Capturing Image from BL680c G5 Server

[Rigoberto Corujo]

Hello Mario,

On May 2nd, I believe you mentioned if failed on step 2 with a "Could not determine BMC" error, but your most recent update says it failed on step 1. I just want to confirm whether you're seeing a different error now.

I also ran into the "Could not determine BMC" error today while doing an image capture. In my case, my managed node did not have the "hp-ilo" RPM installed, so the "hp-ilo" driver wasn't loaded as could be seen by "lsmod | grep ilo".

I reinstalled the PSP with "install800.sh" but noticed that there was an error installing "hp-ilo", which is mentioned in the Release Notes. After installing "hp-ilo" manually and running "service hpasm reconfigure", I ran "Options / Discovery ..." and, afterwards, I was able to do an image capture. Basically, it came down to the iLO driver not being loaded at the time of discovery.

Looking through your previous postings, it seems that you've already checked this, but I would like to confirm that the discovery was done after the ilo driver was loaded.

Rigoberto

[Mario Couthino]

Sorry my mistake(fat fingers). the error is still at step 2.

My HPILO driver is installed
[root@msgtap1 linux]# lsmod | grep ilo
hp_ilo 44156 5

I ran the reinstall again and found no errors
I then ran service hpasm reconfigure and picked all the default prompts with the exception of

This configuration script will configure SNMP to integrate with the HP SIM and
the HP System Management Homepage. The HPASM can also exist in a more secure
SNMP environment (e.g. VACM) that you have previously configured. See the
hpasm(4) man page for specific details on how to configure the VACM entries
in the 'snmpd.conf' file. You may press now to exit now if needed.

Do you wish to use an existing snmpd.conf (y/n) (Blank is n): y



I dont know what happened but after running all the steps it seems to be working.

thanks to all

[Mario Couthino]

I think I spoke to soon.

when running the capture of the image I am now failing on Step Waiting for RAM disk response. My target server is on the # prompt my_modules.

I am also attaching the screen shots of the steps.

[Rigoberto Corujo]

Hello Mario,

Can you confirm that at the "Reset Server" step the target node actually rebooted and did a PXE boot?

In a terminal window, can you run "console " to see the console output from the target node.

Re-run the deploy and verify via the console output whether the node is rebooted and a PXE boot is being done.

Thank you.

Rigoberto

[Mario Couthino]

Hello Mario,

The server did get reset and rebooted. it stopped at waiting for Ram disk response.
On the target box I am at the # my_modules prompt and when I run console (msgtap1) I get console not found.

[Mitchell Kulberg]

Hi Mario,

The # prompt inside the RAMdisk is normal.
You can enter commands while caputres and deploys are running.

If you are stuck at "waiting for ramdisk" my first thought is perhaps a firewall issue. Perhaps the firewall is blocking a port we need. You might want to double check the list of firewall ports you have opened with the list in the getting started guide. In particular, verify port 60000, or whatever port you asked your repository web server to run on. Check all the others too. A simple way to test would be to turn off the firewall "service iptables stop" and try the capture again.

Also verify your DHCP configuration, that you have the option-140 defined properly.

Let me know what happens.

Mitch

[Mario Couthino]

This is the real thing. I was successfully able to capture an exisiting image from my BL server.

Thank you to all

Regards

[Mario couthino_1]

I was successfully able to capture an exisiting image from my BL server.