Integrity Servers
1747980 Members
4518 Online
108756 Solutions
New Discussion

Re: rx8640 Cell ADDing failed - nPar limit?

 
stephen peng
Valued Contributor

rx8640 Cell ADDing failed - nPar limit?

Guys,

There was a single Cell A in this rx8640 box, and I tried to add other three Cell B,C,D to it; With two Cell, A & B, configuration, the nPar was ok to boot to OS, but with A&B&C&D, the four Cell nPar MCAed when it was booting to OS:

Swap device table:  (start & size given in 512-byte blocks)
        entry 0 - major is 64, minor is 0x2; start = 0, size = 134086656
Starting the STREAMS daemons-phase 1
Checking root file system.
file system is clean - log replay is not required
Root check done.
Create STCP device files
Calling function e00000000121efa0 for Shutdown State 8 type 0x4


MCA[26]:MCA occurred!!!.
MCA[26]: lid 0x4030000
MCA[26]: iip 0xe0000000017bb130, ipsr 0x1010084a601a, ifa 0xe00000c238000000
MCA[26]: xip 0xe0000000017bb130, xpsr 0x1010084a601a, xfs 0x8000000000000183
MCA[26]:The processor state register value = 0x20000000fff21120
MCA[26]:Rendezvous status information GR11 = 2
MCA[26]:mca_wakeup() procs MCA'ed 31, procs rendezvous'ed 0,
 procs INIT'ed 0, procs not rendezvoused 1.
MCA, IIP:0xe0000000017bb130 IFA:0xe00000c238000000

 

There were different kind of processors in Cell C, with different minor part number, but when I tried to build a nPar with Cell A&B&D, it still MCAed at the same boot stage.

1, there was no difference between building a 4Cell nPar and a 2Cell nPar, and I don't need further certification, right?

2, could you please tell when the kernel was doing when it MCAed? From what I've been in, if there was  CPU compatibility issue among Cells, it would MCAed as soon as the kernel was loaded.

I have no choice butto suspect Hardware, especially Cell problem.

 

Regards

Stephen

5 REPLIES 5
S_Logan
HPE Pro

Re: rx8640 Cell ADDing failed - nPar limit?

Hi Stephen,

 

It could be hardware issue, Please perform the following steps:

  1. Shutdown the system Operating System, if it is booted.
  2. Connect to the system console in normal manner.
  3. At the EFI Boot Manager menu, select EFI Shell .
  4. At the Shell prompt , enter errdump mca like so: Shell> errdump mca


If there is an MCA error dump log, it will dump the text to the terminal session.
There will be many screens of data. End user can simply cut/paste the output to a text file.

Alternatively, set the terminal emulator to capture the data to a file.

 

Please update with the MCA file , it will be helpful for the analysis.

 

HTH,

Surendar

Was the post useful?

Say Thank you by Clicking the WHITE KUDOS! Star.

For HP Guided Troubleshooting; http://www.hp.com/support/hpgt

 

I work for HPE

Accept or Kudo


[Any personal opinions expressed are mine, and not official statements on behalf of Hewlett Packard Enterprise]
stephen peng
Valued Contributor

Re: rx8640 Cell ADDing failed - nPar limit?

Surendar,
I would upload MCA error dump log ASAP.
Besides, it does not need further license for adding CELL to existing nPar, right?

regards
Stephen
S_Logan
HPE Pro

Re: rx8640 Cell ADDing failed - nPar limit?

Hi Stephen,

 

Adding cells in existing npar doesn't need any licensing, but using the CPUs will be limited to the ICAP license.

 

Hope your system has the same. in case of any limitation with that we cant use the CPUs.

 

HTH,

Surendar

Was the post useful?

Say Thank you by Clicking the WHITE KUDOS! Star.

For HP Guided Troubleshooting; http://www.hp.com/support/hpgt

I work for HPE

Accept or Kudo


[Any personal opinions expressed are mine, and not official statements on behalf of Hewlett Packard Enterprise]
stephen peng
Valued Contributor

Re: rx8640 Cell ADDing failed - nPar limit?

Hi Surendar,
How to verify that where was ICAP limitation?
From my understanding, it would not stop the server from booting to OS, even there was ICAP limitation, right?

regards
Stephen
S_Logan
HPE Pro

Re: rx8640 Cell ADDing failed - nPar limit?

Hi Stephen,

 

Yes, you are correct.

 

There could be other Hardware issue or incompatablity issue is stoping the server boot.

 

you can use " icapstatus" and "parstaus" command will help to identify the icap details.

 

HTH,

Surendar

Was the post useful?

Say Thank you by Clicking the WHITE KUDOS! Star.

For HP Guided Troubleshooting; http://www.hp.com/support/hpgt

 

I work for HPE

Accept or Kudo


[Any personal opinions expressed are mine, and not official statements on behalf of Hewlett Packard Enterprise]