Hello, Description of problem: We use detect_multipath to get the information about Disk Device to build multipathing. Now we have a new Disksubsystem "Hitachi USP 100" and the output with detect_multipath is not ok. With the old Disksubsystem "Hitachi 9980V", the output is ok. Our automatic staging for RHEL3, special building the multipathing is depend on output from detect_multipath. Version-Release number of selected component (if applicable): raidtools 1.00.3-7 and 1.00.3-8 How reproducible: always Steps to Reproduce: 1.With Disksubsystem "Hitachi USP 100" Actual results: - With "Hitachi USP 100": ------------------------- [root@server1 rpm]# detect_multipath /dev/sda, /dev/sdb, /dev/sdc, and /dev/sdd are multiple paths to the same device /dev/md0 = /dev/sda1, /dev/sdb1, /dev/sdc1, /dev/sdd1 /dev/md1 = /dev/sda3, /dev/sdb3, /dev/sdc3, /dev/sdd3 /dev/md2 = /dev/sda4, /dev/sdb4, /dev/sdc4, /dev/sdd4 [root@server1 rpm]# cat /proc/scsi/scsi Attached devices: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: HITACHI Model: OPEN-V Rev: 5004 Type: Direct-Access ANSI SCSI revision: 03 Host: scsi0 Channel: 00 Id: 00 Lun: 01 Vendor: HITACHI Model: OPEN-V*2 Rev: 5004 Type: Direct-Access ANSI SCSI revision: 03 Host: scsi1 Channel: 00 Id: 00 Lun: 00 Vendor: HITACHI Model: OPEN-V Rev: 5004 Type: Direct-Access ANSI SCSI revision: 03 Host: scsi1 Channel: 00 Id: 00 Lun: 01 Vendor: HITACHI Model: OPEN-V*2 Rev: 5004 Type: Direct-Access ANSI SCSI revision: 03 Expected results: - With "Hitachi 9980V": ----------------------- [root@server2 root]# detect_multipath /dev/sda and /dev/sdc are multiple paths to the same device /dev/md0 = /dev/sda1, /dev/sdc1 /dev/md1 = /dev/sda3, /dev/sdc3 /dev/md2 = /dev/sda4, /dev/sdc4 /dev/sdb and /dev/sdd are multiple paths to the same device /dev/md3 = /dev/sdb1, /dev/sdd1 [root@server2 root]# cat /proc/scsi/scsi Attached devices: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: HITACHI Model: OPEN-V Rev: 2109 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 00 Lun: 01 Vendor: HITACHI Model: OPEN-9*7 Rev: 2109 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi1 Channel: 00 Id: 00 Lun: 00 Vendor: HITACHI Model: OPEN-V Rev: 2109 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi1 Channel: 00 Id: 00 Lun: 01 Vendor: HITACHI Model: OPEN-9*7 Rev: 2109 Type: Direct-Access ANSI SCSI revision: 02 Additional info: It's not depend on QLogic Firmware, becose i have the same problem with diverent firmware.
Further debugging of the issue showed, that the problem is indeed in detect_multipath. It checks for same disks in this order: 1) id_eui64 2) id_fcph 3) serial_number 4) id_vendor 5) id_nonunique On the Hitachi 9980V the id_eui64 and id_fcph return 0 and serial_number returns a correct value, which is being used. On the Hitachi USP 100 it returns the same value for every LUN: sda,sdb, sde and sdf on the 9980V and sdc, sdd, sdg and sdh on the USP 100: /dev/sda ->id_fcph[0]==0 ->id_fcph[1]==0 /dev/sdb ->id_fcph[0]==0 ->id_fcph[1]==0 /dev/sdc ->id_fcph[0]==-2146564512 ->id_fcph[1]==3009284 /dev/sdd ->id_fcph[0]==-2146564512 ->id_fcph[1]==3009284 /dev/sde ->id_fcph[0]==0 ->id_fcph[1]==0 /dev/sdf ->id_fcph[0]==0 ->id_fcph[1]==0 /dev/sdg ->id_fcph[0]==-2146564512 ->id_fcph[1]==3009284 /dev/sdh ->id_fcph[0]==-2146564512 ->id_fcph[1]==3009284 serial_number works as well on the USP 100, though it is fooled by the FCPH value to believe it is the same LUN. As a quick fix we could rearrange the order of the tests to check the serial before FCPH, though I am not sure if it might break other storages that fail on serial (same value) and need the FCPH. It would be nice if we could have a similar mechanism as LVM2 DM has, though is is probably out of scope in RHEL3
RHEL 3 issue, bugzilla closed as wontfix, no customer references, closing. Internal Status set to 'Resolved' Status set to: Closed by Tech Resolution set to: 'Rejected' This event sent from IssueTracker by pdemauro issue 80018
This bug is filed against RHEL 3, which is in maintenance phase. During the maintenance phase, only security errata and select mission critical bug fixes will be released for enterprise products. Since this bug does not meet that criteria, it is now being closed. For more information of the RHEL errata support policy, please visit: http://www.redhat.com/security/updates/errata/ If you feel this bug is indeed mission critical, please contact your support representative. You may be asked to provide detailed information on how this bug is affecting you.