Certain Disk Drives Installed in the Sun StorEdge 3510 FC Array may Stop Responding to Requests |
|
| Category : | AvailabilityData Loss |
| Release Phase : | Resolved |
| Product : | Sun StorageTek 3510 FC Array
|
| Bug Id : | 6319828
|
| Date of Workaround Release : | 19-OCT-2005
|
| Date of Resolved Release : | 02-NOV-2005
|
Impact
Certain internal disk drives (listed below) installed in Sun StorEdge 3510 FC Arrays with firmware version 4.11I or higher may stop responding to requests during a controller failover event which results in an initialization of the drive loop.
Firmware version 4.11I and higher has a data integrity enhancement feature called Media Scan. Initialization of the drive loop during Media Scan under some corner case conditions may cause the drive to hang and the controller to mark the drive as "BAD" or offline it.
Logical devices that contain one "BAD/USED" device will become degraded. A logical device with more than one "BAD/USED" device will be flagged as "failed" and data may be lost.
Contributing Factors
This issue can occur on the following platforms:
- Sun StorEdge 3510 FC Array with firmware version 4.11I or later
with one of the following disk drives installed:
- MAP3147FC (146.8GB-10000 RPM FC-AL) without patch 116815-07
- MAP3735FC (73.4GB-10000 RPM FC-AL) without patch 116514-09
- MAS3735FC (73.4GB-15000 RPM FC-AL) without patch 116817-04
- MAS3367FC (36.4GB-15000 RPM FC-AL) without patch 116816-04
Note: The described issue can occur on a Sun StorEdge 3510 FC Array running 3.27R firmware, but the risk is low at this firmware version.
To determine if a Sun StorEdge 3510 FC Array contains one of the drives mentioned above, use the sccli command line as shown:
sccli> show disks
Ch Id Size Speed LD Status IDs
Rev
---------------------------------------------------------------------------
2(3) 6 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N UR00P2900158
WWNN 500000E01015D420
2(3) 7 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N UR00P2900147
WWNN 500000E01015D0B0
2(3) 8 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N 000348R00ES6
WWNN 500000E010677390
2(3) 9 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N UR00P29001B5
WWNN 500000E010165850
2(3) 10 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N UR00P29000SM
WWNN 500000E010154A40
2(3) 11 136.48GB 200MB NONE NEW <Vendor> MAP3147F SUN146G
1601
S/N UR00P29000P1
WWNN 500000E010154100
To determine the Sun StorEdge 3510 FC Array firmware revision, the following command can be used:
sccli> show rev
Vendor: SUN
Product: StorEdge 3510
Revision: 411I
Peripheral Device Type: 0x0
Symptoms
If the described issue occurs, one or more of the affected drives will show as "BAD" when checking "sccli show disks" output after an initialization on the fiber channel loop.
Example:
sccli show disks
Ch Id Size Speed LD Status IDs Rev
----------------------------------------------------------------
2 0 68.37GB 200MB ld0 ONLINE SUN S/N 00050680xxxx
2 1 68.37GB 200MB ld0 ONLINE SUN S/N 00050680xxxx
2 2 68.37GB 200MB NONE USED SUN S/N 00050680xxxx
2 3 68.37GB 200MB ld0 ONLINE SUN S/N 00050680xxxx
Workaround
Drives experiencing this issue can be re-used by re-seating the drive, which will then come up as a "USED" state. A manual rebuild would be needed to make the drive active again on the logical drive. If there is a spare drive configured at the time of failure, the spare drive would have taken over for the failed drive.
Resolution
This issue is addressed in the following disk drives:
- MAP3147FC (146.8GB-10000 RPM FC-AL) with patch 116815-07 or later
- MAP3735FC (73.4GB-10000 RPM FC-AL) with patch 116514-09 or later
- MAS3735FC (73.4GB-15000 RPM FC-AL) with patch 116817-04 or later
- MAS3367FC (36.4GB-15000 RPM FC-AL) with patch 116816-04 or later
Modification HistoryDate: 02-NOV-2005
- State: Resolved
- Updated Contributing Factors, Relief/Workaround, and Resolution sections
AttachmentsThis solution has no attachment