Link to home
Start Free TrialLog in
Avatar of noclav
noclav

asked on

Formatting flash on iLO 4

iLO 4 on a Proliant ML 350 G8 server. I upgraded the firmware from 1.4 to 2.61 and now i get an error for the iLO Health stating Controller firmware revision 2.10.00 Embedded media initialization failed due to media write-verify test failure . I read that some suggest formatting the Embedded Flash but I see this message WARNING: Formatting the Embedded Flash erases all data on the iLO partition and cannot be undone. External providers (like BIOS, Intelligent Provisioning, OneView and FLM) will need to be re-configured. Im not sure if this is going to mess the server up.
Avatar of Casey Weaver
Casey Weaver
Flag of United States of America image

Unless you're using OneView (which you'd know if you are in a datacenter), you should not have any issues. No settings for the server itself (like iLO information or configurations) are stored there. Follow the HPE guide for the issue. I fixed about 14 of these last week (G8 and G9 servers). It goes very smoothly, it's just time consuming due to all the reboots.

https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-c04996097
Avatar of noclav
noclav

ASKER

Thanks Casey for the info.

Im doing this remotely. im concerned about Step 4) Schedule downtime; AC power-cycle and repeat the NAND format
Avatar of noclav

ASKER

no one is at the site so i can reboot the server if that is what they are referring to for AC Power Cycle.
I performed all of mine remotely. In the article I linked shows a link explaining how (with iLO 2.55+) to perform the software based AC power reset in iLO. It's a new and excellent feature they've added.
Avatar of noclav

ASKER

when you ran the rawpost auxcycle.json  did it shut down the server or just the iLO card?
There's a chart there in the iLO link, as to the expected responses of the command. If you run it from the server's running OS, the command will tell you that a reboot is necessary. Perform your normal restart from the OS, and the power cycle will happen as the server completes the shutdown of the reboot, then proceed to fully boot back up. If the OS is not running, the server will complete an auxcycle, and then remain in the off state, ready to be booted by power button or iLO virtual power button.

Keep in mind that the purpose of Auxcycle is to perform a ilo/mainboard reset by removing the power and reapplying it, just as you would shut down the server, unplug its power cable(s), wait, and then reconnect the power cable(s) and boot the server. This action affects the server and iLO card, and doing this NAND flash fix does require downtime, as the server needs a few reboots and the auxcycle for this to be completed.
Avatar of noclav

ASKER

im sorry for the noob questions but i never done this before and dont want to mess it up. i ran that command from another server and this is what i got
iLOrest > rawpost auxcycle.json
The Operation completed successfully.

The server with the issue is on while i ran that command.
Do i now just reboot the server?
Avatar of noclav

ASKER

i have shut the server down and restarted it from iLO and same error. Can i ignore this error? I dont see any issues.
What did the iLO logs show occured during reboot? You should run the NAND flash after doing the auxcycle reboot. You'll most likely need to reboot twice after the NAND flash. The first time will clear the REST API. The second reboot will restore the REST API settings to the NAND.
Avatar of noclav

ASKER

Here are the steps im doing,
1. Press Button Format Embedded Flash and reset iLO
2. After i relogin to the iLO web page and see the same error. I run the auxcycle.json from a remote server.
3. from iLO web interface i launch the remote Console and shut down the server.
4. Have a cigarrette or two.
5. from iLO web interface i power up the server.
6. Press Button Format Embedded Flash and reset iLO (second time)
same error. Controller firmware revision 2.10.00 Embedded media initialization failed due to media write-verify test failure

 
here are the logs

11/17/2018 14:53 11/17/2018 14:53 1 Browser login: admin - 192.168.6.10(DNS name not found).
1109  
11/17/2018 14:52 11/17/2018 14:52 1 iLO reset by user diagnostics.
1108  
11/17/2018 14:49 11/17/2018 14:49 1 On-board clock set; was 01/01/1970  00:06:56.
1107  
[NOT SET]  [NOT SET]  1 Server reset.
1106  
[NOT SET]  [NOT SET]  1 Power-On signal sent to host server by: admin.
1105  
[NOT SET]  [NOT SET]  1 APO: Settings modified by: admin
1104  
[NOT SET]  [NOT SET]  1 REST API Warning: REST API memory cleared (SCHEMA_PROV_ID master storage file set to default. Reboot system.).
1103  
[NOT SET]  [NOT SET]  1 APO: Restored power state to standby.
1102  
[NOT SET]  [NOT SET]  1 System power was cycled for update to take effect.
I see that REST API was cleared, but not restored. It needs another normal reboot.
Avatar of noclav

ASKER

1119  
11/17/2018 15:05 11/17/2018 15:05 1 Embedded Flash/SD-CARD: Failed restart..
1118  
11/17/2018 15:05 11/17/2018 15:05 1 Server power restored.
1117  
11/17/2018 15:05 11/17/2018 15:05 1 Server reset.


i dont see a restore
It's interesting that iLO is not resetting REST. While there's not necessarily anything going wrong, I would open a case with HPE as to restore the server to normal it may need a system board replacement. The NAND can simply be bad and not flashable. After a few reboots I've always seen it restore.
This question needs an answer!
Become an EE member today
7 DAY FREE TRIAL
Members can start a 7-Day Free trial then enjoy unlimited access to the platform.
View membership options
or
Learn why we charge membership fees
We get it - no one likes a content blocker. Take one extra minute and find out why we block content.