This article describes concepts, limitations and sample configurations of link aggregation / NIC Teaming / LACP / Ether-Channel connectivity between ESX and Physical Network Switches, in particular, Cisco and HP.
Archive for the ‘HP’ Category
Sample configuration of EtherChannel / Link aggregation (LACP) with ESX and Cisco/HP switches
Tuesday, August 10th, 2010Probing EDD (edd=off to disable) ….
Saturday, July 24th, 2010Doing a routine firmware upgrade on a ProLiant BL server I came across this error using the HP Firmware CD (v9) on a USB key during the boot process. The system would hang completely and require a reboot.
The solution was to enable the USB controller in the ProLiant BIOS. I have no idea why it was disabled on this particular server but once enabled the firmware update completed without any problems.
EDD stands for Enhanced Disk Drive.
Error upgrading HP Lefthand VSA 8.1 to VSA 8.5
Tuesday, May 25th, 2010If you get the following error when upgrading SAN/iQ on your VSA:
**ERROR**: The OS disk in this node is not large enough for the new software upgrade package.
* You must increase the size of the OS disk before upgrading this node.
* See the Release 8.5 P4000 SAN Solution Upgrade Guide for instructions.
It’s because you haven’t read the upgrade instructions (like me!) – you need to increase the size of the disk files. See here.
If you really don’t want to read the PDF, the relevant bit is below:
To increase the OS disk size on the VSA:
1. Using the CMC, power off the VSA.
2. Open the VI Client and select VSA -> Edit Settings -> Hardware.
3. Select Hard disk 1 (verify that the Virtual Device Node is SCSI (0:0)).
4. Under Disk provisioning, changed the provisioned size to 8 GB.
5. Click OK.
6. Repeat these steps for Hard disk 2 (verify the Virtual Device Node is SCSI (0:1))
7. Using the VI Client, power on the VSA.
8. Find the VSA in the CMC and apply the 8.5 upgrade.
To increase the OS disk size on the Failover Manager:
P4000 SAN/iQ software upgrade user guide 11
1. Using the CMC, power off the Failover Manager.
2. Open the VI Client and select VSA -> Edit Settings -> Hardware.
3. Select Hard disk 1 (verify that the Virtual Device Node is SCSI (0:0)).
4. Under Disk provisioning, changed the provisioned size to 8 GB.
5. Click OK.
6. Repeat these steps for Hard disk 2 (verify the Virtual Device Node is SCSI (0:1))
7. Using the VI Client, power on the Failover Manager.
8. Find the Failover Manager in the CMC and apply the 8.5 upgrade.
c7000 BladeSystem and Wrong Power Input
Friday, May 21st, 2010If ever you find yourself in the position of having the wrong phase power on a HP c7000 BladeSystem installation don’t panic!
We did an install where the customer was adamant they had three-phase power in the computer suite; so we order the following 507016-B21 (3 Phase BLc7000).
On the day of the installation we find that the customer only had a single phase feed to the server rack, and it was 32 Amps instead of 16 Amps. Since the 3 Phase BladeSystem is presented with 2xIEC309 – 5PIN 16A sockets we had no way of powering the unit up.
Fortunately, it is quite straightforward to remove the 3-Phase power module and replace it with a single phase power module (413379-B21). The two minute (circa £100) swap out not only changes the phase but also the presentation; in this case you are left with six c19 Sockets. Problem solved.
We also needed a couple of 252663-B33 power distribution units to convert the under floor 32 Amp commando sockets into a usable format.
P4000 G2 Performance figures
Wednesday, May 5th, 2010vStorage puts a three node cluster through its paces… see here.
HP Lefthand Networks San/IQ 8.5 upgrade.
