Guest

Cisco UCS Manager

Release Notes for Cisco UCS Manager, Release 1.2(1)

  • Viewing Options

  • PDF (356.6 KB)
  • Feedback
Release Notes for Cisco UCS Manager, Release 1.2(1)

Table Of Contents

Release Notes for Cisco UCS Manager, Release 1.2(1)

Contents

Introduction

System Requirements

Updating the Cisco UCS Firmware

Hardware and Software Interoperability

Known Limitations and Behaviors

Resolved Caveats

Open Caveats

Caveats from Previous Releases

Release 1.1(1)

Release 1.0(2)

Release 1.0(1)

New Hardware Features in Release 1.2(1)

Related Documentation

Hardware Documents

Software Documents

Obtaining Documentation and Submitting a Service Request


Release Notes for Cisco UCS Manager, Release 1.2(1)


First Published: March 29, 2010
Latest Update: May 24, 2010
Part Number: OL-22187-01

This document describes the new features, system requirements, and caveats for Cisco UCS Manager, Release 1.2(1b), Release 1.2(1d), and all related firmware on blade servers and other Unified Computing System components associated with that release. Use this document in conjunction with the documents listed in the "Related Documentation" section.


Note We sometimes update the documentation after original publication. Therefore, you should also review the documentation on Cisco.com for any updates. Documentation updates and errata are also in these release notes. The documentation roadmap for this product is available at: http://www.cisco.com/go/unifiedcomputing/b-series-doc


Table 1 shows the online change history for this document.

Table 1 Online History Change 

Part Number
Revision
Date
Description

OL-22187-01

A0

March 29, 2010

Created release notes for Release 1.2(1b).

 

B0

May 24, 2010

Updated release notes for Release 1.2(1d).


Contents

This document includes the following sections:

"Introduction" section

"System Requirements" section

"Updating the Cisco UCS Firmware" section

"Hardware and Software Interoperability" section

"Known Limitations and Behaviors" section

"Resolved Caveats" section

"Open Caveats" section

"New Hardware Features in Release 1.2(1)" section

"Related Documentation" section

"Obtaining Documentation and Submitting a Service Request" section

Introduction

The Cisco® Unified Computing System is a next-generation data center platform that unites compute, network, storage access, and virtualization into a cohesive system designed to reduce total cost of ownership (TCO) and increase business agility. The system integrates a low-latency, lossless 10 Gigabit Ethernet unified network fabric with enterprise-class, x86-architecture servers. The system is an integrated, scalable, multi-chassis platform in which all resources participate in a unified management domain.

System Requirements

Cisco UCS Manager must meet or exceed the following minimum system requirements:

The Cisco UCS Manager GUI is a Java-based application that requires Sun JRE 1.6 or later.

UCS Manager uses web start and supports the following web browsers:

Microsoft Internet Explorer 6.0 or higher

Mozilla Firefox 3.0 or higher

UCS Manager is supported on the following operating systems:

Microsoft Windows XP

Microsoft Windows Vista

Red Hat Enterprise Linux 5.0 or higher

Updating the Cisco UCS Firmware

To update the Cisco UCS software and firmware, refer to the following documents:

For upgrading from release 1.1(1) to 1.2(1) refer to

http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/upgrading/from1.1.1/to1.2.1/b_UpgradingCiscoUCSFrom1.1.1To1.2.1.html

For upgrading from release 1.0(2) to 1.2(1) refer to

http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/upgrading/from1.0.2/to1.2.1/b_UpgradingCiscoUCSFrom1.0.2To1.2.1.html

For upgrading from release 1.0(1) to 1.2(1) refer to

http://www.cisco.com/en/US/docs/unified_computing/ucs/sw/upgrading/from1.0.1/to1.2.1/b_UpgradingCiscoUCSFrom1.0.1To1.2.1.html

The upgrade procedure from earlier releases to Release 1.2.1c will be same as it was to 1.2.1b.

Use the scope firmware and show package filename expand CLI commands to view the contents of a given release package.

Table 2 Package Contents for UCSM Release 1.2(1b)

Package file
Component
Notes

ucs-2100.1.2.1d.bin

Fabric extender firmware

 

ucs-6100-k9-kickstart.4.1.3.N2.1.2d.bin

ucs-6100-k9-system.4.1.3.N2.1.2d.bin

Kernel firmware for the fabric interconnect

System firmware for the fabric interconnect

Both are needed, and both must be updated at the same time.

ucs-b200-m1-bios.S5500.1.2.1.0.030920101143.bin

ucs-b200-m1-k9-bmc.1.2.1d.bin

ucs-b200-m1-sasctlr.01.28.03.00_06.28.00.00_03.12.00.00.bin

ucs-b200-m2-bios.S5500.1.2.1.0.030920101143.bin

B200 M1 BIOS

CIMC firmware

RAID controller

B200 M2 BIOS

 

ucs-b250-m1-bios.S5500.1.2.1.0.030820102351.bin

ucs-b250-m1-k9-bmc.1.2.1d.bin

ucs-b250-m2-bios.S5500.1.2.1.0.030820102351.bin

B250 M1 BIOS

CIMC firmware

B250 M2 BIOS

 

ucs-m71kr-e-cna.1.2.1d.bin

ucs-m71kr-e-hba.2.80A4.bin

ucs-m71kr-e-optionrom.5.03A8.bin

UCS M71KR- E Converged Network Adapter firmware

 

ucs-m71kr-q-cna.1.2.1d.bin

ucs-m71kr-q-optionrom.2.02.bin

UCS M71KR- Q Converged Network Adapter firmware

 

ucs-m81kr-vic.1.2.1d.bin

UCS M81KR Virtual Interface Card firmware

 

ucs-manager-k9.1.2.1d.bin

UCS Manager software

Runs on the fabric interconnect


Hardware and Software Interoperability

For detailed information about storage switch, operating system, adapter, adapter utility, and storage array interoperability, see the Hardware and Software Interoperability Matrix for this release, located at:

http://www.cisco.com/en/US/products/ps10477/prod_technical_reference_list.html


Note VMware ESX and ESXi 3.5 Update 4, and ESX and ESXi 4.0 are not compatible with Intel 56xx processors. 55xx processors are not affected by this limitation. Please refer to the interoperability matrix for this release for OS and other support questions.


Known Limitations and Behaviors

Adapters

A UCS B250 Extended Memory Blade Server can hold two mezzanine cards, but they must be the same type. Mixed configurations are not supported.

To deploy two UCS M81KR Virtual Interface Cards on the UCS B250 Extended Memory Blade Server running ESX 4.0 you will need to upgrade to patch 5 (ESX4.0u1p5) or later release of ESX 4.0.

(CSCtd32884 and CSC71310) The type of adapter in a server impacts the maximum MTU supported. Network MTU above the maximums may cause the packet to be dropped for the following adapters:

· The Cisco UCS CNA M71KR adapter supports a maximum MTU of 9216.

· The Cisco UCS 82598KR-CI adapter supports a maximum MTU of 14000.

Fabric Interconnect

When trying to install ESX on a server blade, Port Security should be disabled on the corresponding vNIC on the UCS Manager service profile. Otherwise, installation will fail as Port Security will secure only the first MAC address and drop traffic that is sourced with a different MAC address.

Recovering the Corrupt BIOS on a Server

Before you attempt to recover a corrupt BIOS, remove all attached or mapped USB storage from a server. If an external USB drive is attached or mapped from vMedia to the server, BIOS recovery fails.

SSD

Seagate SATA disks and Intel or Samsung SATA SSDs are not supported in release 1.2(1) and will not be part of B200 or B250 M1 or M2 Servers using release 1.2(1).

UCS Manager

The display of NIC statistics is actually the average, minimum and maximum of the change since the last collection. If the values are 0, there has been no change since the last collection.

UCSM GUI

HDD metrics (displayed in the GUI Storage Tab) are not displayed after you hot-swap or remove and reinsert the HDD. This is because the UCS Manager reads the HDD metricsat boot time. This does not impact operation of the HDD, the rest of UCS, or UCS Manager functionality. Re-associate a service profile or re-acknowledge a server to cause a boot that will discover the new HDD metrics.

UCSM GUI and Boot Order

The Configured Boot Order tab shows the boot order assigned to the server. The Actual Boot Order tab shows exactly what will boot from the various devices in the configuration.


Note The Actual Boot Order tab always shows "Internal EFI Shell" at the bottom of the boot order list.


Passwords

Choose a strong password. If a password is not strong, UCSM does not allow it to be set as a local user password. UCSM enforces following rules for a strong password:

1. Must be minimum 8 characters.

2. Must contain at least three of the following: lower case letters, upper case letters, digits or special characters.

3. No character can be repeated more than 3 times consecutively.

4. Must not be same as user name or reverse of user name.

5. Must pass password dictionary check.

6. $ and = symbols are not allowed in the password.

Management GUI

When Java Web Start is not configured properly the exception: "BadFieldException" is thrown by Java Web Start when launching the KVM viewer. This is the default behavior of Java Web Start for any application that uses native libraries when the java web start cache is disabled. The workaround is to select control panel -> java -> General tab -> Temporary Internet Files -> Settings -> and then select "Keep temporary files on my computer" -> OK.

After the initial configuration, you can change the management IP port and the related subnet mask in Cisco UCS Manager CLI. You cannot make this change in Cisco UCS Manager GUI.

