Sun Microsystems, Inc.  Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-71-1004870.1
Update Date:2010-07-06
Keywords:

Solution Type  Technical Instruction Sure

Solution  1004870.1 :   Sun Fire[TM] V1280-E15000:Tech Tip:An offlined processor can be involved in a panic in rare cases  


Related Items
  • Sun Fire 3800 Server
  •  
  • Sun Fire 6800 Server
  •  
  • Sun Fire 12K Server
  •  
  • Sun Fire 4800 Server
  •  
  • Sun Fire V1280 Server
  •  
  • Sun Fire 15K Server
  •  
  • Sun Fire 4810 Server
  •  
Related Categories
  • GCS>Sun Microsystems>Servers>Midrange V and Netra Servers
  •  

PreviouslyPublishedAs
206832


Description
An offlined processor can be involved in a panic in rare cases.

Steps to Follow
An offlined processor (as opposed to a blacklisted processor) continues to
run a small amount of code (the idle loop, cross traps etc).  It is
possible for an offlined processor to still trigger L2 SRAM uncorrectable
errors (for example, UCC) in rare cases.
Enhancements in ScApp firmware 5.15.3/SMS 1.4 and Solaris 8[TM] Kernel
Update 24 (KU24) will offline the processor if the Soft Error Rate
Discrimination (SERD) count for correctable (xxC) events is exceeded or if
there is a single uncorrectable (xxU) event.  Solaris indictments are
stored in the FRUID CHS (Component Health Status), and the offlined
processor (including the memory that it controls) will be removed by POST
on the next reboot of the system.
There is a window of opportunity where the potential for a panic exists
from the time when a processor has been offined to when it actually is
deconfigured and removed completely when the offlined processor can take L2
SRAM errors that can subsequently lead to the panic of a system.
Offlined CPUs (as above) do continue to run small amounts of code which
could still suffer errors which lead to the subsequent panic of a system.
Offlining is a guard against fatal UCC events, though it is not risk
free.  It is recommended that the board that contained the offline
processor be removed from the system as soon as practical by rebooting
or scheduling maintenance.
Note: Using Dynamic Reconfiguration is not recommended as this will force
the failing cpu to run additional code which may aggrevate this condition.


Product
Sun Fire V1280 Server
Sun Fire 6800 Server
Sun Fire 4810 Server
Sun Fire 4800 Server
Sun Fire 3800 Server
Sun Fire 15K Server
Sun Fire 12K Server

Internal Comments
Please see bug 4947174, 4884166.

offline, DR, UCC, panic, 3800, 4800, 4810, 6800, 1280, 15K, 12K
Previously Published As
72539

Change History
Date: 2004-05-17
User Name: 25440
Action: Approved
Comment: Republishing as COntract
Version: 0
Date: 2004-05-17
User Name: 25440
Action: Updated
Comment: Pulling back to Draft to reissue as Contract
Version: 0
Date: 2004-05-17
User Name: 25440
Action: Approved
Comment: Republishing as "Public" per request from Omar Del Rio (via Laure Bouvin)
Version: 0
Date: 2004-05-17
User Name: 25440
Action: Accepted
Comment:
Version: 0
Date: 2004-05-14
User Name: 97233
Action: Approved
Comment: Added references to indicate that DR is not a safe option
to remove an offlined processor for the system.
Version: 0
Date: 2004-05-14
User Name: 97233
Action: Updated
Comment: Moving to draft to make some changes.
Version: 0
Date: 2004-03-16
User Name: 13128
Action: Approved
Comment: KE quality checklist applied - Publish
Version: 0
Date: 2004-03-09
User Name: 13128
Action: Accepted
Comment:
Version: 0
Date: 2004-03-08
User Name: 29589
Action: Unlock
Comment:
Version: 0
Date: 2004-03-03
User Name: 29589
Action: Accepted
Comment:
Version: 0
Date: 2004-03-02
User Name: 72607
Action: Approved
Comment: Ok, Doc looks good, changes were made. Sending for final review...
Version: 0
Date: 2004-03-02
User Name: 72607
Action: Accepted
Comment:
Version: 0
Date: 2004-03-02
User Name: 97233
Action: Approved
Comment: Changed the title to reflect that this issue affects all ESP servers. Added SMS 1.4 to the content.
Version: 0
Date: 2004-01-30
User Name: 32657
Action: Rejected
Comment: You've added 12/15k to the product list, but the title doesn't include
that yet. Also, do you want to mention SMS 1.4 (the first release
which supports chs) for the 15k next to the comments re: scapp ?
Version: 0
Date: 2004-01-30
User Name: 97233
Action: Approved
Comment: Add sf12k/sf15k to the Infodoc
Version: 0
Date: 2004-01-30
User Name: 97233
Action: Updated
Comment: Need to add sf12k/sf15k to this infodoc
Version: 0
Date: 2003-11-18
User Name: 43660
Action: Approved
Comment: Added trademarks. Minor grammatical changes.
Version: 0
Date: 2003-11-18
User Name: 72607
Action: Approved
Comment: No comments, doc looks good and useful. Please publish
Version: 0
Date: 2003-11-14
User Name: 97233
Action: Approved
Comment: Pleas review ASAP.
Version: 0
Date: 2003-11-14
User Name: 97233
Action: Created
Comment:
Version: 0
Product_uuid
6a74b2f9-bbd8-4b2c-870d-b6b73d6e224f|Sun Fire V1280 Server
29da7938-0a18-11d6-8a41-9ed1ad6d6779|Sun Fire 6800 Server
29d6f808-0a18-11d6-8aa8-943929fbbdd8|Sun Fire 4810 Server
29d3a694-0a18-11d6-92da-df959df44cdd|Sun Fire 4800 Server
29d05214-0a18-11d6-92b2-a111614865b5|Sun Fire 3800 Server
29e4659c-0a18-11d6-9fa1-e67bbc033df8|Sun Fire 15K Server
077fd4c5-df8f-4320-ad69-7d01603a674d|Sun Fire 12K Server

Attachments
This solution has no attachment
  Copyright © 2011 Sun Microsystems, Inc.  All rights reserved.
 Feedback