Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition | |||
|
|
Solution Type FAB (standard) Sure Solution 1000198.1 : Improved firmware for Seagate 10K.6 disk drives will reduce the incidence of unexpected outages due to a spindle motor issue.
PreviouslyPublishedAs 200251 Bug Id <SUNBUG: 5069315>, <SUNBUG: 4968898> Product Affected Products Listed Below Seagate Cheetah 7 FC-AL disk drives experiencing a higher than expected failure rate (see details below). Part
Impact Seagate Cheetah 7 FC-AL disk drive models ST336607FC, ST373307FC, ST3146807FC, and SCSI disk drive models ST336607LC, ST373307LC, ST3146807LC are experiencing a higher than expected failure rate due to premature spindle motor bearing failure. Engineering analysis of returned disk drives has revealed a higher than normal rate of a particular failure mode. While the overall global install base of Seagate Cheetah 10k.6 family of disk drives is meeting or close to meeting overall quality goals, Sun has been advised of certain instances of Cheetah 10k.6 disk drive issues impacting our customers. The following disk drive models are affected: Seagate Model# Description ST336607FC Cheetah 7 146LDD FCAL (36GB) ST373307FC Cheetah 7 146LD FCAL (73GB) ST3146807FC Cheetah 7 146LP FCAL (146GB) ST336607LC Cheetah 7 147LDD SCSI (36GB) ST373307LC Cheetah 7 147LD SCSI (73GB) ST3146807LC Cheetah 7 147LP SCSI (146GB) For all systems, to determine if a system has one of the Seagate Cheetah 7 drives, do one of the following:
For A1000/A3x00 Arrays attached to a host, an alternative would be to run one of the two following commands: A. #/usr/lib/osa/bin/drivutil -i <raid module specifier> Ex: #/usr/lib/osa/bin/drivutil -i c1t0d0 B. From RM6 GUI, RM6 GUI - Configuration - Module - Profile - drives For T3/T3+ arrays, an alternative would be to use Explorer output or T3 commands to identify the part numbers. The "fru list" command can be used. Another possible way to identify these disk drives for most, but not all, systems and arrays would be to use the "iostat" command. #iostat -En | grep -i vendor When this issue occurs, disk drives can fail with little or no warning. In systems protected by RAID configurations, data is protected. Those systems without RAID could suffer data loss. The specific failure mode is spindle motor failure. This failure mode results in a DNR or Drive Offline condition with no leading indicators or previous error messages. The most frequent messages seen are, "Drive not ready" or the drive may produce excessive read, write or media errors. Sample error messages: Mar 03 15:52:03 arr2 ISR1[1]: N: u1d07 SCSI Disk Error Occurred (path = 0x0) Mar 03 15:52:03 arr2 ISR1[1]: N: Sense Key = 0x1, Asc = 0x3, Ascq = 0x0 Mar 03 15:52:08 arr2 ISR1[1]: W: u1d07 SCSI Disk Error Occurred (path = 0x0) Mar 03 15:52:08 arr2 ISR1[1]: W: Sense Key = 0x3, Asc = 0x3, Ascq = 0x0 Mar 03 15:52:08 arr2 ISR1[1]: W: Sense Data Description = Peripheral Device Write Fault Mar 05 17:48:34 arr2 ISR1[1]: W: u1d07 SCSI Disk Error Occurred (path = 0x0) Mar 05 17:48:34 arr2 ISR1[1]: W: Sense Key = 0x3, Asc = 0x31, Ascq = 0x0 Mar 05 17:48:34 arr2 ISR1[1]: W: u1d07 SCSI Disk Error Occurred (path = 0x0) Mar 05 17:48:34 arr2 ISR1[1]: W: Sense Key = 0x4, Asc = 0x19, Ascq = 0x0 When this issue occurs, the following error codes may indicate that a drive has failed due to a failed spindle motor: "Sense Code 02x04". Sample error messages: /sbus@7,0/QLGC,isp@0,10000/sd@1,0 (sd46): Error for Command: writeError Level: Fatal Sense Key: Hardware Error ASC: 0x2 (no seek complete), ASCQ: 0x0, FRU: 0x0 10098107 c1t8d0 /sbus@7,0/QLGC,isp@0,10000/sd@1,0 (sd46): Error for Command: readError Level: Fatal Sense Key: Vendor Unique ASC: 0x80 (), ASCQ: 0x0, FRU: 0xa 10102117 c2t10d0 /sbus@7,0/QLGC,isp@0,10000/sd@a,0 (sd54): Error for Command: readError Level: Fatal Sense Key: Media Error ASC: 0x11 (unrecovered read error), ASCQ: 0x0, FRU: 0x0 10102117 c1t5d0 /sbus@3,0/QLGC,isp@0,10000/sd@0,0 (sd15): Error for Command: writeError Level: Fatal Sense Key: Media Error ASC: 0x3 (peripheral device write fault), ASCQ: 0x0, FRU: 0x0 10088635 c2t13d0 /sbus@6,0/QLGC,isp@1,10000/sd@d,0 (sd42): Error for Command: load/start/stopError Level: Retryable Sense Key: Not Ready ASC: 0x4 (LUN not ready), ASCQ: 0x0, FRU: 0x0 10096449 c2t9d0 5