When several KVM Consoles are launched, the SUN JRE sometimes reports an error and the KVM Console may fail to launch. Launch the KVM Console again to resolve this issue.

When waking up a computer from sleep, UCS Manager GUI will detect an event sequencing error and display the error: "Event Sequencing is skewed" because the JRE doesn't have a sleep detection mechanism. Always shut down UCS Manager GUI before putting your computer to sleep.

KVM may fail to open by the first try on a blade. This bug is caused by JRE 1.6_11, see: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6741349 To resolve this, upgrade to JRE 1.6_12 or higher.

Boot Order

You should set only one bootable Storage device which is either local disk or SAN boot and to do it only from UCSM. To assure reliable SAN boot operation, the local disks must be physically removed from the server or servers. If the local disk and SAN boot are both present in the boot order, then the association will fail.

In the Cisco UCS Manager GUI, if the Reboot on boot Order Change check box is checked for a boot policy, and if CD-ROM or Floppy is the last device in the boot order, then deleting or adding the device does not directly affect the boot order and the server does not reboot.

The use of the terms primary or secondary boot devices while creating a service profile does not imply a boot order. The effective order of boot devices within the same devie class is determined by PCIe bus scan order.

RHEL

When Using RHEL 4.8, the system cannot see a USB drive mapped through virtual media.The kernel needs to be rebuilt with max_luns option set to 255 in modprobe.conf, refer to the article at http://kbase.redhat.com/faq/docs/DOC-3293

SNMP

SNMP shows the fabric interconnect name rather than the system name. There is no workaround.

An SNMP user name can not be the same as a local user name. Select an SNMP user name that does not match any local user name.

TFTP

Downloads may be slow if TFTP is used. If TFTP performance is slow, use SCP or another protocol.

VM Migration

After a VM is powered off or migrated to a different host, the vSphere Client prevents the power on or migration of the VM to the same host and it reports the error: "Currently connected network interface <x> uses Distributed Virtual Switch (uuid: <y>) which is accessed on the host via a switch that has no free ports." This is most likely to occur when the cumulative number of virtual NICs associated with the VMs powered on a host approaches the number of dynamic vNICs configured in the service profile of the host. UCS Manager configures the number of ports allowed for the DVS on an ESX host to match the number of dynamic vNICs configured in the service profile of the server. After a VM is powered off or migrated to a different ESX host, vSphere Server should recompute the "numPortsAvailable" property in the "hostProxySwitch" object. In the problem scenario, the numPortsAvailable remains zero. After migrating a second VM away from the ESX host, it is possible to power on or migrate a VM on the ESX host.

When attempting to migrate a VM powered on a server to a different data-store on the same server, the vSphere Client prevents the migration and it reports the error: "Currently connected network interface <x> uses Distributed Virtual Switch (uuid: <y>) which is accessed on the host via a switch that has no free ports." This happens when the cumulative number of virtual NICs of the VMs powered on an ESX host matches or exceeds the number of dynamic vNICs configured in the service profile of the server. UCS Manager configures the number of ports allowed for the DVS on an ESX host to match the number of dynamic vNICs configured in the service profile on the server. As result of attempting the VM migration, vSphere Server incorrectly detects an increase in the number of DVS ports used by VMs powered on the host. Shut down VMs on the ESX host and retry the migration.

When changing a template type from initial to updating and at the same time modifying a reboot-causing value, the list of dependencies (i.e. the servers that are affected by the change) is not returned by the back-end so there is no reboot warning. To avoid this. change the template type from initial to updating. After submitting this change (press "Apply Changes") perform the other desired changes, and an appropriate warning with the correct list of dependencies will be displayed.

Resolved Caveats

The following caveats are resolved for the 1.2(1d) release:

Cisco UCS Manager

(CSCtf87136) The Events Log is now correctly updated after reaching 10K entries, and the Audit Log is correctly updated after reaching 20K entries.

(CSCtg61163) BIOS upgrade no longer results in a failure when attempting to activate older versions of the BIOS.

Fabric Interconnect

(CSCtf94452) CallHome notifications saying "SW_CRASH svc_sam_portAG crashed with crash type:134 Severity Level:5" are no longer sent roughly once a week.

(CSCtg10216) Fabric Interconnect activation no longer fails with the error Descr="Kickstart boot variable is not set as expected" because one of the internal log file has grown so large it leaves insufficient space for unpacking the fabric interconnect firmware image.

The following caveats are resolved for the 1.2(1b) release:

Red Hat Linux

(CSCsy24735) When the RHEL 5.3 XEN kernel is booted, the "MCFG area at a0000000 is not E820 reserved" message appears as part of the boot messages. This issue is inherent in RHEL and is resolved in RHEL 5.5.

XML API

(CSCtb55756) From the XML API, for properties that are based on an enum type, users are no longer able to send numeric values that don't correspond to an enum constant, these numeric values are now validated.

Cisco UCS Manager

(CSCta04662) When the IO module is activated with "set startup only" selected, on the UCS Manager activation, the flag no longer changes from "pending-next-boot" to "ready" even though though the IO module was not rebooted.

(CSCtc66393) IOM Activation status no longer gets stuck in "rebooting" even though IOM was not rebooted.

(CSCtd90460) If there are any changes in DNS to the LDAP hostname, it will resolve in UCSM.

(CSCte63284) Cloning of a service profile no longer results in an error message, UCSM will now check for and prevent saving a copied service profile to a name that's already in use.

(CSCte99193) When downloading a firmware bundle using FTP, the download no longer gets stuck at 2%. The download will either fail or succeed.

(CSCtf49065) In release 1.1(1), the static splash image that is briefly shown when the UCS Manager is launched shows 1.3 instead of 1.1. This is visible for a few seconds, and then replaced by the image with the correct version. This problem is fixed in release 1.2 and higher. For release 1.1, this is a purely cosmetic issue.

(CSCta26204) The 12-hour periodic log messages reporting maximum temperature sensor readings are no longer mis-routed as critical message containing the string: "Thermal Maxima ..."

(CSCsz01075) A service profile can no longer be assigned to a server that has a service profile already associated to it.

(CSCta24307) In Service Profile Navigator, when Changing the Boot Policy type, clicking the "Reset Values" now correctly resets the page to the initial status.

(CSCsz91123) Output of keyring commands in 'show configuration' is in the expected sequence. Keyring commands do not need to be performed in specific order for a keyring to be successfully created.

(CSCsz83593) There is no longer a 5 minute delay when trying to connect to the subordinate on out of band interface.

(CSCsz88780) CDP packets are now sent by UCS-6120 on virtual interfaces. Hosts that rely on CDP neighbor information will be able to obtain it from the UCS-6120.

(CSCsz91527) You are no longer allowed to accidentally se a trustpoint name for the default keyring.

(CSCta10001) When the http or https port is set to a port that is already in use by other service, the web process restarts as expected.

(CSCta14482) When you create delete or modify the DNS or domain name or time zone configurations, the operation is now listed in the audit log records.

(CSCta19830) User-defined threshold policies that are set at the vNIC or vHBA level are now evaluated for vNIC and port statistics.

(CSCta32500) UCSM no longer needs to power off the server after a UUID is changed

(CSCta49993) You may now qualify a server adapter by selecting a non-virtualized-eth that does not have support protection.

(CSCta57183) When creating fc-profile from the CLI, any available property may be set as part of the first commit.

(CSCta76741) Removing vHBA from the storage boot order in updating-template will update the boot order on the service profile instance that is based on this template.

(CSCtc29323) Duplicate Intel BIOSes were removed from firmware packages.

(CSCtc75379) When logged in as a non-admin user, the "drop" property is present in the eth-classified CLI and the set drop command is present in the Cisco UCS Manager CLI.

UCS Manager GUI

(CSCte05392) The "Cleared" check box in the fault table now functions as expected.

(CSCtc62337) After launching the Cisco UCS Manager GUI and selecting the Firmware Versions section of the SAS Controller, Cisco UCS Manager reports the complete FW section correctly.

(CSCtc77623) The OS on the server no longer ignores the soft shutdown request by Cisco UCS Manager.

(CSCta16847) Changing the QoS class of service value for FCoE traffic no longer disrupts FCoE traffic.

(CSCta30015) when modifying the Serial Over LAN Policy of a Service Profile, the Serial Over LAN Policy Drawer (in the Policies Tab of the Service Profile Navigator) will immediately reflect the new value.

(CSCta43745) Using the UCSM GUI, World Wide Node Names starting with 5... can now be used in a Service Profile.

(CSCta60324) When changing the boot order of a service profile and clicking "save changes", when navigating away, the message "There are uncommitted changes, do you want to lose your changes?" popup no longer comes up.

(CSCtc23224) The "Parsing error exception" dialog no longer pops up when making changes to a vNIC in a service profile.

(CSCtc32492) The vHBA template VSAN and vNET no longer get out of sync with the VSAN definition.

(CSCtc50229) If a new server is inserted into its slot, and a test service profile was previously created that did not have this server associated to it, and then the test profile was updated to include this otherwise unprovisioned server, the server will be associated to the service profile.

(CSCtc71503) The UCS Manager no longer shows a mismatch between the Actual Boot Order vs. the Configured Boot Order.

