Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition | |||
|
|
Solution Type Problem Resolution Sure Solution 1009358.1 : Pending "disabled" status being reported by "showcomponent" command
PreviouslyPublishedAs 212945 Symptoms When using 5.19.x firmware, system may start to report current "enabled" with pending "disabled" status for some of its components when relevant domain is up: Component Status Pending POST Description --------- ------ ------- ---- ----------- /N0/SB4/P0 enabled - pass UltraSPARC-III+, 1200MHz, 8M ECache /N0/SB4/P1 enabled disabled pass UltraSPARC-III+, 1200MHz, 8M ECache /N0/SB4/P2 enabled - pass UltraSPARC-III+, 1200MHz, 8M ECache /N0/SB4/P3 enabled - pass UltraSPARC-III+, 1200MHz, 8M ECache /N0/SB4/P0/B0/L0 enabled - pass 512M DRAM /N0/SB4/P0/B0/L2 enabled - pass 512M DRAM /N0/SB4/P0/B1/L1 enabled - untest empty /N0/SB4/P0/B1/L3 enabled - untest empty /N0/SB4/P1/B0/L0 enabled disabled pass 512M DRAM /N0/SB4/P1/B0/L2 enabled disabled pass 512M DRAM /N0/SB4/P1/B1/L1 enabled disabled untest empty /N0/SB4/P1/B1/L3 enabled disabled untest empty Resolution Pending "disabled" status means that corresponding component's CHS records are updated to "faulty" value and this component will be excluded from configuration during next POST cycle. Starting from 5.19.0 firmware release, AD (Auto-Diagnosis) mechanism has been extended to cover following types of system faults: - CPU errors (IERR, ISAP, PERR, THCE, TSCE, IPE, DPE); When AD engine registers fault condition for some component, it updates CHS for this component to prevent having it used after next POST. For example, showcomponent output presented above might be preceeded by following messages in domain logs: Oct 09 16:19:41 sf68a-sc0 Domain-A.SC: [ID 893798 local1.warning] [VCM] Event: SF6800.VCMON.1.09.1438 CSN: 0321MM2466 DomainID: A ADInfo: 1.VCMON.19.2 Time: Sun Oct 09 16:19:30 MSK 2005 FRU-List-Count: 1; FRU-PN: 5016178; FRU-SN: A14617; FRU-LOC: /N0/SB4/P1 Recommended-Action: Service action required This is an example of VCMON flagging event resulting in SB4/P1 removal scheduled on next POST run. Each time pending "disabled" status is seen, system's health needs to be examined through logging service call as soon as possible. Relief/Workaround In order to provide a TEMPORARY workaround of re-enabling these CHS-disabled components, service personnel will need to apply for a service mode password and have it ready for the moment when system is scheduled to be rebooted. The service-mode password will be required to enable back CHS-disabled components prior to the system reboot. It needs to be understood clearly that this TEMPORARY workaround of re-enabling components which are previously diagnosed as FAULTY and marked FAULTY in their CHS and/or downgrading system firmware results in a VERY HIGH potential risk of system outages. It is recommended that any re-enabling of a Pending disabled component be done only as temporary workaround before permanent solution of particular system fault is implemented. Utilizing this TEMPORARY workaround should occur only after all platform and domain logs are collected and only with the agreement of a solution center engineer handling the service call noted in the service call's case documentation. Additional Information Sun Blueprint[TM]: AUTO DIAGNOSIS AND RECOVERY ENHANCEMENTS FOR SUN FIRE[TM] MIDRANGE SERVERS UPDATED FOR FIRMWARE RELEASE 5.19.0 http://www.sun.com/blueprints/0805/819-3729.pdf Product Sun Fire 4800 Server Sun Fire V1280 Server Sun Fire E6900 Server Sun Fire E4900 Server Sun Fire E2900 Server Sun Fire 6800 Server Sun Fire 3800 Server Internal Comments Audited/updated 11/24/09 - [email protected], Mid-Range Server Content Team serengeti, chs, pending, disabled, scapp, showcomponent Previously Published As 83138 Change History Date: 2006-01-24 User Name: 71396 Action: Approved Comment: Performed final review of article. No changes required. Attachments This solution has no attachment |
||||||||||||
|