There are some new features in the new VMware released some days back a vCenter Server 5.1 update 2 build 1473063. This update enables the customization for windows 8.1, windows 2012 R2 and SLES 11 SP3 and many more. Additionally there has been multiple fixes released which hopefully might interest you or your clients using 5.1.
To be able to point out the different upgrade scenarios, if you are running on the 5.1 release, you can only use the custom installation option not the easy install, to upgrade to subsequent 5.1 update or patch releases. You will have to first upgrade vCenter sign on, followed by vCenter inventory service and vCenter Server.
In addition to vCenter 5.1 U2 there u=is also new ESXi 5.1 U2 release which matches the newly available features of vCenter.
New features of VMware vCenter Server 5.1 Update 2:
- Possibility to run vCenter Server on Windows Server 2012 R2.
- Database support for Microsoft SQL Server 2012 SP1
- Support for customization of the following guest operation systems:
- Windows 8.1
- RHEL 6.4
- Windows Server 2012 R2
- SLES 11 Service Pack 3
Resolved issues section shows that some of the OpenSSL libraries have been patched:
Update to OpenSSL version openssl-0.9.8y addresses multiple security issues
The OpenSSL library is updated to version openssl-0.9.8y to address multiple security issues. The Common Vulnerabilities and Exposures project (cve.mitre.org) has assigned the names CVE- 2013-0169 and CVE-2013-0166 to these issues.
More resolved issues:
Storage issues- Virtual disk I/O performance might degrade due to the misaligned disks of the vCenter Server Virtual Appliance. this is due to the misalignment in the disk partition of the vCenter Server Virtual appliance, thus there might be an overall degrade in the virtual disk I/O performance.
Attempts to deploy vApp on vCloud Director fails – Attempts to deploy vApp on vCloud Director fails with the Read timed out error message. This issue occurs due to delay in serializing the SPS profile.
- Impossible to log in to the vCenter Server Virtual Appliance through vSphere Client using Windows session credentials.
- Random failures of VMware VirtualCenter Server service.
- Service hostd might stop responding and generates hostd-worker dump.
- Message displayed that the access is unauthorized in the Log Browser in the vSphere Web Client.
- Error after upgrade in the Past Day Stats Rollup job.
- Impossible to create thin disks on a SDRS cluster if the disk size is larger than individual datastore.
- Disconnections of ESXi host from vCenter because of to sfcbd exhausting inodes
- Impossibility accessing the VMFS datastore or some files.
- VMware VirtualCenter Server service might fail if disk I/O usage on the vCenter Server system is high.
There have been a lot of fixes done:
The issues below have been addressed:
- ESXi hosts disconnect from vCenter Server and cannot be reconnected to the vCenter Server. This issue is caused by the hardware monitoring service (sfcdb) that populates the /var/run/sfcb directory with over 5000 files.
- Incorrect error messages similar to the following might be displayed by CIM providers:
- “Request Header Id (886262) != Response Header reqId (0) in request to provider 429 in process 5. Drop response.”
- The Hardware Status tab fails to display health statuses and displays an error message similar to the following: “Hardware monitoring service on this host is not responding or not available.”
- In an ESXi host, when reading the Field Replaceable Unit (FRU) inventory data using the Intelligent Platform Management Interface (IPMI) tool, the vmklinux_9:ipmi_thread of vmkapimod incorrectly displays the CPU usage as hundred percent.
- The host IPMI System Event Log is not getting cleared in a cluster environment.
- Some 3D applications display misplaced geometry on a Windows 7 or Windows 8 virtual machine if the Enable 3D support option is selected.
- Netlogond might consume high memory in Active Directory environment with multiple unreachable Domain Controllers. As a result, Netlogond might fail and ESXi host might lose Active Directory functionality.
- Virtual machine becomes unresponsive due to a signal 11 error, while executing SVGA code in svga2_map_surface.
- An ESXi host creating virtual machines on a remotely mounted vSphere Storage Appliance (VSA) datastore might stop responding. This is due to the VSA Manager plugin not handling network errors correctly.
- ESXi host experiences a purple diagnostic screen with errors for E1000PollRxRing and E1000DevRx when the rxRing buffer fills up and the max Rx ring is set to more than 2. The next Rx packet received that is handled by the second ring is NULL, causing a processing error.
- Whenever you reset a management network where the management traffic is enabled on multiple VMkernel ports, the IP address displayed on the Direct Console User Interface (DCUI) changes.
- In vCenter performance chart, the net.throughput.usage related value is in Kilobytes, but the same value is returned in bytes in the vmkernel. This leads to incorrect representation of values in the vCenter performance chart.
- When you provision and customize a virtual machine from a template on a vDS with Ephemeral Ports, the virtual machine might lose connection from the network.