(CSCtc75362) When the 'Confirm Successful Operations' confirmation messages checkbox for the UCS Manager properties has been disabled, confirmation messages stop.

(CSCte70720) When making changes on the "Specific Boot Policy" page of a Service Profile, problematic reconfigurations are no longer allowed and implemented.

(CSCta30226) A node-wwnn object's config-state is no longer displayed as 'not-applied' even though it is deployed on the server during association.

(CSCta76749) When defining a threshold in a given direction (above-normal or below-normal), if the current update is in the opposite direction, a fault for the opposite direction will not be raised.

(CSCtb83759) When a MAC address moves out of a UCS-6100, flushing of MAC addresses no longer allows duplicate MAC addresses to be assigned.

(CSCtc23170) When a vHBA's WWPN (World Wide Port Name) is defined as h/w derived, the WWNN (World Wide Node Name) will still be derived as specified in the service profile.

(CSCtc29117) When a service profile is bound to a template, previously created vNICs, vHBAs and contained policies are deleted.

(CSCtc41465) When vHBA templates are modified, the changes immediately propagate to the affected vHBA instances bound to the template.

(CSCtc50661) When binding a service profile to a template the boot order settings are no longer set incorrectly if the service profile had local boot policy settings. The service profile will get all its boot order settings from the template.

(CSCtd41935) A service profile will no longer associate if the VLAN-Port instances maximum of 3140 is reached.

(CSCtd86292) Removing servers that have not been previously decommissioned no longer leads to undefined or missing slots if a new server is subsequently inserted.

(CSCtc55259) While changing Cluster IP (VIP), there is no longer a problem reaching the new IP.

BIOS

(CSCsz37847) The system is now able to boot from HDD when the following boot order is set in the service profile: CDROM, FDD, HDD.

(CSCsz47261) Local Disk is no longer present in the boot order even when it is not present in the server profile.

(CSCta10138) If the BIOS scrub policy is enabled for a service profile, the BIOS reset will no longer cause the Association or Disassociation time to increase to as much as 15 minutes.

Fabric Interconnect

(CSCsz03030) The BMC interface to the IO Module ethernet x/1/9 no longer shows in pinning border or server output.

(CSCta02631) SNMPv3 is now supported on UCS 6120.

(CSCta29227) The UCS 6120 management interface IP address is now propagated to an upstream MDS.

(CSCta36817) When multiple SNMP trap hosts are configured, SNMP traps are sent to all hosts.

(CSCta84769) The link between the UCS 6100 fabric interconnect and the UCS 2100 fabric extender no longer fails to come up resulting in a "FEX-fabric sfp invalid" message.

Adapters

(CSCta29118) Broadcast packets behave as expected when the multicast table is full.

(CSCtc58037) Applying a host profile to a host which doesn't have all the vNICs of the host profile will no longer fail with the error that NIC does not exist.

(CSCtd32228) A target will be discovered as expected when configured in an IVR Zone.

VMWare

(CSCtc58014) After a host profile is applied to ports in VMware Center, vCenter will allow the user to remove a pNIC.

Open Caveats

The following caveats are open for Release 1.2(1):

Adapters

Symptom    When ESX4i is installed on a UCS blade with a M81KR Adapter, or on a Fiber Channel LUN, or ESX4i is upgraded from Update1 to Update1 Patch03/Patch04/Patch05, the Upgrade or Install of ESX4i or Cisco VEM module may fail with the following error mesage: "Message - Bootbank preparation failed: /sbin/bootbankstage-install.sh -v prep returned (1) Errno - 17 Description - There was an error setting up ESXi installation destination"

Workaround   None. (CSCtf73779)

Red Hat Linux

Symptom    Loading multiple driver disks during a RHEL 5.x installation fails.

Workaround   Please refer to the article at http://kbase.redhat.com/faq/docs/DOC-17753 (CSCte73015)

Symptom    When using the Cisco UCS M81KR Virtual Interface Card and Large Receive Offload (LRO), you may encounter a panic when traffic forwarding and LRO are enabled with RHEL 5.4 - xen kernel.

Workaround   For steps and more information on this issue, refer to the article http://kbase.redhat.com/faq/docs/DOC-22147 (CSCte53336)

Symptom    Network usage (scp, ftp, web browsing, etc.) causes a kernel panic on the RHEL5.4 Xen kernel (2.6.18-164.el5xen)

Workaround   Please refer to http://kbase.redhat.com/faq/docs/DOC-23816 for the workaround. (CSCte11502)

Symptom    When a vNIC is not in failover mode and a link down event occurs, the network traffic on the blades is disrupted with a system running RHEL 5.3.

Workaround   This is a known issue with the ixgbe driver in RHEL 5.3 and because RHEL 5.4 is the latest release, Red Hat recommends upgrading the systems to the RHEL 5.4. If you cannot upgrade to RHEL 5.4, below are a few suggestions that has been found to work.

1. Restart the network.

service network restart

or

ifdown ethx 
ifup ethx  

2. Run your system with nomsi.

Edit /etc/grub.conf

Add pci=nomsi to the kernel line

Restart the system with this kernel

Note that network performance may be affected since the system is running in legacy mode. (CSCte44548)

Symptom    Pass-through DMA Support selection is not available in the BIOS if it was Disabled prior to the BIOS upgrade. If you had this option Disabled in previous version, the selection is hidden in the updated BIOS.

Workaround   When you are in the BIOS Setup utility, press F9 to Load the BIOS defaults (CSCtb96792)

Local Disk/RAID

Symptom    A mirrored RAID cluster moved from one blade to another blade will not boot correctly and will be reported as a foreign RAID.

Workaround   Array activation is necessary to convert the foreign RAID to a native RAID. Instructions on moving and activating the array are in the troubleshooting guide. (CSCtd59353)

Symptom    After the removal or insertion of one or more local disks, their full discovery fails.

Workaround   Re-acknowledge the server to complete the full discovery. (CSCsy80888)

Symptom    The Disk Fault/Error Codes, Disk Status, Alarms and the failures forwarded by the SAS Controller are not received by Cisco UCS Manager.

Workaround   None. (CSCsy76853)

Symptom    FSM gets stuck in an Error Configuring the Local Disk Controller state due to various underlying conditions. Those can include but are not limited to the following:

The Local Disks not getting discovered correctly or are "available/presence-Equipped" but not in a Ready state.

Failures that can't be correctly communicated to Cisco UCS Manager can get reported as this type of error.

Workaround   Remove and insert all of the local disks from the failing server, then re-acknowledge the server. (CSCta45805)

Symptom    The disk scrub policy does not meet DoD compliance.

Workaround   None. (CSCsy20036)

BIOS

Symptom    Hubs ethat only use USB 1.0 may not properly present an attached USB device to the UCS server.

Workaround   Avoid using USB hubs that are exclusively USB 1.0 capable. Virtually all USB hubs sold today are USB 1.0/2.0 capable. (CSCtb20301)

Symptom    With the B-250 blade server, the displayed ESX and Linux OS HDD Boot Device Order is the reverse of the BIOS HDD Boot Order.

Workaround   Review both the disks (and drive labels as applicable) during installations of ESX and Linux versions and choose the correct disk for installation. (CSCtd90695)

Symptom    When memory mirroring configuration is disabled by removing a DIMM, the BIOS will switch to the Performance mode, and will not log a SEL that mirroring was disabled.

Workaround   Check the status of the memory mirroring in BIOS Setup->Advanced -> Memory Configuration -> Memory RAS and Performance Configuration. (CSCsy54097)

Symptom    When a faulty DIMM is detected in early BIOS POST (e.g. the blade was powered on with a faulty DIMM), two SEL entries will be sent to the BMC. One entry will be logged for each DIMM.

Workaround   Enter BIOS Setup and navigate to Advanced -> Memory Configuration. This menu will help to distinguish a faulty DIMM from its neighbor. (CSCsy97698)

Symptom    When hot plugging or removing USB devices at a BIOS Setup -> Advanced -> USB screen, the Setup Utility may freeze.

Workaround   Reboot the server. (CSCsz41907)

Symptom    When a NIC is not present in the system, or is not part of the boot order, the BIOS produces the prompt "Press F12 to boot from the Network".

Workaround   None. (CSCsz44683)

Symptom    If a single DIMM in a channel is marked faulty during BIOS POST, the entire channel is marked faulty. Even empty DIMM slots can be marked faulty.

Workaround   None. This is a false positive error message, it can be safely ignored. (CSCsz73464)

Symptom    Under certain conditions after clearing the CMOS, Service Profile disassociation might take up to 20 FSM iterations to finish, the virtual media KVM interface might incorrectly indicate it will boot to the EFI Shell, or the BIOS boot-order table might contain only the EFI-shell as a boot option.

Workaround   Enter the following commands:

enter scrub-policy <any name>
         set bios-settings-scrub no <---
         set descr ""
         set disk-scrub no

(CSCtc44331)

Symptom    Installing EFI Native SLES 11 is not supported in this release.

Workaround   Currently, there is no workaround. (CSCsz99666)

Symptom    Disabling USB 2.0 through Advanced->USB could result in inconsistent information on that particular page. Some devices may not show up as expected. This does not results in functional degradation during POST.

Workaround   Either don't disable USB 2.0 (we are not aware of any need to disable it), or ignore the resulting artifacts in the BIOS setup. (CSCta21849)

Symptom    After resetting the CMOS the system date needs to be reset to current.

