Due to a bug in the authentication module in multi-domain environments VMware released another update of vCenter: vCenter 5.1 Update 1a:
VMware releases another update for vCenter, 5.1 Update 1b:
The long anticipated first major Update bundle of ESXi and vCenter 5.1 has finally been released. Download them at the usual place.
The insane list of fixes confirms my gut feeling again that unfortunately, many VMware products only start getting usable after the first (or sometimes even the second) Update bundle. (Remember vCenter 5.1a and 5.1b or the loads of support alerts?)
ESXi 5.1 Update 1
Go and check the release notes. Seriously.
No real new features or enhancements have been added apart from a few new supported Guest OSes.
But huge loads of important issues and bugs have been fixed, a few of which were anticipated since a long time. Here are some excerpts to highlight some of the important or interesting fixes:
- ESXi 5.x host appears disconnected in vCenter Server and logs the ramdisk (root) is full message in the vpxa.log file
If Simple Network Management Protocol (SNMP) is unable to handle the number of SNMP trap files (.trp) in the /var/spool/snmp folder of ESXi, the host might appear as disconnected in vCenter Server. You might not be able to perform any task on the host.
- Use of the invoke-vmscript command displays an error
When you use the invoke-vmscript PowerCLI command scripts on a virtual machine, the script fails with the following error message:
The guest operations agent could not be contacted.
— Mentioned this issue here —
- ESXi hosts might fail with a purple diagnostic screen when you attempt to plug in or unplug a keyboard or mouse through a USB port
When you attempt to plug in or unplug a keyboard or a mouse through the USB port, the ESXi host might fail with the following error message:
PCPU## locked up. Failed to ack TLB invalidate.
- Component-based logging and advanced configurations added to hostd log level
To avoid difficulties in getting appropriate logs during an issue, this release introduces component-based logging by dividing the loggers into different groups and prefixing them. Also, new advanced configuration allows you to change hostd log’s log level without restarting.
- ESXi hosts might fail if hostd-worker thread consumes 100% CPU resources
Under sufficiently high workload on the ESXi host, hostd-worker thread might get stuck consuming 100% CPU while fetching the virtual machine screenshot file for vCloud Director UI. This issue might result in the failure of the ESXi host.
- Long running vMotion operations might result in unicast flooding
When using the multiple-NIC vMotion feature with vSphere 5, if vMotion operations continue for a long time, unicast flooding is observed on all interfaces of the physical switch. If the vMotion takes longer than the ageing time that is set for MAC address tables, the source and destination host start receiving high amounts of network traffic.
- ESXi host stops responding with a purple diagnostic screen during arpresolve
The ESXi host might stop responding during arpresolve and display a purple diagnostic screen
— ARP is serious business. Is that some single-frame layer 2 DoS vector? —
- Network connectivity on IPv6 virtual machines not working with VMXNET3
When more than 32 IPv6 addresses are configured on a VMXNET3 interface, the unicast and multicast connectivity to some of those addresses are lost.
- Virtual machine might lose network connectivity from external environment after vMotion with vNetwork Distributed Switch environment
A virtual machine might lose network connectivity from the external environment after vMotion with vNetwork Distributed Switch environment.
- Attempts to apply host profile might fail with an error message indicating that the CIM indication subscription cannot be deleted
- Hardware Status tab might stop displaying host health status
On an ESXi 5.1 host, Small-Footprint CIM Broker daemon (sfcbd) might fail frequently and display CIM errors. As a result, Hardware Status tab might stop displaying host health status and syslog.log might have error message similar to the following:
Timeout (or other socket error) sending request to provider.
- Unable to delete files from the VMFS directory after one or more files are moved to it
After moving one or more files into a directory, an attempt to delete the directory or any of the files in directory might fail.
Accessing corrupted metadata on VMFS3 volume might result in ESXi host failure
If a file’s metadata is corrupted on a VMFS3 volume, ESXi host might fail with a purple diagnostic screen while trying to access the file. VMFS file corruption is extremely rare but might be caused by external storage issues.