- The ESXi userworld OpenSSL library is updated to version openssl-0.9.8y to resolve multiple security issues.
- The ESXi userworld libxml2 library has been updated to resolve a security issue.
- VMware ESXi and ESX contain a NULL pointer dereference in the handling of the Network File Copy (NFC) traffic. To exploit this vulnerability, an attacker must intercept and modify the NFC traffic between ESXi/ESX and the client. Exploitation of the issue might lead to a Denial of Service.
- When you attempt to remove multiple NFS datastores using the Host Profile option, an error occurs because there might be a datastore that has already been deleted.
- On an ESXi host when bandwidthCap and throughputCap are set at the same time, the I/O throttling option might not work on virtual machines.
- When a SCSI Request Sense command is sent to Raw Device Mapping (RDM) in Physical Mode from a guest operating system, sometimes the returned sense data is NULL (zeroed).
- VMware Tools is updated to provide pre-built modules for SUSE Linux Enterprise 11 SP3, Oracle Linux 5.x with 2.6.39-200/300/400 kernels and Oracle Linux 6.x with 2.6.39-200/300/400 kernels
VMware vCenter Server
The issues below have been addressed:
- After you deploy ESXi 5.1 on a host which previously had ESX 4.1 installed, the ESXi host might fail to reconnect to the vCenter Server and display a licensing error.
- License capacity monitoring alarms are triggered in vCenter Server 5.x
- Attempts to deploy a virtual machine on a vSphere Distributed Switch result in an error message stating that no free ports are available
- You might lose access to NFS mounted data stores after you reboot a stateless ESXi host
- Virtual disk I/O performance might degrade due to the misaligned disks of the vCenter Server Virtual Appliance
- After performing a maintenance upgrade on vCenter Server 4.x, if you upgrade to vCenter Server 5.x, ESX/ESXi 4.x hosts might not connect to vCenter Server
- Syntax errors encountered when you run VCDB_oracle.sql from the vCenter Server 5.1 install media
- Attempts to upgrade vCenter Server from version 4.x to 5.x might fail with error 25004
- Attempts to migrate custom attributes with duplicate or identical values to tags fails
- After you replace SSL certificates, vCenter Server 5.1 might fail to start if multiple service IDs are included in the vpxd.cfg file.
- Unable to sort virtual machines by provisioned or used space in the vSphere Web Client
- Data store performance chart data missing if the data store cluster contains more than 10 data stores
- vSphere Web Client displays incorrect IPMI/iLO information
- Instead of displaying the device or file system identifier, vSphere Web Client displays } in the All Paths Down (APD) event log.
- If you restart the vCenter Server after creating a snapshot, the snapshot might incorrectly point to the base VMDK until you rename the snapshot.
- In an environment where ESXi hosts are connected to a vSphere Distributed Switch (vDS) version 5.1 and vDS health check is enabled, incorrect vDS health status might be displayed for the ESXi hosts after you reconfigure the MTU or VLAN setting of the vDS.
- If JGroups in Inventory Service does not converge fast, the vSphere Web Client might report connection failure to the ESXi host.
- In an environment where a cluster has advanced configuration options for HA that affect admission control, after you restart the vpxd service, the HA admission control might not detect the user settings until you reconfigure the cluster.
VMware vCenter Update Manager
The issues below have been addressed:
- When you attempt to remediate a host that is in a DRS cluster and contains powered on virtual machines that are connected to datastore ISO files, if both the virtual machine and the datastore ISO files are on a shared storage, Update Manager fails to migrate these virtual machines to another host. As a result, Update Manager cannot enter the host into maintenance mode, and remediation fails.
- When you upgrade UMDS 5.1 to 5.1 Update release version, the information about third-party portals for downloading ESX/ESXi patches is deleted in the process of upgrading the UMDS.
- When you use SQL Server Native Client 11.0 to connect to the Microsoft SQL Server 2012 database that Update Manager uses, the Update Manager service stops if your vSphere inventory contains virtual appliances.
- When you restart the VMware VirtualCenter Server service on a Windows or a Linux system, the settings about excluded patches in dynamic patch baselines are lost.
- While creating a new baseline or editing an existing baseline, if you add bundles to the baseline that you imported offline, the New Baseline or the Edit Baseline wizard might stop responding on the Criteria page.
- If your inventory is empty, the Update Manager Service takes about 5 seconds to start or restart. If you have nested vApps and resource pools in your inventory, the Update Manager Service might take an excessive amount of time to start or restart. The larger the number of nested vApps and resource pools in your vSphere inventory, the longer it takes the Update Manager Service to start or restart. The amount of time might be fifteen or more minutes.