Workaround   None. (CSCtb12390)

Red Hat Linux

Symptom    Network usage (e.g. SCP, FTP, web browsing, etc.) causes a kernel panic on RHEL5.4 Xen kernel (2.6.18-164.el5xen).

Workaround   Please refer to http://kbase.redhat.com/faq/docs/DOC-23816 for the workaround. (CSCte11502)

Adapters

Symptom    The link between the Switch and the Adapter appears to be flapping continuously. This typically occurs when a large number of VMs (30) are "Powered On" simultaneously from the Virtual Sphere Client GUI.

Workaround   Do not power on a large number of VMs simultaneously. To avoid this entirely, power off all VMs, flap the link and power them on again. (CSCte73499)

Symptom    With a server booting from SAN, a server crash may occur when an FC HBA is reset, or when a WWPN is zoned in or out on a core fabric switch, or a link flap event occurs either between the NPV switch and the host, or between the NPV switch and the core switch, or the core switch and storage. The symptoms of this problem include server hangs during normal operation, whether or not I/Os happen. It may either remain hung, or may panic and reboot. Upon a reboot, a vmcore file may or may not be present in /var/crash/ depending on whether sync to disk succeeded or not. Alternatively, the server may not crash or hang but can encounter a journal commit I/O error and the boot partition (located on a SAN LUN) will become a read-only file system. The server stays up, but nothing can be written to the necessary partition.

Workaround   There is no workaround other than not booting from SAN. If the OS is RHEL, the fix for this issue will be in RHEL 5.5.z, and upgrading to that kernel is recommended when it releases. (CSCtf81596)

Symptom    When a DCBXP peer on a physical interface sends two different unique identifiers in the Protocol data unit in the same session, a DCBXP process error disables the port.

Workaround   Issue the following commands to enable the port:

scope chassis <Chassis Id> 
scope server <Server Id> 
scope adapter <Adapter Id> 
scope ext-eth-if <Id> 
set adminstate reset-connectivity 
commit-buffer 
 
   

Wait for a minute, the port will come up.(CSCsx42435)

KVM

Symptom    Sometimes keyboard strokes pressed during the boot sequence are ignored by the KVM Application. This occurs because the KVM Tab is not currently selected.

Workaround   Before pressing an key in the KVM Tab, always click the mouse inside the KVM Tab itself. This ensures that the Tab has the focus and that the keys pressed by the user are captured and sent to the Server. (CSCte36910)

Symptom    During the Server Power-State Management, the KVM session may be aborted with a message displaying "Network Connection Dropped".

Workaround   Close the KVM session(s) that have been aborted and open a new KVM session. (CSCtc53253)

Cisco UCS Manager

Symptom    Information about the number of CPU cores enabled is unavailable. Along with the other processor information, the UCS manager GUI and CLI don't show this info.

Workaround   None (CSCte12172)

Symptom    A SMASH Command with an incorrect option should result in an"INVALID OPTION" error.

Workaround   No work-around required. The message displayed is "COMMAND SYNTAX ERROR". (CSCsv87256)

Symptom    The PCIe Address for the Cisco UCS M81KR Virtual Interface Card is not seen in the GUI (or CLI). It causes no functional impact.

Workaround   The only work around is to boot some host OS onto the blade and then determine the PCI address and map it to the MAC address (and subsequently to the VNIC). In a 2.6 kernel based Linux for instance, the /sys/class/net/<device> directory has relevant information. (CSCte58483)

Symptom    Modification of trusted CoS policy in Service Profile does not get immediately applied to the server. If you modify the trusted CoS policy of an adapter profile in a service profile that is currently attached to a physical server, a server reboot is needed. Since it is unsafe to automatically reboot an associated server, UCSM currently does not.

Workaround   Manually reboot the server or disassociate and reassociate the server to get the CoS policy to be applied. (CSCte44668)

Symptom    Environment statistics for IO Module 1 are not shown.

Workaround   None. (CSCtd14585)

Symptom    For each Cisco UCS 82598KR-CI 10 Gigabit Ethernet Adapter, 2 interfaces show up in the OS and ethtool reports Link Detected = yes for both of them. This is only seen on Cisco UCS B250 servers.

Workaround   Use the MAC that has the value provisioned in the service profile. (CSCtd14055)

Symptom    One vNIC defined in the service profile boot order results in two BIOS vNICs.

Workaround   Avoid defining two different pxelinux.cfg/<MAC> files that have different boot/install instructions. When booted, both vNICs should execute the same PXE configuration. (CSCsz41107)

Symptom    After a full restore of the primary fabric interconnect, the subordinate fabric interconnect installation may temporarily fail with the following message:

Enter the admin password of the peer switch:
    Connecting to peer switch... unable to connect! Password could be wrong.
    Hit enter to try again or type 'restart' to start setup from beginning...
 
   

Workaround   Retry the fabric interconnect installation. (CSCsz85876)

Symptom    When the time-based UUID generation method is used in the UUIDgen tool, the system reports duplicate UUIDs because UUIDs are validated based on their suffix, whereas the time-stamp method creates UUIDs with unique prefixes, but non-unique suffixes.

Workaround   Use the random method in the UUIDgen tool to ensure that suffixes are unique. (CSCta40790)

Symptom    For a given port profile with existing VIFs, if the "Max-Ports" setting is reduced from the currently configured value to a value less than the "Used-Ports" value reported for that port profile by VMware vCenter, this is a mis-configuration. The new value for "Max-Ports" for that port profile will only be updated in Cisco UCS Manager and its update in VMware Center will fail, causing a inconsistency between Cisco UCS Manager and VMware Center Server.

Workaround   If the need arises to reduce the value of "Max-Ports" of a given port profile, the new value should be at least the value of "Used-Ports" reported by the VMware Center for all the DVSes for that port profile (not lower than maximum of all the "Used-Ports" values). This constraint has to be ensured manually. (CSCte12163)

Symptom    When a service profile containing 2 vNICs and having failover enabled is applied to some adapters, the fail back timeout specified in the adapter policy for the second vNIC has no effect. The fail back timeout specified in the adapter policy and applied to the first vNIC is applied to the whole adapter and effective for both vNICs.

Workaround   Specify the desired timeout in the adapter policy and apply to the first vNIC. (CSCsz68887)

Symptom    Cisco UCS Manager reports incorrect v NICs or VIFs associated with a given Virtual Machine if and only if the Virtual Machine is deployed from VMware Center in Fault-Tolerance (Active-Standby) mode and the Virtual Machine happens to be the standby or FT Virtual Machine.

Workaround   None. (CSCte45010)

Symptom    When firmware is updated, UCSM accepts another firmware update before the completion of the current update. The new request will fail silently and lock out the selected firmware package for the next update.

Workaround   Create a new host firmware package with a different name from the one that gets locked out which points to the same BIOS package. The new package with a different name, using the same BIOS will not be locked out. (CSCtc26149)

Cisco UCS Manager GUI

Symptom    When upgrading from releases prior to 1.1.1, OS-specific default adapter policies will not have the current recommended default values.

Workaround   After an upgrade from a release prior to 1.1.1, we recommend manually changing the adapter policy parameters to the following values:

Eth VMWare->RSS: Disabled
Eth VMWarePassThru->RSS: Enabled
Eth default->RSS: Enabled
 
   
FC (all)->FCP Error Recovery: Disabled
FC (all)->Flogi Retries: 8
FC (all)->Flogi Timeout: 4000
FC (all)->Plogi Timeout: 20000
FC (all)->IO Throttle Count: 16
FC (all)->Max LUNs Per Target: 256 

(CSCte58155)

Symptom    All DIMMs are reported to be in array 1, and the max memory for a blade is reported as 192GB.

Workaround   DIMMs may be physically located in array 1 or array 2. This is a display-only issue, and does not affect functionality. Ignore the incorrect report. (CSCta56527)

Symptom    The UCSM GUI could show all DIMMs to be in array 1, and maximum memory for a blade of 192GB for a B200.

Workaround   This is a display-only issue, and does not affect functionality. DIMMs may be physically located in array 1 or array 2. (CSCta56527)

Symptom    Hardware revision numbers for fabric interconnect components are not populated in the Cisco UCS Manager.

Workaround   Enter the connect nxos command to connect to the native NX-OS CLI, then issue the appropriate show sprom component command and look for H/W Version: field in the command output. (CSCta12005)

Symptom    In the Cisco UCS Manager GUI, you can only select port channels and individual uplink ports as pin targets, but the Cisco UCS Manager CLI allows you to also select port channel member ports as pin targets.

Workaround   Port channel member ports should not be selected as pin targets, even if the Cisco UCS Manager CLI allows you to. (CSCta60495)

Symptom    Even though there are no POST failures in the chassis, Cisco UCS Manager may sometimes display the overall status of the chassis as POST-failure.

Workaround   Check the chassis POST messages for errors, and if there are none, ignore the Cisco UCS Manager status message. (CSCsz01878)

Symptom    Statistics counters cannot be cleared using Cisco UCS Manager CLI.

Workaround   Clear the counters using the Cisco UCS Manager GUI. (CSCsz47512)

Symptom    The assignment of servers to dynamic pools do not automatically occur.