- Adding new ESXi host to a High Availability cluster and subsequently reconfiguring the cluster might result in the failure of any other host in the cluster with purple diagnostic screen
When a new ESXi host is added to a High Availability (HA) cluster and the HA cluster is subsequently reconfigured, any other host in the existing HA cluster might fail with a purple diagnostic screen
- When the quiesced snapshot operation fails the redo logs are not consolidated
When you attempt to take a quiesced snapshot of a virtual machine, if the snapshot operation fails towards the end of its completion, the redo logs created as part of the snapshot are not consolidated. The redo logs might consume a lot of datastore space.
- iSCSI LUNs do not come back online after recovering from the APD state
After recovering from the All-Paths-Down (APD) state, iSCSI LUNs do not come up until a host reboot. This issue occurs on Broadcom iSCSI offload-enabled adapters configured for iSCSI.
- ESXi host might fail with a purple diagnostic screen if you run the vmware-vimdump command from DCUI
When you run the vmware-vimdump command from Direct Console User Interface (DCUI), the ESXi host might fail with a purple diagnostic screen. This might also result in missed heartbeat messages. This issue does not occur when the command is run by connecting through an SSH console.
— What. —
- Reinstallation of ESXi 5.1 does not remove the Datastore label of the local VMFS of an earlier installation
Reinstallation of ESXi 5.1 with an existing local VMFS volume retains the Datastore label even after the user chooses the overwrite datastore option to overwrite the VMFS volume.
- Microsoft Windows Deployment Services (WDS) might fail to PXE boot virtual machines that use the VMXNET3 network adapter
Attempts to PXE boot virtual machines that use the VMXNET3 network adapter by using the Microsoft Windows Deployment Services (WDS) might fail with
- resxtop fails when upgraded from vSphere 5.0 to vSphere 5.1
In vSphere 5.1, SSL certification checks are set to ON. This might cause resxtop to fail in connecting to hosts and displays an exception message similar the following:
HTTPS_CA_FILE or HTTPS_CA_DIR not set.
- VMRC and vSphere Client might stop responding when connected to a failed virtual machine
On an ESXi 5.1 host, VMware Remote Console (VMRC) and vSphere Client might stop responding when connected to a failed virtual machine or virtual machine with failed VMware Tools.
- Time synchronization with the ESXi server might result in an unexpected reboot of the guest operating system when an ESXi host is configured as an NTP server
When an ESXi host is configured as an Network Time Protocol (NTP) server, the guest operating system might unexpectedly reboot during time synchronization with the ESXi host. This issue occurs when the virtual machine monitoring sensitivity level is set to High on a High Availability cluster and das.iostatsInterval option is set to False.
- VMware Tools might fail while taking a quiesced snapshot of a virtual machine
If non-executable files are present in the backupScripts.d folder, VMware Tools might fail while taking a quiesced snapshot of a virtual machine.
- After VMware Tools installation the guest operating system name changes from Microsoft Windows Server 2012 (64-bit) to Microsoft Windows 8 (64-bit)
After you create Microsoft Windows Server 2012 (64-bit) virtual machines and install VMware Tools, the guest operating system name changes from Microsoft Windows Server 2012 (64-bit) to Microsoft Windows 8 (64-bit).
- VMware Tools might leak memory in Linux guest operating system
When multiple VLANs are configured for network interface in Linux guest operating system, VMware Tools might leak memory.
- On an ESX/ESXi host earlier than version 5.1, upgrading only VMware Tools to version 5.1 results in a warning message
On an ESX/ESXi host earlier than version 5.1 and with a virtual machine running Windows guest operating system, if you upgrade only VMware Tools to version 5.1, a warning message similar to the following might be displayed in Windows Event Viewer:
[ warning] [vmusr:vmusr] vmware::tools::UnityPBRPCServer::Start: Failed to register with the host!
- Attempts to install VMware Tools might fail with Linux kernel version 3.7
VMware Tools drivers are not compiled as the VMware Tools installation scripts are unable to identify the new kernel header path with Linux kernel version 3.7. This might cause VMware Tools installation to fail.
