere FAQs and Solutions

1. The virtual machine file is locked and cannot be properly power on

Fault Status:

Starting a virtual machine 95%, pausing and process interrupted, tip: ubable to access files since it is locked.

Bane: HA

Workaround:

(1) First, the HA function in cluster is closed. If the feature is not turned off, it is easy to cause a deadlock, and the VM is constantly beating, constantly again different in the esx inner loop being locked, acts vain.

(2) The disk file is locked, to solve, you must know exactly which ESX has locked him, this is the key.

Method: See/var/log/vmkernel However, before doing this, prepare some other work.

(3) In the VC, remove the locked VM from the inventory. The reason is very simple, this is a unregister process.

(4) According to/var/log/vmkernel, search for owner, you can find a statement similar to the following:

OCT 04:23:33 esx-hostname vmkernel:3:06:29:47.992 cpu6:1656) fs3:1975:checking If lock holders is live for lock [Typ E 10c00001 offset 52008960 v 380, HB offset 3554304 Oct, 04:23:33 esx-hostname vmkernel:gen, Mode 1, owner 48f5f637- 462688bc-fd28-0e1a6434b6f8 Mtime 38112]

The 48f5f637-462688bc-fd28-0e1a6434b6f8 behind Ok,owner is your target. Because he is the host that locks the VM.

(5) According to the following command, find out which esx uuid is 48f5f637-462688bc-fd28-0e1a6434b6f8

[[email protected] root]# esxcfg-info |grep-i ' System uuid '

(6) After locating the target host, it is of course the process of killing him to lock the VM. The reason for this is that HA has migrated VMs from other hosts, but there is no unregister and register process, so at the 3rd step, when you look at the VM's summary, the host IP still belongs to the problematic host. But the VM is also the new host Overlord hard power on, registration is not registered, and how to start it. Find the PID with the following command:

Ps-efwww|grep VIRTUALMACHINE.VMX

After the PID is found, the kill-9 PID

(7) At this time, also to determine one thing,. vswp file things. This is the experience of dealing with problems for the Taiwan guests. Because of ignoring this, so after killing import, re-register the VM, also said no swap files, start or failure.

When the VM is started, swap is generated automatically, and there is no swap file, because the swap exists, because the duplicate name causes the non-normal generation.

Enter into the/vmfs/volumes/lunid/vm_path/, vmkfs-d virtual_machine.vswp or into the datastore Browser, in the inside to remove the swap file can also.

(8) Complete the strategy, you can also go to the VM's settings--options--swapfile location, set the position of the save.

(9) Re-register the VM. Go to datastore Browser and find Vm.vmx,add to inventory.

(10) Start the VM. Good Luck.

　　2. Ignore esxi/vcenter server tips for SSH events

(1) VSphere client connects to VC or ESXi server;

(2) at home-> Inventory-> Hosts and clusters select your ESX server;

(3) Select the configuration on the right, then click on the Advanced Settings in the software column;

(4) In the Advanced settings, select the Uservars in the list on the left;

(5) After selecting Uservars in the list on the left, drag to the bottom on the right, change the uservars.suppressshellwarning value to 1, and do not need to restart.

　3. Attempt to migrate a VM with a USB device failed

Fault Status:

When you perform the Virtual Machine Migration Wizard, if the system detects that an incompatible USB device exists, the system prompts the following error message:

Currently Connecteddevice ' USB 1 ' uses backing ' PATH:1/7/1 ', which is not accessible.

Fault Analysis:

This problem typically occurs when a USB Passthrough devices function is turned on for the host with Vmdirectpath I/O support, and then a USB device is assigned to the specific VMS, such as a dongle;

Solution:

(1) Confirm that the USB device can be recognized and supported by the virtual machine and ensure that the WITH VMOTION option is checked when adding the USB device to the VMS;

(2) Re-attempt to add the USB device to the VMS before performing the vmotion action;

(3) Verify that the ESXi host has not been rebooted because the original supported vmotion Withvmotion function will be invalidated after the ESXi host restarts.

4. Troublshooting process of Convert Linux system

(1) Confirm that the source conversion Linux machine OS is in the official support list;

(2) have root authority;

(3) Confirm the DNS settings have no problem, note: should be both in Linux and Windows are added;

(4) Confirm that the source Linux can ping the IP with ESX or vcenter. If you fail at 2%, the most likely problem is permissions or firewall blocking;

(5) Confirm that Linux allows SSH to log in. This can help us to log on to the Linux system when converting.

(6) Confirm that the static IP is set for helper virtual machine, not DHCP (if there is no DHCP server in the network);

(7) Confirm that the source and destination are on the same subnet. If the different subnets through the routing link may be faulted;

(8) Note that converter does not support a Linux system that does a soft array. It can be done with a cold clone disc, which sets the soft array to/dev/md0.

(9) Log directory for VMware Converter standalone: C:\Documents and Settings\All Users\Application Data\vmware\vmware VCenter Converter Standalone, used for scheduling.

　5. VCenter Service Status page failure: Unable to retrieve health status

Fault Status:

The VCenter Server Status page prompts you for the following error message:

Unable to retrieve health status for VCenter inventory service

Unable to retrieve health status for VMware vSphere profile-driven Storage Service

When you perform a search action for vcenter server, you are prompted with the following error message:

Unable to connect to WebServices to perform query.

