HW error detected on CMC not on ILO

[testmado662]

hello,

When I tried to update Lefthand from 12.6 to 12.8, CMC refuse to update until all HW problems have been solved. What I don't understand, why those problems appears in CMC and not in ILO.

HW problem don't looks like real, all Fan with 0 speed and both Power Supply unknown. After P4000 restarted, same problem. We'll try to stop P4000, unplug power cords and wait few seconds in order to unload all HW components.

Does somebody already get that kind of incident, HW problems seen in CMC not in ILO ?

ILO4 : last version

BIOS : P71 07/01/2015

Both P4000 of same frame get aqual version and HW is OK on first node

Thanks

[Rachna-K]

@testmado662 

We need to know the Model of the StoreVirtual Node. Is it G2 or G3? I am providing the following information for the G3 Nodes.

During boot of the StoreVirtual Node, the hp-health daemon is started to communicate with the ILO to retrieve hardware status information including fan and power supply. This status information is passed on to the mcelog daemon who will report any problems. If hp-health takes too much time to collect the information then the status information is not yet available when requested by the mcelog daemon hence a status unknown is displayed.

I do believe you have already tried to perform a Power Cycle on the Node.

We can try the following:
1. Reset iLO
2. Upgrade iLO to the latest version and then try to downgrade it to 2.50. Run the Diagnostic Test in CMC
3. If issue persists then request you to raise a Support Ticket as we might have to look into the daemon services

[testmado662]

Hello,

The product Number is B7E18A and looks like à P4330 G2.

P4330 was recently restarted but I would like now, to unplug PS a few seconds beore to restart node.

I attach snapshot and I think you're right, we don't get communication between P4330 HW and CMC. I've upgraded ILO and new version dont appears in CMC HW Information.

Next Step, we need to power off node but have we ability to verify or launch deamon manualy ?

Thanks

[Rachna-K]

@testmado662 

Thank you for the information.

B7E18A is a StoreVirtual 4330 Node which is G3.

CMC is not updated with the latest iLO version as you said. I would recommend you to perform the a Power Cycle on the Node.

If this does not pass the Diagnostics Test in CMC then we need to change the daemon services order in the backend.

For this a Support Ticket is required. 

[testmado662]

Hello Rachna,

When you spoke about "we need to change the daemon services order", it's a software modification ? Does an update, at last Lefthand version (12.8) will fix the problem ?

Regards

[Rachna-K · Accepted solution]

@testmado662 

Yes correct its a software modification at the backend. This issue is fixed in 12.8

[testmado662]

Hello Rachna,

To update in 12.8 or other version, all HW incident have to be solved, else no update possible.

If reboot don't fix the problem, we'll have to ask HPE to fix problem before to update.

Thanks