![]() | Oracle System Handbook - ISO 7.0 May 2018 Internal/Partner Edition | ||
|
|
![]() |
||||||||||||||||
Solution Type Problem Resolution Sure Solution 1937173.1 : Oracle ZFS Storage Appliance: ISCSI LUNs Allocated to Linux Clients Fail Discovery with "Logical Unit Not Accessible" Errors
In this Document
Created from <SR 3-9586859109> Applies to:Sun ZFS Storage 7320 - Version All Versions and laterSun Storage 7210 Unified Storage System - Version All Versions and later Sun ZFS Storage 7420 - Version All Versions and later Oracle ZFS Storage ZS3-2 - Version All Versions and later Oracle ZFS Storage ZS3-4 - Version All Versions and later 7000 Appliance OS (Fishworks) SymptomsMultiple iSCSI LUNs are created and managed from a Project on an Oracle ZFS Appliance. The LUNs are assigned to a unique target group and initiator group, such that they are exclusively available to a Linux client. When the Linux client is rebooted, enumeration of any LUN after LUN 0 fails. The log files on the client will report the following: ------------------
Attached 34:0:0:0: attached scsi generic sg5 kernel: scsi 34:0:0:1: Done Success Host Byte=DID_OK Driver=Driver_ok CDB Inquiry: 12 00 00 00 2400 Sense Key: Not Ready Add. Sense: Logical Unit Not accessible, Target port in unavailable state 34:0:0:0: Unexpected response from lun 1 while scanning. scan aborted. ------------------ This error indicates
Once LUN 0 is enumerated successfully, the Linux command sg_luns is run against LUN 0 to gather inquiry information about what else is being presented from the appliance. The output below shows that 5 LUNs (0, 1, 2, 3, 4, and 5) were presented from the ZFS appliance to the Linux client. #sg_luns /dev/sdX
Lun list length = 40 Report luns [select_report=0]: 000000000000000 000100000000000 000200000000000 000300000000000 000500000000000 CauseThe problem is with the iSCSI configuration on the ZFS appliance. For unknown reasons, several, previously deleted iSCSI LUNs are still being presented to the client. The LUNs do not appear in output from the CLI, but they do appear in output from the shell. ZFS-NODE1:> shares Linux-Proj> list
LUNs: NAME VOLSIZE GUID lun20 20G 600144F0C0D6CC01000054453D2F0003 lun0 5G 600144F0C0D6CC01000054453C430001 lun40 40G 600144F0C53A5DCE000054453EFF0002 Shell output. # oracle only restrcited-command
LU Name: 600144F0C0D6CC01000054453C430001 LU Name: 600144F0C0D6CC01000054453CFB0002 LU Name: 600144F0C0D6CC01000054453D2F0003 LU Name: 600144F0C53A5DCE000054453EF20001 LU Name: 600144F0C53A5DCE000054453EFF0002
SolutionThe problem is resolved by opening an SR with Oracle Support to manually remove the iSCSI LUNs from the configuration. Once the iSCSI LUNs are removed, the Linux client should be rebooted. "Warning : The following command(s)/instruction(s) may result in changes to appliance settings/configuration, and could
therefore result in serious damage if not fully understood and implemented correctly. These steps must be carried out only by a highly experienced Oracle TSC or Field Support engineer, trained to work at the Solaris Shell on the ZFS Storage Appliance. If you are at all unsure whether to proceed, please stop and consult your manager or supervisor immediately.". From the shell, list out the iSCSI LUNs with stmfadm. You can then identify and delete phantom LUNs. ZFS-NODE1# stmfadm list-lu LU Name: 600144F0C0D6CC01000054453C430001 LU Name: 600144F0C0D6CC01000054453CFB0002 LU Name: 600144F0C0D6CC01000054453D2F0003 LU Name: 600144F0C53A5DCE000054453EF20001 LU Name: 600144F0C53A5DCE000054453EFF0002 ZFS-NODE1# stmfadm delete-lu 600144F0C0D6CC01000054453CFB0002 ZFS-NODE1# stmfadm delete-lu 600144F0C53A5DCE000054453EF20001 ZFS-NODE1# stmfadm list-lu LU Name: 600144F0C0D6CC01000054453C430001 LU Name: 600144F0C0D6CC01000054453D2F0003 LU Name: 600144F0C53A5DCE000054453EFF0002 Attachments This solution has no attachment |
||||||||||||||||
|