Workaround   Re-acknowledge the server to enable it as a candidate for pool assignment. (CSCta06882)

Symptom    When a cluster configuration is set up such that I/O module 1 is cabled to fabric interconnect B and I/O module 2 is cabled to fabric interconnect A, then the Ethernet devices are given ports 1 and 0. However if the setup is straight, with I/O Module 1 connected to fabric interconnect A and I/O Module 2 to fabric interconnect B, then the devices are assigned ports 0 and 1.

Workaround   Connect IOM1 to fabric-interconnect A, and IOM2 to fabric-interconnect B. (CSCtb35660)

Symptom    When checking for UUID uniqueness, currently only the UUID suffix is checked.

Workaround   Use a single UUID prefix. (CSCtc59481)

Symptom    System takes several minutes to initialize when a large number of virtual interfaces is configured. When a system with a large number of virtual interfaces is rebooted, it can take several minutes to initialize and complete configuration of the virtual interfaces

Workaround   None. (CSCtd59681)

Symptom    In a Cisco UCS instance with five or more chassis, the following sequence of events may cause the system to not be in high availability ready state for a period of up to five minutes:

1. Discover all chassis

2. Wait for HA READY

3. Decommission all chassis

4. Recommission all chassis

Workaround   Wait for the HA READY state. (CSCsz40462)

Symptom    Logon access is denied for user accounts where the password field was left blank during user account creation.

Workaround   When creating a user account, ensure that a secure password for the account is specified. (CSCta21326)

Symptom    When a fabric interconnect moves from standalone mode to cluster mode the virtual IP Address may be overwritten by the out of band IP of the primary fabric-interconnect. This issue happens only if the out of band IP was modified in stand alone mode before changing the mode to cluster.

Workaround   After using the cluster enable option, change the cluster IP again. This should fix the cluster IP. (CSCta73919)

Symptom    When more than 10 Cisco UCS Manager GUI sessions are open at the same time with remote authentication for a long time (typically for a few hours), one of the Cisco UCS Manager GUI sessions may fail to re-authenticate. This causes the session that fails the re-authentication to close. The problem does not happen when local authentication is in use.

Workaround   Re-login to the Cisco UCS Manager GUI session when it closes. (CSCtb05260)

Symptom    If you move from subnet A to subnet B then do a cluster failover and move back to subnet A, a Virtual IP is no longer pingable.

Workaround   Set the VIP to a different VIP by using the following CLI commands to set it back to the original address:

scope system 
set set virtual-ip <new address> 
commit 

(CSCtc55636)

Fabric Interconnect

Symptom    Without pin group configuration, server interfaces are pinned to uplink interfaces of the fabric interconnect dynamically and the pinning is redistributed as uplink interfaces go up or down. In some situations, the distribution of server interfaces across uplink interfaces is not balanced. Potential impact is some uplinks are under utilized.

Workaround   None. (CSCsv92356)

Symptom    At bootup of the fabric interconnect, the following message will be displayed on the console: "The startup-config won't be used until the next reboot."

Workaround   None, just ignore the message. (CSCsx13134)

Symptom    Console login is treating the admin and ADMIN account as the same. Console login name on the fabric interconnect is not case sensitive, so there is no differentiation between e.g. 'admin' and 'ADMIN'.

Workaround   Implement usernames that are not case sensitive. (CSCsy15489)

Symptom    When you set a TFTP core exporter IP address through Cisco UCS Manager, the fabric interconnect accepts the address even if it is malformed, e.g 1.2.3.412. This does not pose any functional impact to core exporter with valid IP addresses.

Workaround   Remove the malformed core exporter IP address. (CSCsz75747)

Symptom    Under high stress on the system with repeated port flapping and a default or native VLAN changing simultaneously, a process may cause the fabric interconnect to reload.

Workaround   None. (CSCta09325)

Symptom    The show cdp neighbor command does not display information for CDP neighbors seen from the management interface, nor does it display the fabric interconnect CDP information corresponding to the management interface.

Workaround   None. (CSCta25287)

Symptom    Under some circumstances, system messages are not shown on the terminal monitor even though logging on terminal monitor is enabled globally and on a particular session.

Workaround   Use some other destination such as a console, remote server, or local log file. If using a terminal monitor as a syslog destination, re-issue the terminal monitor command from the NXOS CLI every time after changing any syslog configuration. (CSCta31689)

Symptom    With FM/DM version 5.0(0.295) and UCS 1.1 release, a security user defined in Cisco UCS Manager does not get displayed on FM/DM.

Workaround   None. Please verify user security from Cisco UCS Manager. (CSCte25876)

UCS Manager CLI

Symptom    The UUID of the VM changes in VMware vCenter. After a VM restarts, the virtual machine node on the VM tab shows multiple instances of the same VM with one online and one offline.

Workaround   After the VM retention period configured in the VM life cycle policy has passed, Cisco UCS Manager deletes the offline instance automatically. (CSCtc86297)

Symptom    If multiple vHBA initiators are configured in the same zone so that they are visible to each other and can login to each other, and storage multipathing is not configured with active and standby paths through the Cisco UCS fabric interconnects (There is only one path through the primary fabric interconnect), and the primary fabric interconnect is rebooted, the vHBA driver can be exposed to a situation in which two initiators can try to login to each other at the same time, and cause a host crash.

Workaround   Do not configure multiple vHBA initiators in the same zone, to ensure that they are not visible to each other. (CSCte36784)

Symptom    ESX Read and Write commands sent on a vHBA keep timing out and being aborted even if the target device is logged into the network. This can happen if the link flaps very fast from up to down to up and memory allocation for the link event fails. The vHBA driver misses the link down event and it does not re-login to the fabric interconnect. The fabric interconnect, however, has no login state for the vHBA and so drops all packets from the vHBA.

Workaround   Disable and re-enable the port on the UCS fabric interconnect that corresponds to the vHBA. (CSCte08092)

TFTP

Symptom    When downloading an image or bundle using tftp, the download task appears to be stuck at 2% or sometimes fails completely. TFTP download can fail or appear stuck if the bundle is too large. It almost always appears to fail with the current bundle which is 450MB in size. If any image is downloaded individually, should succeed.

Workaround   Either download individual images and not the bundle or use a different protocol like SCP, SFTP, or FTP. Sometimes having the TFTP server close to the fabric interconnect (on the same VLAN or subnet) may help. (CSCtf66646)

Caveats from Previous Releases

Release 1.1(1)

The following caveat is open for Release 1.1(1l), most Release 1.1(1j) caveats also apply unless they are listed as resolved for Release 1.1(1l):

Symptom    Pass-through DMA Support selection is not available in the BIOS if it was Disabled prior to the BIOS upgrade. If you had this option Disabled in previous version, the selection is hidden in the updated BIOS.

Workaround   When you are in the BIOS Setup utility, press F9 to Load the BIOS defaults (CSCtb96792)

This section lists the open caveats in release 1.1(1j).

BIOS

Symptom    Low Voltage DIMMs marked as Identity Not Established and are not included in the PID catalog.

Workaround   None. Disregard the message, the DIMMs are fully functional and the error state is a false positive. Closed in 1.1(1m). (CSCtg86520)

Symptom    Sometimes keyboard strokes pressed during the boot sequence are ignored by the KVM Application.

Workaround   Before pressing a key in the KVM Tab, always click the mouse inside the KVM Tab itself. (CSCte36910)

Symptom    With the B-250 blade server, the displayed ESX and Linux OS HDD Boot Device Order is the reverse of the BIOS HDD Boot Order.

Workaround   Review both the disks (and drive labels as applicable) during installations of ESX and Linux versions and choose the correct disk for installation. (CSCtd90695)

Symptom    When memory mirroring configuration is disabled by removing a DIMM, BIOS will switch to the Performance mode, and will not log a SEL that mirroring was disabled.

Workaround   Check the status of the memory mirroring in BIOS Setup->Advanced -> Memory Configuration -> Memory RAS and Performance Configuration. (CSCsy54097)

Symptom    When a faulty DIMM is detected in early BIOS POST (e.g. the blade was powered on with a faulty DIMM), two SEL entries will be sent to the BMC. One entry will be logged for each DIMM.

Workaround   Enter BIOS Setup and navigate to Advanced -> Memory Configuration. This menu will help to distinguish a faulty DIMM from its neighbor. (CSCsy97698)

Symptom    When hot plugging or removing USB devices at a BIOS Setup -> Advanced -> USB screen, the Setup Utility may freeze.

Workaround   Reboot the server. (CSCsz41907)

Symptom    When a NIC is not present in the system, or is not part of the boot order, the BIOS produces the prompt "Press F12 to boot from the Network".

Workaround   None. (CSCsz44683)

Symptom    If a single DIMM in a channel is marked faulty during BIOS POST, the entire channel is marked faulty. Even empty DIMM slots can be marked faulty.

Workaround   None. This is a false positive error message, it can be safely ignored. (CSCsz73464)

Symptom    Installing EFI Native SLES 11 is not supported in this release.

Workaround   Currently, there is no workaround. (CSCsz99666)

Symptom    Disabling USB 2.0 through Advanced->USB could result in inconsistent information on that particular page. Some devices may not show up as expected. This does not results in functional degradation during POST.