Symptoms Root Cause Root cause analysis has determined that spindles at the end of the tolerance distribution may result in less oil, which may lead to inadequate lubrication and ultimately cause the drive to fail. The vendor has identified this as a 1% issue, though this may vary at individual sites. At the component level Seagate has implemented a new spindle motor design. This fix was implemented at Sun via ECO WO_27575 and DPCO 434. At the FW level this issue is addressed by new disk drive firmware patches. The new drive firmware monitors drive activity on Seagate Cheetah 10K.6 disk drives, and has an enhanced SMART feature to predict a premature failure of the spindle motor bearings. This will allow for graceful replacement before an outage occurs. An Upon Failure FCO has been released to facilitate hardware replacement for disk drives that experience this spindle motor issue. See FCO A0256-1. Resolution Update disk drive firmware as needed with the following patches: FC-AL Disk Drives T3 36GB/72GB - 109115-17 T3+ 36GB/72GB - 112276-13 (2.1.6) 36GB/72GB - 115180-07 (3.1.4) SE6120, Minnow, Non-T3 36GB - 113671-03 72GB - 114708-06 146GB - 114709-06
SCSI Disk Drives (including Minnow) 36GB - 116369-08 72GB - 116370-08 146GB - 116512-04
Note: (June 2005 update) patches are now released that INCLUDE Minnow. Modification History Major modifications listed below. Date: 01-SEP-2005
Date: 24-JUN-2005
Date: 03-NOV-2004
Date: 14-May-2008
Previously Published As 100664 Internal Comments References listed below Related Information
Internal Contributor/submitter Bob Bihler Internal Services Knowledge Engineer [email protected] (previously [email protected]) Internal Eng Business Unit Group NWS (Storage) Internal Resolution Patches 114708-06, 114709-06, 116369-08, 116370-08, 116512-04, 113671-03, 109115-17, 112276-13, 115180-07 Internal Kasp FAB Legacy ID 100664, I1136-1 (FIN) Internal Sun Alert & FAB Admin Info Critical Category: Significant Change Date: Avoidance: Patch Responsible Manager: null Original Admin Info: null Product_uuid 1fbdf362-2ff1-11d7-9690-eff42e52a30b|Netra st D130 Storage Array (AC) 299bd26e-0a18-11d6-85ae-b00895a52321|Netra ct 400 Server 29a04d1c-0a18-11d6-8e54-e4fcf1214b85|Netra ct 800 Server 29ad9292-0a18-11d6-8ad4-b5e221808747|Netra st D130 Storage Array (DC) 2a5eee34-0a18-11d6-831b-82d1fe840b21|Sun StorageTek D240 Media Tray 2a660f34-0a18-11d6-98ca-dfa8c950698b|Sun StorageTek MultiPack Desktop Array 2a6a1264-0a18-11d6-85f5-ebacf433bde4|Sun StorageTek S1 Array 2a6d7d50-0a18-11d6-8e0b-f0bd33b24928|Sun StorageTek T3 Array 2a714b10-0a18-11d6-86e2-d56b387d4fbf|Sun StorageTek T3+ Array 2a753f90-0a18-11d6-96d0-f5e96014c515|Sun StorageTek D1000 Array 2a792916-0a18-11d6-8d0a-c3d03933af3c|Sun StorageTek A1000 Array 2a8022d4-0a18-11d6-8043-ee5a180fdb7f|Sun StorageTek A3500 Array 2a8acc34-0a18-11d6-87ef-bd9528a0b933|Sun StorageTek A5200 Array 2cd2e7d2-2980-11d7-9c3f-c506fe37b7ef|Sun StorageTek 6120 Array 3db30178-43d7-4d85-8bbe-551c33040f0d|Sun StorageTek 3310 SCSI Array 49f7ad4a-aa28-47c7-935a-b971312469ea|Netra st A1000 Array 4de60cc2-a08e-4610-b8bf-6a1881cb59c6|Sun StorageTek 6320 System 5475c17c-c204-11d7-9de1-080020a9ed93|Sun StorageTek 6020 Array 58553d0e-11f4-11d7-9b05-ad24fcfd42fa|Sun StorageTek 3510 FC Array 63ff680d-0598-11d8-84cb-080020a9ed93|Sun Fire E2900 Server 6a74b2f9-bbd8-4b2c-870d-b6b73d6e224f|Sun Fire V1280 Server b216445d-b1a6-4b97-8860-8f6019982dfa|Sun StorageTek D2 Array b228c446-c104-419b-ae2e-2788d245ab9f|Netra st D1000 Array b648cdf0-efb8-4d4f-93d4-b17c1baf1935|Sun StorageTek A3500 FC Array e41a7084-3dbf-472d-918b-efb50dcbc220|Netra 1280 Server Attachments This solution has no attachment |
||||||||||||
|