- Customization of guest operating system might fail when deployed from some non-English versions of Windows guest operating system templates
Customization of guest operating system might fail when deployed from some non-English versions of Windows guest operating systems templates, such as the French version of Microsoft Windows 7, the Russian version of Microsoft Windows 7 and the French version of Microsoft Windows Server 2008 R2 guest operating systems. This issue occurs when the VMware Tools service vmtoolsd.exe fails.
- Virtual machines with vShield Endpoint Thin Agent might encounter performance-related problems when you copy network files to or from a CIFS share
You might encounter performance-related problems with virtual machines while copying network files to or from a Common Internet File System (CIFS) share.
This issue occurs when virtual machines running vShield Endpoint Thin Agent available from the VMware Tools bundle are used.
Be aware of the lengthy list of still present known issues too:
Contrary to what’s stated in the release notes, the HCL does not display information for ESXi 5.1 Update 1 yet. But the Product Interoperability Matrix has been updated.
It shows that the 5.1 U1 VMware Tools are supposedly not supported for 5.0 hosts, which is odd since they are still listed as supported on 4.1 U3 and 4.0 U4. The 5.1 U0 Tools were fully supported on 5.0 hosts too. The information is probably just not updated properly yet.
vCenter Server 5.1 Update 1(a)
Go and check the release notes. Seriously.
A couple of new enhancements have been added with Update 1, namely:
- vCenter Server is now supported on Windows Server 2012
- vCenter Server now supports the following databases:
Microsoft SQL Server 2012
Microsoft SQL Server 2008 R2 SP2
- vCenter Server now supports customization of the following guest operating systems:
Windows Server 2012
(Actually, Windows 8 and 2012 were supported previously too – though there was a bug with the customization process)
- vCenter Essentials no longer enforces vRAM usage limit of 192 GB
With vSphere 5.1 Update 1, the Essentials and Essentials Plus licenses no longer restrict virtual machine power-on operations when the vRAM usage limit of 192 GB is met
Like with ESXi 5.1 U1, there is a long list of fixed bugs and issues. A couple of serious security vulnerabilities, especially in the vCenter Appliance department have been fixed too. Here are a few taken from the release notes:
- Backup of the Inventory Service database fails
A backup operation of the Inventory Service database while the Inventory Service is running fails due to a bad_certificate error.
— Certificate issues in VMware land – I’m lovin’it. —
- Unable to add ESXi 5.1 hosts to existing vSphere Distributed Switch versions 4.0, 4.1, and 5.0 in vCenter Server 5.1 with compatibility issue
When you upgrade from vCenter Server 4.0, 4.1, or 5.0 to vCenter Server 5.1, adding ESXi 5.1 hosts to the existing vSphere Distributed Switch (vDS) versions 4.0, 4.1, and 5.0 might fail. However, if you create new vDS switch versions 4.0, 4.1, 5.0 or 5.1 after upgrading to vCenter Server 5.1, you will be able to add ESXi 5.1 hosts.
- vCenter Server when deployed in an environment that uses Active Directory (AD) with anonymous LDAP binding enabled doesn’t properly handle login credentials
In this environment, authenticating to vCenter Server with a valid user name and a blank password might be successful even if a non-blank password is required for the account.
The Common Vulnerabilities and Exposures project (cve.mitre.org) has assigned the name CVE-2013-3107 to this issue.
- vCenter VAMI UI allows arbitrary code execution
The vCenter Server Appliance (vCSA) VAMI web interface contains a vulnerability that allows an authenticated remote attacker to upload files to an arbitrary location creating new files or overwriting existing files. Replacing certain files may result in a denial of service condition or code execution. In the default vCSA setup, authentication to vCSA is limited to root since root is the only defined user.
- Storage operations such as cold migration, storage vMotion, and cloning of virtual machine fail at 99%
Storage operations, including cold migration, storage vMotion, and cloning of a virtual machine with IDE disks and change block tracking enabled might fail at 99% when you use vCenter Server 5.1 to manage ESX/ESXi 4.x hosts.
An error message similar to the following is displayed:
A general system error occurred: Configuration information is inaccessible.