Workaround   Either don't disable USB 2.0 (we are not aware of any need to disable it), or ignore the resulting artifacts in the BIOS setup. (CSCta21849)

Symptom    FSM gets stuck in an Error Configuring the Local Disk Controller state due to various underlying conditions. Those can include but are not limited to the following:

The Local Disks not getting discovered correctly or are "available/presence-Equipped" but not in a Ready state.

Failures that can't be correctly communicated to Cisco UCS Manager can get reported as this type of error.

Workaround   Remove and insert all of the local disks from the failing server, then re-acknowledge the server. (CSCta45805)

Symptom    The Disk Fault/Error Codes, Disk Status, Alarms and the failures forwarded by the SAS Controller are not received by Cisco UCS Manager.

Workaround   None. (CSCsy76853)

Symptom    After resetting the CMOS the system date needs to be reset to current.

Workaround   None. (CSCtb12390)

Symptom    After moving a mirrored RAID cluster from one server to another server, it fails to boot and is reported as a foreign RAID.

Workaround   Array activation is necessary to convert the foreign RAID to a native RAID by utilizing the RAID configuration Utility within the BIOS. (CSCtd59353)

Red Hat Linux

Symptom    Network usage (e.g. SCP, FTP, web browsing, etc.) causes a kernel panic on RHEL5.4 Xen kernel (2.6.18-164.el5xen).

Workaround   Please refer to http://kbase.redhat.com/faq/docs/DOC-23816 for the workaround. (CSCte11502)

Symptom    When a vNIC is not in failover mode and a link down event occurs, the network traffic on the blades is disrupted with a system running RHEL 5.3.

Workaround   This is a known issue with the ixgbe driver in RHEL 5.3 and because RHEL 5.4 is the latest release, Redhat recommends upgrading the systems to the RHEL 5.4. If you cannot upgrade to RHEL 5.4, below are a few suggestions that has been found to work.

1. Restart the network.

service network restart

or

ifdown ethx 
ifup ethx  

2. Run your system with nomsi.

Edit /etc/grub.conf

Add pci=nomsi to the kernel line

Restart the system with this kernel

Note that network performance may be affected since the system is running in legacy mode. (CSCte44548)

Symptom    When using the Cisco UCS M81KR Virtual Interface Card and Large Receive Offload (LRO), you may encounter a panic when traffic forwarding and LRO are enabled with RHEL 5.4 - xen kernel.

Workaround   For steps and more information on this issue, refer to the article http://kbase.redhat.com/faq/docs/DOC-22147 (CSCte53336)

Adapters

Symptom    When a DCBXP peer on a physical interface sends two different unique identifiers in the Protocol data unit in the same session, a DCBXP process error disables the port.

Workaround   Issue the following commands to enable the port:

scope chassis <Chassis Id> 
scope server <Server Id> 
scope adapter <Adapter Id> 
scope ext-eth-if <Id> 
set adminstate reset-connectivity 
commit-buffer 
 
   

Wait for a minute, the port will come up.(CSCsx42435)

KVM

Symptom    During the Server Power-State Management, the KVM session may be aborted with a message displaying "Network Connection Dropped".

Workaround   Close the KVM session(s) that have been aborted and open a new KVM session. (CSCtc53253)

Cisco UCS Manager

Symptom    UCSM 1.1.(1j) firmware activation during UCS firmware upgrade will result in a server. reboot. Customers upgrading their systems from the 1.0.1 or 1.0.2 release to the 1.1(1j) release will experience an unexpected server reboot upon UCSM component activation. This is a result of an incorrect setting of the internal VNIC/VHBA property called lifecycle, that makes the system believe that reconfiguration of these objects is required. This reconfiguration triggers a server reboot.

Workaround   No workaround currently exists. Customers can install the 1.1.(1j) release if a server reboot is acceptable for their firmware upgrade scenario and a maintenance window is scheduled. Customers that do not consider a server reboot as an acceptable condition during UCSM FW activation, should install release 1.1(1m). (CSCtf02353)


Note Please note that BIOS and interface card firmware upgrade still require a server reboot. Customers are expected to continue their operations after UCS firmware upgrade with a new BIOS and Interface Card firmware. Therefore the server reboot described in CSCtf02353 is an unexpected condition on UCSM activation, but still fits in the current firmware upgrade completion requirements.



Note CSCtf02353 will not be a problem on newly purchased systems that ship with Release 1.1.(1j).


Symptom    The disk scrub policy does not meet DoD compliance.

Workaround   None. (CSCsy20036)

Symptom    After the removal or insertion of one or more local disks, their full discovery fails.

Workaround   Re-acknowledge the server to complete the full discovery. (CSCsy80888)

Symptom    One vNIC defined in the service profile boot order results in two BIOS vNICs.

Workaround   Avoid defining two different pxelinux.cfg/<MAC> files that have different boot/install instructions. When booted, both vNICs should execute the same PXE configuration. (CSCsz41107)

Symptom    After a full restore of the primary fabric interconnect, the subordinate fabric interconnect installation may temporarily fail with the following message:

Enter the admin password of the peer switch:
    Connecting to peer switch... unable to connect! Password could be wrong.
    Hit enter to try again or type 'restart' to start setup from beginning...
 
   

Workaround   Retry the fabric interconnect installation. (CSCsz85876)

Symptom    When the time-based UUID generation method is used in the UUIDgen tool, the system reports duplicate UUIDs because UUIDs are validated based on their suffix, whereas the time-stamp method creates UUIDs with unique prefixes, but non-unique suffixes.

Workaround   Use the random method in the UUIDgen tool to ensure that suffixes are unique. (CSCta40790)

Symptom    For a given port profile with existing VIFs, if the "Max-Ports" setting is reduced from the currently configured value to a value less than the "Used-Ports" value reported for that port profile by VMware vCenter, this is a mis-configuration. The new value for "Max-Ports" for that port profile will only be updated in Cisco UCS Manager and its update in VMware Center will fail, causing a inconsistency between Cisco UCS Manager and VMware Center Server.

Workaround   If the need arises to reduce the value of "Max-Ports" of a given port profile, the new value should be at least the value of "Used-Ports" reported by the VMware Center for all the DVSes for that port profile (not lower than maximum of all the "Used-Ports" values). This constraint has to be ensured manually. (CSCte12163)

Symptom    When a service profile containing 2 vNICs and having failover enabled is applied to some adapters, the fail back timeout specified in the adapter policy for the second vNIC has no effect. The fail back timeout specified in the adapter policy and applied to the first vNIC is applied to the whole adapter and effective for both vNICs.

Workaround   Specify the desired timeout in the adapter policy and apply to the first vNIC. (CSCsz68887)

Symptom    Cisco UCS Manager reports incorrect v NICs or VIFs associated with a given Virtual Machine if and only if the Virtual Machine is deployed from VMware Center in Fault-Tolerance (Active-Standby) mode and the Virtual Machine happens to be the standby or FT Virtual Machine.

Workaround   None. (CSCte45010)

Cisco UCS Manager GUI

Symptom    The UCSM GUI could show all DIMMs to be in array 1, and maximum memory for a blade of 192GB for a B200.

Workaround   This is a display-only issue, and does not affect functionality. DIMMs may be physically located in array 1 or array 2. (CSCta56527)

Symptom    Hardware revision numbers for fabric interconnect components are not populated in the Cisco UCS Manager.

Workaround   Enter the connect nxos command to connect to the native NX-OS CLI, then issue the appropriate show sprom component command and look for H/W Version: field in the command output. (CSCta12005)

Symptom    In the Cisco UCS Manager GUI, you can only select port channels and individual uplink ports as pin targets, but the Cisco UCS Manager CLI allows you to also select port channel member ports as pin targets.

Workaround   Port channel member ports should not be selected as pin targets, even if the Cisco UCS Manager CLI allows you to. (CSCta60495)

Symptom    Even though there are no POST failures in the chassis, Cisco UCS Manager may sometimes display the overall status of the chassis as POST-failure.

Workaround   Check the chassis POST messages for errors, and if there are none, ignore the Cisco UCS Manager status message. (CSCsz01878)

Symptom    Statistics counters cannot be cleared using Cisco UCS Manager CLI.

Workaround   Clear the counters using the Cisco UCS Manager GUI. (CSCsz47512)

Symptom    The assignment of servers to dynamic pools do not automatically occur.

Workaround   Re-acknowledge the server to enable it as a candidate for pool assignment. (CSCta06882)

Symptom    In Cisco UCS Manager GUI, if the Reboot on boot Order Change checkbox is checked for a boot policy, and if CD-ROM or Floppy is the last device in the boot order, then deleting or adding the device does not directly affect the boot order and the server does not reboot.

Workaround   None. (CSCta54895)

Symptom    When a cluster configuration is set up such that I/O module 1 goes to fabric interconnect B and I/O module 2 goes to fabric interconnect A, then the Ethernet devices are given ports 1 and 0. However if the setup is straight, with I/O Module 1 connected to fabric interconnect A and I/O Module 2 to fabric interconnect B, then the devices are assigned ports 0 and 1.

Workaround   Connect IOM1 to fabric-interconnect A, and IOM2 to fabric-interconnect B. (CSCtb35660)

Symptom    When checking for UUID uniqueness, currently only the UUID suffix is checked.

