|
This document describes changes and updates that have occurred since the release of BMC Performance Manager Express for Hardware v2.7.03.
What's New
Coverage
| • | Cisco UCS B-Series Blade Chassis, through the UCS Manager |
| • | Data Domain Storage Appliance, through the Data Domain SNMP agent |
| • | HP StorageWorks MSA external disk enclosures, directly attached to an HP ProLiant system and managed through the HP Insight Management Agents |
| • | Version 6.1.2 of the IBM Director Agent on IBM xSeries servers (Note: IBM Director 6.1.0 and 6.1.1 are not supported as they do not provide the necessary hardware information) |
| • | Adaptec SAS controllers on Linux and Sun Solaris systems, through the arcconf command line utility |
| • | HP Smart Array RAID controllers on HP-UX systems, through the sautil command line utility |
| • | LSI RAID controllers managed through the MegaCli and lsiutil command line utilities on Windows and Sun Solaris systems |
| • | Fiber channel HBA adapters on VMware ESXi systems |
Parameters
| • | "Degrees Below Warning" in the Capacity Report class (SEN_HW_CAPACITYREPORT) represents the number of degrees before reaching the closest warning threshold. It enables administrator to optimize the data center temperature. |
| • | "ConnectedToPhysicalAddress" (SEN_HW_NETWORK) provides, when available, the physical address of the port to which the network interface is connected (MAC address for an Ethernet port, WWN address for an FC Port). |
Input Properties
| • | The input property "SNMP Port" can now be used to specify the port number performing SNMP queries. |
Changes and Improvements
| • | On EMC disk arrays, temporary volumes created for the sole purpose of mirrors, snapshots and clones are no longer discovered and monitored |
| • | On SMI-S compliant disk arrays (EMC, Hitachi, HP EVA), dormant devices no longer trigger a warning |
| • | When monitoring a remote Linux system, the “Linux – Network” connector’s availability is now fully tested. In case of a failure (credentials, network connection, etc.), the corresponding network card objects triggered alarms instead of being put offline and marked as “failed connector” |
| • | On IBM xSeries servers running Linux, logical volumes (RAIDs) are no longer reported as physical disks |
| • | On NetApp filers, aggregates flagged as “redirect” are no longer reported in warning |
| • | On Sun Solaris systems, the number of errors that occurred on memory modules is reported with the ErrorCount parameter |
| • | On Sun Solaris systems, illegal requests on physical disks are no longer reported and therefore no longer trigger alerts on the ErrorCount parameter |
| • | On Sun Solaris systems, unplumbed network interfaces are now also discovered and monitored |
| • | On Sun M3000, M4000, M5000, M8000 and M9000 systems, the following objects are now monitored: DC-to-DC converters, I/O expansion units (I/O boards, power supplies, uplink and downlink cards), CPU and memory boards, Crossbar and Clock units, Operator panel and XSCF units. Therefore, the power consumption evaluation is now more accurate |
| • | On Cisco Telnet Switches, monitoring of the administrative status of SFPs (Isolation Errors, SFP Suspended, etc...) was added (Only link status was previously monitored). Support for MDS 9020 model has also been implemented; the preferred method for monitoring MDS 9020 switches is now Telnet / SSH |
| • | Additional information for BIOS, firmware and driver versions is now displayed in the Locator parameter of the Disk Controllers |
Fixed Issues
Cisco
| • | On Cisco MDS fiber switches, the status of clocks is now properly reported |
| • | On Cisco MDS fiber switches with a high number of ports, the product failed to discover the fiber ports |
Fujitsu-Siemens
| • | On Fujitsu-Siemens PRIMERGY systems, instances were created for disabled temperature sensors |
| • | On some Fujitsu-Siemens PRIMERGY systems, failed batteries in the RAID controllers were not properly reported |
Hitachi
| • | Hitachi AMS Disk Arrays could not be monitored because the Performance Manager failed to recognize the HiCommand Device Manager as a valid SMI-S compliant agent |
HP
| • | On older HP ProLiant systems, an error message was displayed in the System Output Window of the Portal about the collection of the enclosure status |
| • | On some HP ProLiant systems with no SNMP agent enabled, the status of power supplies was incorrectly interpreted |
| • | On some HP NetServer systems, the HP Instant TopTools agent was not properly located |
| • | On HP-UX systems, some physical disk objects were not properly attached to the appropriate disk controller instance or sometimes were not discovered at all |
IBM
| • | On some IBM xSeries systems with the IBM Director Agent 5.2.x, incorrect thresholds were set for voltage sensors, leading to false alerts |
| • | On some IBM AIX systems, the Performance Manager was not able to report the correct number of Memory modules |
| • | On IBM AIX systems, some disk controllers were not properly detected |
| • | On IBM AIX systems, physical disks were sometimes undetected or marked as missing |
| • | On IBM DS disk arrays managed through SMCli, devices reported as “Near expiration” didn’t trigger a warning |
| • | On IBM 3584 tape libraries, the status of robotic pieces was not properly collected |
Sun
| • | On Sun Solaris systems with SNMP enabled, some network interfaces were incorrectly reported as physical adapters, e.g. lo0, lpfs0, jnet, dman, aggr0, clprivnet, sppp0 and similar |
| • | On Sun Solaris systems, the status of physical disks with no s2 slice was not properly tested (typically when hosting a ZFS filesystem) |
| • | On Sun Solaris systems, a failure of the picld daemon (or the underlying instrumentation chip and driver) caused false alarms on fans, LEDs and power supplies |
| • | On Sun Solaris systems, Intel and AMD multi-core processors were not completely discovered |
| • | On Sun Solaris systems, tape drives were sometimes not properly detected |
| • | On Sun Solaris systems, false alarm that was triggered on self-regulating fans |
| • | On Sun Solaris systems, Fan Speed calculation did not always return the correct value |
| • | On Sun M8000 and M9000 systems, processors and memory modules on separate boards were not properly discovered and monitored |
| • | On Sun M3000, M4000, M5000, M8000 and M9000 systems, power supplies were sometimes reported with an unknown status |
VMware
| • | On some servers running VMware ESXi, the discovery failed to recognize SMI-S or SMASH-compliant disk controllers |
| • | Depending on the model and vendor of the server, systems running VMware ESXi (3.0, 3.5 or 4.0) were not fully supported and disk controllers, disks and volumes were not properly monitored |
|