- Single Sign On (SSO) upgrade from vCenter Server 5.1 0a to vCenter Server 5.1 0b does not replace the sspiservice.exe and rsautil.cmd files
After you upgrade SSO from vCenter Server 5.1 0a to vCenter Server 5.1 0b, the sspiservice.exe and rsautil.cmd files are not replaced. When you run the rsautil -v command, the resulting version number is of vCenter Server 5.1 0a.
- Unable to edit the settings of a virtual machine that is a member of a datastore cluster
In vCenter Server, you cannot edit the settings of a virtual machine that is a member of a datastore cluster by using a user account that does not have the permission to configure a datastore cluster in vCenter Server but has full permissions at the virtual machine level. The following error message is displayed:
Permission to perform this operation was denied.
You do not hold privilege “Datastore cluster > Configure a datastore cluster on Datastore cluster Cluster Name”
- Unable to see CPU, memory, disk metrics on the cluster view in the performance chart
The CPU, memory, disk, and network metrics are absent from the cluster view under the performance option of the Advanced tab in vSphere Client.
— Whoa I always wondered where these stats were on the cluster level. —
- vCenter Server becomes non responsive when validating permissions for unknown objects
When you move a host that has permissions on child entities or itself out of a cluster without removing permissions and also remove the host from the vCenter inventory, the vCenter Sever freezes and become non responsive whenever periodic validation takes place.
- vCenter Server 5.1 service fails to start after you restart the server
The vCenter Server service fails to start if the Single Sign On service cannot connect to the database. This occurs when the Single Sign On service starts before the database service.
- Updating the Base DN for groups when editing an identity source is not working correctly
In the Edit Identity Source dialogue for SSO, after you information for Base DN for groups such that it differs from that for the Base DN for users, save the changes, and return to the Edit Identity Source dialog box again, you notice that the Base DN for groups text box displays the same as the Base DN for users. This issue does not occur when you initially add the identity source.
- Performance history for past year might contain only 30 days of information in vSphere 5.1
When you attempt to view the vSphere performance history for past year in the past year view tab, you can see only one month of performance history.
- Logging in to vCenter Server through the vSphere Web Client fails if you specify a non-ASCII user name
If you provide a valid vCenter Server user name composed of non-ASCII characters and attempt to log in using the vSphere Web Client, the login attempt fails with the following error:
Provided credentials are not valid.
- vSphere 5.1 Web Client advanced performance charts are slow to display
In the vCenter 5.1 web client, the advanced performance charts takes several minutes to load.
— Yeah, I really hope the dogslowness has been fixed. Maybe this fixes this annoying for input string error too?—
- Attempt to install Red Hat Enterprise Linux using physical media on Windows XP fails when the installer is running
When you connect to any ESX/ESXi host through vSphere Client running on Windows XP, attempts to install Red Hat Enterprise Linux by using client-side physical media redirected from vSphere Client running on Windows XP fails when the installer is running.
— This specific issue made my day. —
- Virtual Machine snapshot size (GB) and VM Total Size on Disk (GB) alarms are triggered incorrectly
In vCenter Server 5.1 you can configure the vCenter Server to send alarms when virtual machine size on disk exceeds a limit or when virtual machine snapshot size exceeds a limit. These alarms are falsely triggered when the virtual machine size or snapshot size are within the set limits. The alarm for virtual machine snapshot size might even be triggered when no snapshot exists.
- Cloning a virtual machine through vSphere Client or vSphere Web Client causes the resulting virtual machine to have its disk pointing back to the source virtual machines disk
In vCenter Server, when you clone a virtual machine through vSphere Client or vSphere Web Client, you have the option to edit the hardware of the destination virtual machine. If you choose to edit a disk and adjust its size, the disk of the resulting virtual machine points back to the source virtual machine disk. This results in the destination virtual machine using the source virtual machine disk.
- Virtual machines are unable to connect to the network after reverting to a snapshot
In vCenter Server, when a snapshot is reverted and virtual machine is powered on, Connect At Power On is selected on the network adapter but Connected is deselected. If you select the Connected check box an error similar to the following is displayed:
Invalid Configuration for device 0.
- vSphere 5 Storage vMotion is unable to rename virtual machine files on completing migration
In vCenter Server , when you rename a virtual machine in the vSphere Client, the VMDK disks are not renamed following a successful Storage vMotion task. When you perform a Storage vMotion task for the virtual machine to have its folder and associated files renamed to match the new name, the virtual machine folder name changes, but the virtual machine file names do not change.
