New update available - Opslogix VMware Management Pack (V.26.10.4540.0)

We are happy to announce that a new release of our Opslogix VMware Management Pack for SCOM (V.26.10.4540.0) is now available.

You can download the latest version from our Customer Portal.

New updates:

• Added: CustomFieldsInclude and CustomFieldsExclude overrides on the VMware Virtual Machine discovery to choose which vCenter custom attributes are stored in the CustomFields property (regex on the attribute name). Use CustomFieldsExclude ^NB_ to leave out NetBackup attributes such as NB_LAST_BACKUP, which otherwise cause a configuration change for every VM after each backup. Default: all attributes, as before.
• Added: Automatic license assignment. A new "Automatically assign available licenses" checkbox on the Assign tab of the VMware MP Configuration dashboard enables the new rule "VMware license auto-assign" (disabled by default, runs every 8 hours on the All Management Servers Pool). It licenses ESXi hosts that are not licensed yet while free licenses remain, never exceeding the license count, and releases the licenses of hosts that no longer exist in vCenter. Optional assignment rules set which hosts are licensed first (priority regexes on the host path) and which are never licensed automatically (exclude regex). A warning alert lists hosts that could not be licensed. The setting is saved in the unsealed management pack OpsLogix.IMP.VMWare.LicenseAutoAssign.Overrides; changing it requires the rights to create management packs (normally SCOM Administrator).
• Added: Set-SCOMVMwareHostLicences -AutoAssign runs the same automatic assignment from PowerShell (supports -WhatIf), with -PriorityPatterns, -ExcludePattern and -ReleaseMissingHosts.
• Changed: Collector appsettings.json default log level lowered from Trace to Warning.
• Changed: The attributes in the VM CustomFields property are now ordered by name. After the upgrade, VMs with two or more custom attributes can get a one-time configuration change at the first discovery.
  
• Changed: The license inventory now includes the datacenter and cluster path of each host, shown in the Assign tab and used by the assignment rules.
• Fixed: Inventory/Whitelist Check monitor probe failed to initialize with NotImplementedException: check because the "check" command had no routing case in GetWorkflowInfo, leaving the monitor in an uninitialized state for every vCenter connection.
• Fixed: VM Guest OS FullName property was empty for VMs where VMware Tools is not installed or not running; it now falls back to the guest OS configured on the VM.
  
• Fixed: Set-SCOMVMwareHostLicences counted hosts instead of sockets, assigned every host of the connection instead of only the matching ones, stopped at the first host that did not fit with -SkipOverlicensed, did not support scaled-out connections, and failed when removing licenses from a connection without whitelist.
• Fixed: Add-SCOMVMwareConnection failed with a NullReferenceException because the license key was not loaded.
• Fixed: The per-workflow UseCaching registry setting (HKLM\SOFTWARE\OpsLogix\VMware\Client Modules\<ACTION>\<TYPE>) was never read because the key path used a forward slash; only the global setting worked. Boolean registry settings (UseCaching, IsBatchLogicalSet, EmulationMode) now also accept 1/0 (REG_DWORD or REG_SZ) besides true/false.
• Fixed: The vCenter Server Appliance (VCSA) health monitors never received data because the health summary used the object ID Appliance.<product> while the appliance is discovered as Appliance-<product>. The monitors now reflect the appliance health.
• Fixed: The ESX Host HBA multipath Status monitor never received data because it read the property host.config.multipathState.path.all while the collector sends host.config.multipathInfo. It now evaluates the dead and disabled paths: no dead or disabled paths = Healthy, any dead path = Critical, only disabled paths = Warning. Before, a host with active and dead paths could match Healthy and Critical at the same time. Custom overrides of HealthyValue/CriticalValue on this monitor must be reviewed.
• Fixed: The ESX Host in Standby Mode monitor could never detect standby because runtime.standbyMode (entering, exiting, in, none) was parsed as a boolean. A host that is in standby or entering standby is now reported as in standby.
• Fixed: The VMware Appliance Network ETH0 Drop rule collected the receive activity (net.rx.activity.eth0) instead of dropped packets. It now collects net.rx.drop.eth0 under the new counter name "Network ETH0 RX Drop (Average)", and a new rule collects net.tx.drop.eth0 as "Network ETH0 TX Drop (Average)". Historical data under "% Network ETH0 Drop (Average)" is receive activity in kB/s, not drops.
• Fixed: Property bag caching (UseCaching, on by default for monitoring) never skipped anything when a workflow sent no more property bags than the batch size (350 for monitoring): sent property bags were only remembered in the batched code path. Unchanged monitoring data is now skipped for the cache TTL (CacheTtlSec, default 30 minutes), which lowers the load on the management servers. Guest disk data is not cached, so the guest disk free space monitors still get every sample.

Haven't had the chance to try the Opslogix VMware Management Pack yet?

Start your 30-day free and non-binding trial below!

Are you unable to find the new version in the portal? Contact us to try the newest version or to get assistance updating your license key.