Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition | |||
|
|
Solution Type Sun Alert Sure Solution 1019704.1 : Sun SPARC Enterprise M8000 and M9000 Servers With Certain Firmware May Experience Unexpected Platform Outage
PreviouslyPublishedAs 244206 Bug Id <SUNBUG: 6716245> Product Sun SPARC Enterprise M8000 Server Sun SPARC Enterprise M9000 Server Date of Resolved Release 23-Oct-2008 Sun SPARC Enterprise M8000 and M9000 Servers With Certain Firmware May Experience Unexpected Platform Outage 1. Impact Sun SPARC Enterprise M8000 and M9000 Servers with XSCF Control Package (XCP) firmware versions prior to 1072 may experience unexpected platform outage as result of a fan tray failure. 2. Contributing Factors This issue can occur on the following platforms: SPARC Platform
XSCF> version -c xcpIf the "Current" value is less than 1072, the system is vulnerable to this issue. 3. Symptoms All platform domains will be shut down and messages similiar to the following will be captured in the XSCF platform monitor log: May 24 00:56:26 xscf0 Alarm:
/FAN_A#2:SCF:Abnormal FAN rotation speed. Insufficient rotation
May 24 00:56:35 xscf0 last message repeated 2 times May 24 00:57:51 xscf0 monitor_msg: SCF:DomainID 0 state change (shutdown started, detail#2) May 24 00:57:51 xscf0 monitor_msg: SCF:Domain issued power-off request to RCI target (DomainID 0) May 24 00:57:55 xscf0 monitor_msg: SCF:All domains shutdown started May 24 00:58:10 xscf0 monitor_msg: SCF:DomainID 0 state change (Powered off, detail#2) May 24 00:59:22 xscf0 monitor_msg: SCF:System powered off Key items to note are a fan failure and a monitor message indicating "all domains shutdown started." 4. Workaround There is no workaround for this issue. Please see the Resolution section below. 5. Resolution This issue is addressed in the following release: SPARC Platform
http://www.oracle.com/technetwork/server-storage/sun-sparc-enterprise/downloads/index.htmlNote: The changes implemented in XCP 1072 and later shut down the platform based on exceeded temperature thresholds rather than the loss of a fan tray. Internal Comments Please send technical questions to the following email: sunalert-tech-questions@sun.com and CC the following persons: Internal Contributor/Submitter Internal Eng Responsible Engineer Internal Services Knowledge Engineer CR 6716245 - XSCF should shutdown platform by exceeded temperatures rather than fan loss Support Personnel: M8000 and M9000 servers manufactured before October, 2008 were built with structurally inadequate fan retention brackets. ECO @41211 modified and strengthened the fan tray retention bracket design. Manufacturing began a phased-in release of reworked chassis commencing in October 2008. As such, all chassis prior to October 2008 have the original design retention bracket. Chassis manufactured between October 2008 and December 21, 2008 may or may not have the redesigned brackets, and chassis manufactured after December 21, 2008 were built with the upgraded retention bracket. Visual inspection is necessary for final determination. Symptoms: Visual inspection of the fan trays may evidence that the tray is not fully seated into the chassis - A small gap of 1-3 mm may be evident between the fan tray face and retaining bracket.The gap may allow the fan tray to become unseated from the fan tray backplane. Flex by the retaining bracket and insufficient bumper height may not push the fan tray fully into the chassis. Root Cause: The original design of the fan tray retaining bracket was structurally inadequate to assure fan tray fully seating into the chassis. A redesigned retaining bracket prevents flex within the bracket and taller bumper offsets push the fan tray fully into the chassis Corrective Action: Supported Workaround (if available): XCP 1072 includes the fix for CR 6716245. This release of XCP and higher modifies software behavior to only shutdown the platform for over temperatures and not solely by the loss of a fan tray. XCP 1072 and higher mitigates the loss of a fan tray. Customers should upgrade to XCP 1072 or higher on all M8000 and M9000 XSCFU. Final Resolution: At the discretion of the customer, the Field Service team may obtain a fan tray retention arm retrofit kit. The kit will allow the chassis to be upgraded to the redesigned style retaining bracket. Order the necessary kit as follows: M8000: 555-1946 DC1 FANTRAY STRAP FCO KIT M9000-32: 555-1947 DC2 FANTRAY STRAP FCO KIT M9000-64: 555-1947 DC2 FANTRAY STRAP FCO KIT (order two kits) Follow the instructions available in the below linked document to implement the replacement procedure: http://webdocs.central/pas/uploadpa/archive/ PA004-21464.D_01_820-5635-10.pdf Identification of Affected Parts (how to): Older style fan tray retaining brackets may be visually identified by a flat metal strip design. Newer fan tray retaining brackets are U-shaped. A depiction of the new fan tray retaining bracket is on page 3 of the above linked replacement procedure document. Internal Contributor/submitter David.Lafko@sun.com Internal Eng Responsible Engineer David.Southwell@sun.com Internal Services Knowledge Engineer david.mariotto@sun.com Internal Eng Business Unit Group SSG ES (Enterprise Systems) Internal Escalation ID 1-24028944, 1-448931102, 1-456255601, 1-461397004 |