HP c-Class Infrastructure come in handys

I’ve been working with HP c-Class infrastructure components (Onboard Administrator and Virtual Connect)  for a few years now and collected some useful commands that helped me out several times while doing new installations or maintenance.

Update all Blade iLO firmware’s from the OA CLI

OA with connection to the Internet:

• Logon to the Onboard Administrator with SSH/Telnet (CLI)
• Issue the command: hponcfg all ftp://iss:tools@hprc.external.hp.com/iLOfw/ilo-update.xml

OA with no connection to the Internet:
(Steps below describe the use of TFTP, but this can also be FTP or HTTP, change the locations accordingly)

• Start a local TFTP server and place two files in the root directory:

ilo-update.xml and the latest iLO2 firmware (.BIN)

(Either get the .BIN from ftp://iss:tools@hprc.external.hp.com/iLOfw or whenever that’s not updated yet download the “Online ROM Flash Component for Windows – HP Integrated Lights-Out 2” and extract the package, this also gets you the latest .BIN file)

• Edit the ilo-update.xml to match your local TFTP’s IP address, path and filename

IMAGE_LOCATION=”ftp://<LOCAL_TFTP_IP>/<FILENAME>.BIN”/>

• Logon to the Onboard Administrator with SSH/Telnet (CLI)
• Issue the command: hponcfg all tftp://<LOCAL_TFTP_IP>/ilo-update.xml

Rebooting a Blade iLO Controller

The procedure of rebooting (actually HP call’s it resetting) an iLO controller could come in handy whenever the web interface isn’t reachable anymore.

• Logon to the Onboard Administrator with SSH/Telnet (CLI)
• Issue the following commands from the Onboard Administrator Command Line (CLI):

connect server X (x corresponds to the blade bay)
cd /map1
reset

Re-insert a Blade (removal from chassis)

While implementing the c-Class infrastructure it oftenly happens that iLO controllers get unreachable or other strange error messages appear for whatever cause that may be. Re-inserting the blade in the bay causes a full reset which in most circumstances solves these problems.

This re-inserting can also be done via the Onboard Administrator CLI.

• Logon to the Onboard Administrator with SSH/Telnet (CLI)
• Issue the command: reset bay X  (x obviously corresponds to the blade bay)

Unreachable Virtual Connect Domain Manager

My experience while implementing Virtual Connect modules is that oftenly the Virtual Connect Manager isn’t reachable while the modules  are working fine. In that case you can restart Virtual Connect Domain Manager process.

• Logon to the first Virtual Connect Module with SSH/Telnet (CLI)
• Issue the command: reset vcm –failover (this could take a few minutes)