Workaround   Use a single UUID prefix. (CSCtc59481)

Symptom    System takes several minutes to initialize when a large number of virtual interfaces is configured. When a system with a large number of virtual interfaces is rebooted, it can take several minutes to initialize and complete configuration of the virtual interfaces

Workaround   None. (CSCtd59681)

Symptom    In a Cisco UCS instance with five or more chassis, the following sequence of events may cause the system to not be in high availability ready state for a period of up to five minutes:

1. Discover all chassis

2. Wait for HA READY

3. Decommission all chassis

4. Recommission all chassis

Workaround   Wait for the HA READY state. (CSCsz40462)

Symptom    Logon access is denied for user accounts where the password field was left blank during user account creation.

Workaround   When creating a user account, ensure that a secure password for the account is specified. (CSCta21326)

Symptom    When a fabric interconnect moves from standalone mode to cluster mode the virtual IP Address may be overwritten by the out of band IP of the primary fabric-interconnect. This issue happens only if the out of band IP was modified in stand alone mode before changing the mode to cluster.

Workaround   After using the cluster enable option, change the cluster IP again. This should fix the cluster IP. (CSCta73919)

Symptom    When more than 10 Cisco UCS Manager GUI sessions are open at the same time with remote authentication for a long time (typically for a few hours), one of the Cisco UCS Manager GUI sessions may fail to re-authenticate. This causes the session that fails the re-authentication to close. The problem does not happen when local authentication is in use.

Workaround   Re-login to the Cisco UCS Manager GUI session when it closes. (CSCtb05260)

Fabric Interconnect

Symptom    Without pin group configuration, server interfaces are pinned to uplink interfaces of the fabric interconnect dynamically and the pinning is redistributed as uplink interfaces go up or down. In some situations, the distribution of server interfaces across uplink interfaces is not balanced. Potential impact is some uplinks are under utilized.

Workaround   None. (CSCsv92356)

Symptom    At bootup of the fabric interconnect, the following message will be displayed on the console: "The startup-config won't be used until the next reboot."

Workaround   None, just ignore the message. (CSCsx13134)

Symptom    Console login is treating the admin and ADMIN account as the same. Console login name on the fabric interconnect is not case sensitive, so there is no differentiation between e.g. 'admin' and 'ADMIN'.

Workaround   Implement usernames that are not case sensitive. (CSCsy15489)

Symptom    When you set a TFTP core exporter IP address through Cisco UCS Manager, the fabric interconnect accepts the address even if it is malformed, e.g 1.2.3.412. This does not pose any functional impact to core exporter with valid IP addresses.

Workaround   Remove the malformed core exporter IP address. (CSCsz75747)

Symptom    Under high stress on the system with repeated port flapping and a default or native VLAN changing simultaneously, a process may cause the fabric interconnect to reload.

Workaround   None. (CSCta09325)

Symptom    The show cdp neighbor command does not display information for CDP neighbors seen from the management interface, nor does it display the fabric interconnect CDP information corresponding to the management interface.

Workaround   None. (CSCta25287)

Symptom    Under some circumstances, system messages are not shown on the terminal monitor even though logging on terminal monitor is enabled globally and on a particular session.

Workaround   Use some other destination such as a console, remote server, or local log file. If using a terminal monitor as a systlog destination, re-issue the terminal monitor command from the NXOS CLI every time after changing any syslog configuration. (CSCta31689)

Symptom    With FM/DM version 5.0(0.295) and UCS 1.1 release, a security user defined in Cisco UCS Manager does not get displayed on FM/DM.

Workaround   None. Please verify user security from Cisco UCS Manager. (CSCte25876)

UCS Manager CLI

Symptom    The UUID of the VM changes in VMware vCenter. After a VM restarts, the virtual machine node on the VM tab shows multiple instances of the same VM with one online and one offline.

Workaround   After the VM retention period configured in the VM lifecycle policy has passed, Cisco UCS Manager deletes the offline instance automatically. (CSCtc86297)

Symptom    If multiple vHBA initiators are configured in the same zone so that they are visible to each other and can login to each other, and storage multipathing is not configured with active and standby paths through the Cisco UCS fabric interconnects (There is only one path through the primary fabric interconnect), and the primary fabric interconnect is rebooted, the vHBA driver can be exposed to a situation in which two initiators can try to login to each other at the same time, and cause a host crash.

Workaround   Do not configure multiple vHBA initiators in the same zone, to ensure that they are not visible to each other. (CSCte36784)

Symptom    ESX Read and Write commands sent on a vHBA keep timing out and being aborted even if the target device is logged into the network. This can happen if the link flaps very fast from up to down to up and memory allocation for the link event fails. The vHBA driver misses the link down event and it does not re-login to the fabric interconnect. The fabric interconnect, however, has no login state for the vHBA and so drops all packets from the vHBA.

Workaround   Disable and re-enable the port on the UCS fabric interconnect that corresponds to the vHBA. (CSCte08092)

Release 1.0(2)

The following caveats were opened in UCS software release 1.0(2) and are still unresolved.

Adapters

Symptom    If default adapter policies are used, windows OS can take a long time to boot due to non-ideal VHBA related settings.

Workaround   Create an adapter policy with optimal values and use that in the service profile. (CSCtb99003)

BIOS

Symptom    With various Local Disk Configurations, the LSI SAS Configuration Utility fails to launch while in BIOS.

Workaround   The LSI SAS Controller Utility should not be used and all of the Local Disk Policy and Service Profile operations must be executed using UCSM.(CSCtc21336)

Symptom    Disabling USB 2.0 is disabled through Advanced->USB, results in various artifacts on that particular page. Some devices may not show up as expected. This results in no functional degredation during POST.

Workaround   Either don't disable USB 2.0 (we are currently are not aware of any need to disable it), or ignore the resulting artifacts in the BIOS setup. (CSCta21849)

Symptom    When the memory mirroring configuration is destroyed by removing a DIMM, the BIOS will switch to the Performance mode, and will not log a message that mirroring was disabled.

Workaround   Check the status of the memory mirroring in BIOS Setup->Advanced -> Memory Configuration -> Memory RAS and Performance Configuration. (CSCsy54097)

Symptom    When plugging or removing USB devices at BIOS Setup -> Advanced -> USB, the Setup Utility may hang.

Workaround   Reboot the server. (CSCsz41907 )

Fabric Interconnect

Symptom    When a fabric interconnect moves from standalone mode to cluster mode it is possible that the virtual IP gets overwritten by the out of band IP of the primary fabric-interconnect. This issue happens only if the OOB IP was modified in stand-alone mode before changing the mode to cluster.

Workaround   After using the cluster enable option, change the cluster IP again. This should fix the cluster IP. (CSCta73919)

HTTP

Symptom    HTTPD process crashed, with the following event log:

Process crashed. Core file 1253640662_SAM_ucs-6120-1-A_httpd_log.3114.tar.gz (SAM/Switch Core Dump) detected on fabric interconnect A

Workaround   None. (CSCtc13234)

Pinning

UCS Manager GUI

Symptom    When more than 10 GUI sessions are open at the same time with remote authentication for a long time (typically for few hours), it has been observed that one of the GUI sessions fails to re-authenticate. This causes the session that fails re-authentication to close. Problem does not happen when local authentication is in use.

Workaround   Re-login to the GUI session when it closes. (CSCtb05260)

Symptom    When waking up from sleep, the UCS Manager GUI will detect an event sequencing error and display the error: "Event Sequencing is skewed" because the JRE doesn't have a sleep detection mechanism.

Workaround   Always shut down the UCSM GUI before putting your computer to sleep. (CSCta94641)

UUID

Symptom    When checking for UUID uniqueness, currently only the UUID suffix is checked.

Workaround   Use a single UUID prefix. (CSCtc59481)

Symptom    Downloads may be slow if TFTP is used.

Workaround   If TFTP performance is slow, use SCP or another protocol. (CSCtb45761)

Release 1.0(1)

The following caveats were opened in UCS software release 1.0(1e) and are still present.

AAA

Symptom   Local user passwords can not contain "$" character.

Workaround   Do not include the "$" character in local user passwords. (CSCsz44814)

Symptom   When using a fully qualified domain name (FQDN) of an LDAP provider, the FQDN is not resolved with DNS, and user authentication using the LDAP provider does not work.

Workaround   Use an IP address instead of an FQDN when creating LDAP providers. (CSCta09972)

Symptom   The IPMI user database in not backed up in the UCS Manager when you export a configuration backup.

Workaround   After the configuration import is done, manually configure the IPMI user profile. (CSCta48483)

Adapters

Symptom   With fabric failover being enabled in a service profile associated with a QLogic or Emulex CNA, there is a potential traffic loss in the network-to-host direction if the fabric interconnect is operating in End Host mode. The duration of loss depends on how soon after failover the host sends a packet towards the network.

Workaround   The traffic loss ends as soon as the host sends a packet after the failover. If the host is a silent host, inject traffic from the host towards the network. (CSCsw39341)

Symptom   When a service profile containing two vNICs and having failover enabled is applied to QLogic or Emulex CNAs, the failback timeout specified in the adapter policy for the second vNIC has no effect. The failback timeout specified in the adapter policy and applied to the first vNIC is applied to the whole adapter and is effective for both vNICs.

Workaround   Specify the desired failback timeout in the adapter policy and apply to the first vNIC. (CSCsz68887)

Symptom   When a DCBXP peer on a physical interface sends two different unique identifiers in the protocol data unit in the same session, a DCBXP process error disables the port.