Verify that the "VMware virtualcenter Management webservices" service running onhttps://<vcenter-host-name>:10443

Fault Analysis:

This problem generally occurs because the Vcenter Server service has changed or a new vcenter server has been installed, but the database is still the original database result;

Solution:

Replace the Vws.jar, Jointool.jar, and Ds.jar files with the following steps:

Download the Vws.zip file in the attachment of this article and unzip the Vws.jar, Jointool.jar and Ds.jar files;

Stop VirtualCenter Server service and VirtualCenter Management webservices service; copy Vws.jar and Jointool.jar to C:\ProgramFiles\VMware\ Infrastructure\tomcat\webapps\web-inf\lib overwrite the original file; Copy the Ds.jar file to C:\Program files\vmware\infrastructure\inventory Service\lib overwrite the original file, restart the related service or the Vcenter server server.

6. The connection to the VMRC console has been disconnected ... Trying to reconnect

Fault Status:

When a vsphere client connects to an ESXi 5.0 host and the virtual machine is started, the console cannot be connected, and after the console is opened, a line "VMRC console connection is disconnected ... Trying to reconnect. ”

Fault Analysis:

As a matter of case, a DEP policy similar to a Windows system is a problem that is caused by an open state. But this is the case that all virtual machines are prompting for this error, troubleshooting DEP, and logging on to a VC platform with the local vsphere client, the problem remains the same. In order to eliminate the problem, changed a notebook login VC, suddenly found that the problem is missing. The original is the local vsphere client out of the question, and repeatedly thinking, I found that the problem before the local WIN7 with 360 security guard upgrade patch, whether it is a patch destroyed vsphere client a file it

Solution:

Reproduce the installation of vsphere Client, problem solving.

7. Conflict between VCenter Server and IIS on port 80

Fault Status:

Both VCenter Server and Microsoft Internet information Service (IIS) use port 80 as the default port for direct HTTP connections. This conflict causes VCenter Server to fail to restart after installing VSphere authentication Proxy. After the VSphere authentication Proxy installation is complete, VCenter Server cannot be restarted.

Fault Analysis:

If IIS is not installed when you install VSphere authentication Proxy, the Setup program prompts you to install IIS. Because IIS uses port 80, which is the default port for the vcenter Server direct HTTP connection, vcenter Server is in VSphere authentication Proxy. cannot be restarted after installation is complete. See page 32nd, "Ports required for VCenter Server".

Solution:

To resolve conflicts between IIS and VCenter Server for Port 80, do one of the following.

If IIS was installed before installing Vcenter Server, change the port of the Vcenter Server direct HTTP connection from 80 to a different value. If you have installed Vcenter server before installing IIS before restarting Vcenter Server, change the binding port of the IIS Default Web site from 80 to another.

8. The host fails to boot when ESXi is installed in UEFI mode

Fault Status:

In UEFI mode, reboot may fail when ESXi is installed on the host and rebooted. When this problem occurs, an error message that resembles the following is also displayed: An unexpected network error occurred. No boot device available (unexpected network error.) No boot device available).

Fault Analysis:

The host system does not recognize the disk on which ESXi is installed as a boot disk.

Solution:

(1) When an error message is displayed on the screen, press F11 to display the boot options.

(2) Select an option similar to the Add boot option. The text for this option may vary, depending on your system.

(3) Select File \efi\boot\bootx64.efi on the disk where ESXi is installed.

(4) Change the boot order so that the host boots from the added option.

9. Setting the Microsoft SQL database to an unsupported compatibility mode causes VCenter Server installation or upgrade to fail. VCenter Server installation using the Microsoft SQL database fails when the database is set to a version that is not supported for compatibility mode.

Fault Status:

The following error message is displayed: The database user you entered does not have the necessary permissions to install and configure VCenter Server with the selected database. Please correct the following error:%s

Fault Analysis:

The database version must be a supported version of VCenter Server. For SQL, this error can occur even if the database is a supported version, but if it is set to run in a version compatibility mode that is not supported. For example, this error can occur if SQL 2008 is set to run in SQL 2000 compatibility mode.

Solution:

Make sure that the VCenter Server database is a supported version and that it is not set to run in compatibility mode with unsupported versions.

10. Mistakenly delete the running virtual machine, through the XX-FLAT.VMDK recovery method

Fault status: Mistakenly deleted the running virtual machine, into the directory to view, only the Xx-flat.vmdk file, from the file type, only the files format, not virtual disk format, new virtual, choose to add existing disk, the hint does not exist

Solution:

(1) Create a new virtual machine instead of creating a hard disk

(2) SSH login host to find the location and directory of the XX-FLAT.VMDK files,

(3) Create a new XXX.VMDK file in the same directory as the above file, the size is as large as the xx-flat.vmdk file,

Use Ls-la to view the xx-flat.vmdk file size,

Create a new disk file with Vmkfstools-c file size-a lsilogic XXX.VMDK

(4) Add this disk file to the newly created virtual machine;

(5) Overwrite the new XXX-FLAT.VMDK with the original file XX-FLAT.VMDK (note must be-FLAT.VMDK), use the MV command

(6) Turn on the virtual machine when you are done.

Vmware vsphere FAQs and Solutions