Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition | |||
|
|
Solution Type Problem Resolution Sure Solution 1006476.1 : Sun StorEdge[TM] 6130: Disk reports a status of "Predictive Failure"
PreviouslyPublishedAs 209071 Symptoms A disk within an SE6130 array reports a status of "Predictive Failure", e.g. Sun Storedge[TM] 6130 Configuration Service GUI : Physical Storage -> Disks, Name Tray Role Virtual Disk State Status Capacity Type Firmware t0d03 0 Data Disk 1 Enabled Predictive Failure 68.366 GB Fibre Channel 0407 Sun Storedge[TM] 6130 Configuration Service CLI : # /opt/se6x20/cli/bin/sscs list -a my-se6130 disk t0d03 Tray: 0 Disk: t0d03 Capacity: 68.366 GB Type: Fibre Channel Speed (RPM): 10033 Status: Predictive Failure State: Enabled Role: Data Virtual Disk: 1 Firmware: 0407 Serial number: 3HZ7MTKD00007451GCY2 WWN: 20:00:00:0C:50:EB:EF:49 and Sun Storage Automated Diagnostic Environment (StorADE) reports a critical alarm for the disk : 1 device_error(s) found in logfile (related to 6130 my-se6130a/192.168.128.101) : Sep 14 05:44:46 SUN.54062390100.0428AWF0ND Tray.0.Drive.3: [ID 0x1010] WARNING: Impending drive failure (PFA) detected (1/5d/0):: Resolution A disk reporting a status of "Predictive Failure" should be physically replaced, but a disk will not be failed by the array just as the result of logging a PFA, so the disk could still be in use. Schedule a disk replacement for a convenient time : Use the "Service Advisor" disk drive removal and replacement procedure documented within the Sun StorEdge[TM] 6130 Management Host software to replace the disk. "Service Advisor" is part of StorADE. Provided the version of the Management Host software is v1.3 or higher, then the disk can be manually failed prior to replacement. To manually fail a disk : Sun Storedge[TM] 6130 Configuration Service GUI : Physical Storage -> Disks -> Click the disk name (e.g. t0d03) to see it's Disk Details screen -> "Fail" button Sun Storedge[TM] 6130 Configuration Service CLI : # /opt/se6x20/cli/bin/sscs fail -a my-se6130 disk t0d03 If a disk is manually failed when the array has a free hotspare configured, then (for a RAID 1, RAID 3 or RAID 5 VDISK) a reconstruction will be performed to the hotspare. This reconstruction may be unnecessary. To prevent it, temporarily unassign all hotspare disks configured on the array. Unassign them before manually failing the disk, and then reassign them as hotspares again after the disk has been physically replaced. Additional Information Prior to v1.3 of the Management Host software there was no way to manually fail a disk. The disk will be detected as failed after it is physically removed from the array. To allow this to occur, wait at least 1 minute after removing the disk, before inserting it's replacement. Product Sun StorageTek 6130 Array Internal Comments A copy of the "Service Advisor" disk drive removal and replacement procedure is available internally at : http://pts-storage.west/products/SE6130/ServiceAdvisor/tfdriverr.html RFE are open to implement "copy and replace" or "clone and replace" type functionality on these arrays :
Bug ID: 6255051
Bug ID: 6319945 SE6130, 6130, Predictive, Failure, Disk, PFA, Impending, Drive, 5d, 0x5d, PREDICTION, THRESHOLD, EXCEEDED Previously Published As 82641 Change History Date: 2006-08-01 User Name: 95826 Action: Approved Comment: no change required. republishing. Version: 6 Date: 2006-08-01 User Name: 95826 Date: 2006-08-01 User Name: 75028 Action: Approved Comment: Hi Robert, thanks for approving this document. I've just updated the "internal only" section with a couple more lines about RFE that are open about how the array handles PFAs - please could you approve it again? Attachments This solution has no attachment |
||||||||||||
|