12-28-2013 02:44 PM
I hope I am not introducing a message that has been answered previously but here goes. We have an RX3600 that gets error 55 "Invalid or inaccessible system ID(s)" when swapping out the display board.
I have been doing some reading here and see that a replacement part from HP has no UUID on it and should cause you to get notification at the EFI prompt to update the system information after replacing a board that is retaining such information.
If a part already has a UUID on it must you get into "MFG Mode" and update the system information every time a part is replaced that may have a UUID on it? Is this the only solution or is it possible to clear the NVRAM on the replacement part of this UUID information so that you get prompted for update?
The reason for replacing is due to a cooling issue we are having that seems invalid as when I look at live event it is telling me that FAN 1 is lagging, fan 2 and 3 are failed. Within 30 seconds or so the error clears and live event shows all operating normal. I know that my fans are working (though perhaps not speeding up) as while watching live event and as I get the error I opened the unit to see all fans functioning properly. Previoius to this all fans had been replaced. The temperature via MP shows steady at 25C.
We have plans to replace this board (Has ambient Sensor) and then replace IO Backplane if problem persists since it drives the fans . I am not onsite nor do I have a connection to the server at this time. I do have minimal information from system log after this replacement showing the following:
14 SFW *5 0xC152BC6926020130 FF3F4070000F0300 SYSTEM_FIRMWARE_ERROR 26 Dec 2013 17:36:38
This could be due to the old firmware on the MP/Core I/O previously replaced that we are updating now.
13 SFW 0 *3 0x778002F100E00110 0000000000000000 EFI_SYSID_INVALID 26 Dec 2013 17:35:41
Trying to confirm, I believe this is due to the Display board that we installed and that it already has a UUID on it. After putting original board back we do not get error any longer and brought server back up.
12 SFW 0 2 0x40801CBB00E000F0 0000000000000000 BOOT_SWITCH_INSECURE_MODE 26 Dec 2013 17:35:27
It indicates we are leaving SAL and entering EFI, and as a result we are leaving the trusted area of firmware.
Unfortunately noone ran "sysset" afterward to see what that command was showing and I have been told that all firmware is up to date and at the same revision though I will verify this upon return after Holiday.
Solved! Go to Solution.
12-29-2013 11:53 PM
False fan errors will appear most of the times due to the fan controller. Maybe this part needs to be replaced (I/O backplane).
Regarding the values:
Go into EFI, run
Shell> sysmode service
and compare the values.
If the values are different (not "magic" - all zero) on the new part, you need to call HP support for assistance.
Hope this helps!
There are only 10 types of people in the world -
those who understand binary, and those who don't.
No support by private messages. Please ask the forum!
If you feel this was helpful please click the KUDOS! star in the left column!
12-30-2013 02:28 PM
I have also seen the fan lagging issue be resolved by replacing the midplane assembly. All of the fan cables connect here and perhaps a poor connection design leads to these errors.
01-02-2014 04:31 PM
All of the respnses are greatly appreciated. I have all parts I believe to be necessary in correcting this issue. Once we get downtime scheduled for replacement I will do so. I will be back to respond here to let all of you know what I find.
Thanks to all and I hope that you had a great Christmas and New Year!!
01-10-2014 10:34 AM
A thanks to all for your input. We have replaced the backplane which solved this cooling issue. Previously an FE had replaced the Core I/O board thinking a possible BMC issue and configured for web access. He used the console port to configure but did not try to log in to the LAN connection/web access port afterward and recently I was advised that we cannot connect to it.
I have connected my laptop to it and I am able to ping and telnet to it but still unable to get web access GUI from browser. I have connected to another like server to test my browser connectivity and was succesful. I have used LC command to vailidate configuration. Also used the SA command to verify web access is enabled. I disabled and enabled just in case to no avail.
At this point I am thinking the board that was installed may be bad as far as the LAN port connection for WEB GUI or the web application itself. Will double check and compare config options from my working servers Core I/O_WeB Access and schedule downtime to swap the core I/O board back if config looks good.
Will solve this, many thanks to all of you for your input on teh cooling issue.