Friday, December 17, 2010

5108 Chassis Slot Sagging Issue Solved by TAC


It has come to my attention that Cisco has finally addressed the known issue of blade slot sagging in the older 5108 Chassis. The issue only effects the earlier 5108 products and would appear when the center support is removed and a full width blade is added to the chassis. If you call Cisco TAC and provide the serial number of your chassis, they will ship out the newer blade slot support spare parts and this should resolve the issue.


Friday, November 5, 2010

VMware ESXi 4.1 Problem - Hangs During Reboot at "Initializing Scheduler"

Help Offered from the Forum:

Cisco is aware of this problem and has been working with VMware to address this issue - as this issue has also been seen on other vendor's servers.
Additionally, it only seems to be affecting platforms running ESXi.

On the Cisco side, this issue is being tracked via CSCtj19224 - ESXi stuck at Initializing Scheduler - (CCO Account Required to view details)

The workaround is to disable legacy USB within the BIOS.

Thank you to Clint Pearson for the lead on this one!


Friday, October 22, 2010

The new B230 M1 blade

Hey all,

I just found a great article from M. Sean McGee on the new Cisco UCS B230 M1 blade. Here is the link to Sean's site and he does a great job of explaining the differences between the Cisco B230, IBM HX5, HP BL620c & the Dell M910. The blade is very dense and gives Cisco the edge in memory, CPU & core density.


Monday, October 11, 2010


 Feel like iSCSI on UCS? Here how easy it is,

1. Creating a QoS System Class on UCS with an MTU of 9000 & apply to the iSCSI vNIC(s) (Service Profile -> vNIC ->)
2. Enable Jumbo Frames (MTU 9000) on all switches between Initiator (UCS) and iSCSI Target (Equalloqic).
3. Enable Jumbo Frames on OS Adapter (Vmkernel port on ESX orNetwork Adapter Advanced Properties on Windows)


Cisco blade solution vs. the HP blade solution

I just found a great article the other day on Cisco UCS performance vs. HP. Cisco C-Class comissioned Principled Technologies to pitch the HP blade solution vs the UCS. As this is a 3rd party, the results are impartial and speak for themselves.


Wednesday, October 6, 2010

Monday, October 4, 2010

When to use updating templates on vNICS or vHBAs

To update or not to update? With vNIC/vHBA templates is always a question.

When creating a vNIC or vHBA template I always temped to choose the "updating template" option. By far there are more good than bads but you as a UCS admin must explore the risks. The main risk is causing an outage without any notice, not good from any prospective. With updating templates you must know that all virtual interfaces bound to this template will be updated immediately with any change made to the template. This can be very powerful with it comes to adding new VLANs, but can be very dangerous when it comes to policy changes.

Take for instance you have a new VLAN that needs to be added to your UCS environment. If the template is updating, all you do is add the new VLAN to the global VLAN list within UCS and then update your interface template VLAN list. Done! The new VLAN is operational immediately with no reboot required. Though most changes do not require a reboot, always plan for it and follow the directions of any dialog box presented to you.

Happy updating! 


UCS Manager v1.4 coming in Octorber

·         UCSM 1.4 release – due by end of October
Balboa (1.4) Tentative Features (as of 20 Sept 2010 – could change by Balboa FCS):
§  Support for the new B230 server
§  Direct connect UCS C-Series and manage with by UCS Service Profiles
§  Simplified adapter management by controlling PCIe form factor of Cisco Virtual Interface Card
§  Advanced chassis and multi-chassis power capping capabilities
§  Decouple future server and adapter upgrades with UCSM Server and adapter packs
Ethernet and Fibre Channel
§  Higher VLAN scalability, from 512 in previous release to 1024 in 1.4(1)
§  Enhanced monitoring with SPAN support from the UCS 6100
§  Private VLAN support within UCS Manager
§  Better FC bandwidth utilization with FC Port Channeling and FC Port Trunking in FC NPV Mode
§  Support for NAS appliances connected direct to UCS 6100 in Ethernet End Host Mode (EHM)
§  Enable direct connection of FC storage arrays (zoning configuration not supported)
§  vNIC Failover for all virtual environments (including Hyper-V) Full with the Cisco VIC (AKA Palo)
Authentication and Security
§  Simpler integration with Microsoft Active Directory by use of AD groups in UCSM role assignment
§  Support for multiple, simultaneous, authentication systems
§  Multiuser KVM security enhancements – ownership, RW, and RO privileges
Service Profiles/Stateless Computing
§  CIMC IP address configured by Service Profiles
§  Migration validation for service profiles
§  Impact analysis of service profile configuration changes
§  Scheduling of service profile configuration changes
Management, Monitoring and APIs
§  Comprehensive SNMP GET support for everything in UCS
§  Better troubleshooting with syslog enhancements
§  Consumption and reporting of port licenses in UCS Manager

NIC teaming made easy with UCS

Want to NIC team within UCS? There is a simple answer, enable vNIC failover. First edit your vNIC or vNIC template (my choice for vNICs).

On the "General" tab check the "Enable Failover"

That's it! Your vNIC will be active/passive and will failover via the Cisco UCS hardware teaming feature with no OS driver required. This makes teaming a snap in Windows, Linux and even VMware.


Manual failover via command line

I have customers all the time wanting to fail over their 6120's from primary to subordinate or vice versa. To do this its pretty easy. Below is the process

Process to failover
1. Login to the primary 6120 via the UCS cluster IP address.
2. Verify the current primary 6120 in the cluster.
3. Enter local management on the primary via “connect local-mgmt” command.
4. Issue the “cluster lead x” command to make the subordinate switch become the primary. Replace “x” with the correct switch letter.
5. Verify that the role has changed on the previously subordinate switch by SSH’ing into it and issuing the “show cluster state” command. The switch show now show up as the primary.

Happy failover - UCS TECH