This issue is resolved in this release. To enable this renaming feature, you need to configure the advanced settings in vCenter Server and set the value of the provisioning.relocate.enableRename parameter to true.
— Remember having to set this manually. —
- Unable to configure High Availability on vCenter Server 5.1 with only IPv6 networks
You cannot configure High Availability (HA) on vCenter Server 5.1 that uses only an IPv6 environment, as vCenter Server does not take IPv6 addresses (from default gateway or given by you through the vSphere Client) as isolation addresses.
- Virtual machine power on fails with InsufficientFailoverResources if High Availability Admission Control is disabled
After all the hosts are inaccessible and only one host is left in a High Availability (HA) or Distributed Resource Scheduler (DRS) cluster, when you attempt to power on any more virtual machines even with the admission control is set to OFF, the following error message is displayed:
Check out the even longer list of currently known issues too:
It completely flew under my radar until Dennis commented on it, but there is finally a VMware Update Manager Web Client Plugin available with this release. There are separate Release notes for Update Manager 5.1 Update 1 available which I didn’t notice at all and expected to be mentioned in the vCenter notes.
Miscellaneous updated stuff
ESXi HP extensions
Meanwhile, HP has also updated their April vibsdepot release and now provides a full repository release including the Utility Bundle and NMI driver: http://vibsdepot.hp.com/hpq/apr2013/
But they still haven’t updated the main metadata XML, so you still won’t get this in Update Manager automatically. Currently you need to configure this additional Download Source URL manually:
To continue the confusing version management, the “Complete Bundle” is now at version “04.25.13”.
A new customized HP ISO featuring ESXi 5.1 U1 as well as the April HP updates is also available already:
Also, if you haven’t noticed until recently like me, VMware also released an update to the vSphere Management Assistant (vMA) a few weeks ago:
Updating will be a major PITA because of course, there is no real upgrade path – you need to deploy a completely fresh Virtual Appliance from scratch again.
Also of note, the VDDK has been updated too:
The initial 5.1 release of the VDDK apparently contained some serious bugs, which made some major backup vendors not implement support 5.1 until now:
VMware Converter Standalone 5.1
He’s not dead yet, Jim. This totally caught me off guard but it’s awesome to see the free Converter finally getting the proper update it deserved:
- Support for virtual machine hardware version 9
- Guest operating system support for Microsoft Windows 8 and Microsoft Windows Server 2012
- Guest operating system support for Red Hat Enterprise Linux 6
- Support for virtual and physical machine sources with GUID Partition Table (GPT) disks
- Support for virtual and physical machine sources with Unified Extensible Firmware Interface (UEFI)
- Support for EXT4 file system
VMware vCloud Director 5.1.2
Here comes a small vCD update too:
- Rights for creating, reverting, and removing snapshots: Rights for creating, reverting, and removing snapshots have been added, allowing system administrators to configure these rights for all roles.
- Allocation pool organization virtual datacenters can be elastic or non-elastic: Starting with vCloud Director 5.1.2, system administrators can configure Allocation Pool organization virtual datacenters with Single Cluster Allocation Pool (SCAP), making them non-elastic. This is a global setting that affects all Allocation Pool organization virtual datacenters. By default, Allocation Pool organization virtual datacenters have Single Cluster Allocation Pool enabled. Systems upgraded from vCloud Director 5.1 that have Allocation Pool organization virtual dataceters with virtual machines spanning multiple resource pools have Single Cluster Allocation Pool disabled by default.
- vCloud Director is now supported on Red Hat Enterprise Linux 6.3
- Support for Microsoft SQL Server 2012: vCloud Director now supports Microsoft SQL Server 2012 databases.
- Additional guest operating system customization support: vCloud Director now supports customization of the following guest operating systems:
Microsoft Windows Server 2012
The vSphere PowerCLI has not been updated, Release 2 from February is still the most recent version.
There is no Update for vShield/vCloud Networking and Security (apart of the vShield Endpoint driver shipped with the VMware Tools)