Workaround   Enter the following commands and then wait for a minute for the port to come up:

scope chassis chassis-id
scope server server-id
scope adapter adapter-id
scope ext-eth-if ext-eth-if-id
set adminstate reset-connectivity
commit-buffer

(CSCsx42435)

BIOS

Symptom   When a blade server is powered on with a faulty DIMM, the BIOS POST detects the faulty DIMM and two SEL entries are sent to the BMC. One entry is logged for each DIMM.

Workaround   Enter the BIOS setup and navigate to the Advanced > Memory Configuration menu to distinguish the faulty DIMM from its neighbor. (CSCsy97698)

Symptom   Installing EFI Native SLES 11 is currently not supported in this release.

Workaround   None. (CSCsz99666)

Symptom   One vNIC defined in the UCS Manager service profile boot order results in two BIOS vNICs.

Workaround   Avoid defining two different pxelinux.cfg/<MAC> files that have different boot/install instructions. When booted, both vNICs should execute the same PXE configuration. (CSCsz41107)

Fabric Interconnect

Symptom   Without pin-group configuration, server interfaces are dynamically pinned to fabric interconnect uplink interfaces, and the pinning is redistributed as uplink interfaces go up or down. In some situations, the distribution of server interfaces across uplink interfaces is not even.

Workaround   None. (CSCsv92356)

Symptom   When a fabric interconnect boots, the "The startup-config won't be used until the next reboot" message appears on the console. Fabric interconnect configuration is controlled by the UCS Manager, so this message has no meaning on the fabric interconnect configuration and has no functional impact.

Workaround   None. (CSCsx13134)

Symptom   Console logon user names on the fabric interconnect are not case sensitive. For example, there is no differentiation between admin and ADMIN.

Workaround   Use case insensitive user names. (CSCsy15489)

Symptom   When you set TFTP core exporter IP address through the UCS Manager, the fabric interconnect accepts the address even if it is malformed; for example, 1.2.3.412. This has no functional impact to core exporter with valid IP addresses.

Workaround   Remove the malformed core exporter IP address. (CSCsz75747)

Symptom   When the system is under high stress, with repeated port flapping (ports rapidly going up and down) and default (native) VLAN change, the FWM process may core and cause the fabric interconnect to reload.

Workaround   None. (CSCta09325)

Symptom   The show cdp neighbor CLI command does not display information for CDP neighbors seen from the management interface, nor does it display the fabric interconnect CDP information corresponding to the management interface.

Workaround   None. (CSCta25287)

Symptom   Under some circumstances, syslog messages are not shown on a terminal monitor even though logging on the terminal monitor is enabled globally and on a particular session.

Workaround   Use a different syslog destination, such as console, remote server, or local log file. (CSCta31689)

Faults and Alerts

Symptom   Even though there are no POST failures in the chassis, the UCS Manager sometimes displays the overall status of the chassis as POST-failure.

Workaround   Check the chassis POST messages for errors, and if there are none, ignore the UCS Manager status message. (CSCsz01878)

Symptom   In rare cases the UCS Manager reports the link absence fault between the fabric interconnect server port and the fabric extender during the internal inventory collection. The following is an example of such a fault:

*************************** 
Severity: Cleared 
Code: F0367 
Last Transition Time: 2009-07-15T11:47:49 
ID: 646445 
Status: None 
Description: No link between fabric extender port 2/1/1 and switch A:1/9 
Affected Object: sys/chassis-2/slot-1/fabric/port-1 
Name: Ether Switch Intfio Satellite Connection Absent Cause: Satellite Connection 
Absent 
Type: Connectivity 
Acknowledged: No 
Occurences: 1 
Creation Time: 2009-07-15T11:46:49 
Original Severity: Major 
Previous Severity: Major 
Highest Severity: Major 
******************************* 
 
   

Workaround   Ignore the fault message; it will automatically get cleared after one minute. This will not impact the data path. (CSCta76573)

High Availability

Symptom   On a system with five or more chassis, the following sequence of events causes the system to not be HA ready for up to five minutes:

1. Discover all chassis

2. Wait for HA READY

3. Decommission all chassis

4. Recommission all chassis

Workaround   Wait for HA READY. (CSCsz40462)

Symptom   After a full restore of the primary fabric interconnect, the subordinate fabric interconnect installation may temporarily fail with the following message:

Enter the admin password of the peer switch:
    Connecting to peer switch... unable to connect! Password could be wrong.
    Hit enter to try again or type 'restart' to start setup from beginning...
 
   

Workaround   Retry the fabric interconnect installation. (CSCsz85876)

Inventory

Symptom   Hardware revision numbers for fabric interconnect components are not populated in the UCS Manager.

Workaround   Perform the following steps to determine the revision number for a fabric interconnect component:

1. Enter the connect nxos command to connect to the native NX-OS CLI.

2. Enter the appropriate show sprom component command and look for H/W Version: field in the command output. (CSCta12005)

Symptom   All DIMMs are reported to be in array 1, and maximum memory for a blade is reported as 192 GB.

Workaround   DIMMs may be physically located in array 1 or array 2, and the maximum memory is 96 GB. This is a display-only issue, and does not affect functionality. (CSCta56527)

Pinning

Symptom   In the UCS Manager GUI, you can only select port channels and individual uplink ports as pin targets, but the UCS Manager CLI allows you to also select port channel member ports as pin targets.

Workaround   Port channel member ports should not be selected as pin targets, even if the UCS Manager CLI allows you to. (CSCta60495)

Pools and Policies

Symptom   The assignment of servers to pools in a dynamic pool are not dynamically assigned. To have a server assigned to a pool, it must be re-acknowledged. This should happen automatically based on the server pool policy qualifications as soon as the qualification is associated to a pool.

Workaround   Re-acknowledge the server to enable it as a candidate for pool assignment. (CSCta06882)

Server

Symptom   Local disk removal and insertion is not detected.

Workaround   Select the Re-Acknowledge Server option in UCS Manager GUI to discover the server. (CSCsy80888)

Symptom   The disk scrub policy needs enhancements to meet DOD compliance.

Workaround   None. (CSCsy20036)

SNMP

Symptom   SNMP shows the fabric inteconnect name rather than system name.

Workaround   None. (CSCta22029)

Symptom   An SNMP user name can not be the same as a local user name.

Workaround   Select an SNMP user name that does not match any local user name. (CSCta24034)

SMASH

Symptom   Any SMASH command entered with wrong option should give "INVALID OPTION" error message.

Workaround   None. (CSCsv87256)

UCS Manager CLI

Symptom   Statistics counters cannot be cleared using the UCS Manager CLI.

Workaround   Clear the counters using the UCS Manager GUI. (CSCsz47512)

UCS Manager GUI

Symptom   When several KVM Consoles are launched, the SUN JRE sometimes reports an error and the KVM Console fails to launch.

Workaround   Launch the KVM Console again. (CSCta38463)

Symptom   In the UCS Manager GUI, if the Reboot on boot Order Change checkbox is checked for a boot policy, and if CD-ROM or Floppy is the last device in the boot order, then deleting or adding the device does not directly affect the boot order and the server does not reboot.

Workaround   None. (CSCta54895)

Symptom   Fibre Channel port and server port events do not appear on the Fibre Channel port and server port Events tabs.

Workaround   Look on the Admin Events tab for Fibre Channel port and server port events. (CSCta66375)

UUID

Symptom   When the time-based UUID generation method is used in the uuidgen tool, the system reports duplicate UUIDs because UUIDs are validated based on their suffix, whereas time-stamp method creates UUIDs with unique prefixes, but non-unique suffixes.

Workaround   Use the random method in the uuidgen tool to ensure that suffixes are unique. (CSCta40790)

New Hardware Features in Release 1.2(1)

This release adds support for the:

B200-M2

B250-M2

Intel Xeon 5600 processor s

Low voltage DIMMs

UCS instances in this release scale to up to 10 UCS 5108 chassis.

Related Documentation

For related documentation, refer to this document:

Cisco UCS Documentation Roadmap

Hardware Documents

Cisco UCS Site Preparation Guide

Cisco UCS 5108 Server Chassis Hardware Installation Guide

Cisco UCS 6100 Series Fabric Interconnect Hardware Installation Guide

Regulatory Compliance and Safety Information for Cisco UCS

Software Documents

There were no functional changes to the user interface in this release. Configuration guides for Release 1.1(1) can and should be used.

Cisco UCS Manager CLI Configuration Guide, Release 1.1(1)

Cisco UCS Manager GUI Configuration Guide, Release 1.1(1)

Cisco UCS Manager CLI Command Reference, Release 1.1(1)

Cisco UCS Manager XML API Programmer's Guide

Obtaining Documentation and Submitting a Service Request

For information on obtaining documentation, submitting a service request, and gathering additional information, see the monthly What's New in Cisco Product Documentation, which also lists all new and revised Cisco technical documentation, at:

http://www.cisco.com/en/US/docs/general/whatsnew/whatsnew.html

Subscribe to the What's New in Cisco Product Documentation as an RSS feed and set content to be delivered directly to your desktop using a reader application. The RSS feeds are a free service. Cisco currently supports RSS Version 2.0.


This document is to be used in conjunction with the documents listed in the "Related Documentation" section.