Wednesday, May 5th, 2010SAN/iQ 8.5 is available for download – here are the details (from HP)
SANIQ version 8.5 allows customers to:
1 Reclaim additional capacity with network RAID 5 and 6 (up to 50% additional capacity utilization)
2 Simplify business continuity with application integrated snaps for scheduled local and remote sites
3 Analyze their existing configuration to ensure adherence to industry best practices
4 Support external FC/iSCSI/SAS via VSA, allowing customers to repurpose 3rd party storage
For more information on how to download and upgrade:
To download our SAN/iQ 8.5 software go to:
To download full SAN/iQ release upgrades go to:
www.itrc.hp.com/service/sum/home.do
Click on “Software Update Manager (SUM)”
Click on “Download Updates”
Registration on ITRC and linking a support agreement to a user ID is required to access full upgrades
**An update to this: SAN/iQ update files (for upgrading 8.1 to 8.5) are available here:
ftp://saniq85:SAN85eap@ftp.usa.hp.com/
vStorage has some great write-ups on some of the new features:
SAN/iQ availability and download
SAN/iQ 8.5 Best Practice Analyzer
Applying Access Control Lists (ACL) to Procurve Switches
Wednesday, February 10th, 2010This little reminder is on how to restrict access to particular VLANs on HP ProCurve Switches. In our office network we have a number of VLANs (Internet, Internal, Malware, Guest, etc.).
Within our Malware VLAN we are allowed access to the Internet only and certainly not to any other VLAN within our corporate network. We have a HP ProCurve 5406zl switch with a relatively recent firmware loaded up on it.
Our default VLAN is 192.168.1.0;
Our Internet VLAN is 192.168.2.0 (only contains the Firewall).
Our Guest VLAN is 192.168.3.0
Our Malware VLAN is 192.168.6.0 (VLAN no. 6)
The commands you need are:
ip access-list extended Malware
deny ip any 192.168.1.0/24
deny ip any 192.168.3.0/24
permit ip any any
vlan 6
ip access-group Malware in
You can access the relevant manual page here.
Processing AVATAR
Tuesday, February 9th, 2010Here is a very informative article on what kind of processing power it takes to make the latest Hollywood blockbuster.
MSA 2312sa G2 — Firmware Trouble
Tuesday, February 9th, 2010Over the past weekend, I was with a customer installing a HP Modular Smart Array (MSA) 2312sa G2 SAN. This was the dual controller model, with two HP StorageWorks BL 3Gb SAS Switches installed in the c7000 HP BladeSystem.
I noticed a potential bug with the latest version of the MSA Storage Controller (SC) Firmware M110R25-01. When I actioned the firmware update through the SMU (Storage Management Utility) the first SC updated without a problem and upon completion it automatically kicked off the firmware update for the second storage controller which also went through the firmware update process.
Once completed, I restarted both management controllers and logged back in to the SMU. Within a couple of minutes it kicked off the firmware update process again… this continued again, and again, and again….. even after removing the storage controllers and re-seating them.
I connected up the console cable to the SC and noticed that running commands like show status and show controllers would result in long delays and return “Cannot obtain data lock. Will continue to wait”.
I also noticed the following errors when running the show enclosure-status command:
“Error: Communications Error with Storage Controller (-2194) – Could not read SAS info, so skipping enclosure 1. ”
and also
“Error: Unable to communicate with Storage Controller. Please retry the command.”
Meanwhile, the SMU continued in its firmware update loop so I wasn’t able to get any logs from the unit itself. In the end, I disabled the automatic partner firmware update option from the console using the following command:
set advanced-settings partner-firmware-upgrade disabled
I also removed one of the storage controllers (top one) and restarted the MSA with only a single controller. I was able to get into the SMU and perform a full firmware update. Once the update completed I went through some console commands with no error messages or delays – things seemed back to normal.
I then removed the second controller and re-inserted the first controller and went through exactly the same process. Interestingly enough, it told me that the firmware was already up-to-date so it appears the the problem was different firmware versions on the storage controlllers – for whatever reason, the automatic partner firmware update just didn’t work and looped itself into a frenzy.
I re-inserted the storage controllers back into their original positions and sure enough everything worked without a single delay or error. So my advice to you is, when setting the IP addresses of the unit also disable partner firmware update and update the firmware manually.
HP’s latest Proliant Firmware CD
Tuesday, February 2nd, 2010Was recently updated to version 8.70 (22 Jan 2010). You can download it from here.