5.4.2 Enabling the removal of an I/O device
5.4.2 Enabling the removal of an I/O device
This section describes the procedure for enabling the removal of an I/O device by using the hot plug function.
To use the hot plug function, the hotplug service of Oracle Solaris needs to be enabled. Check the status of the service. If it is disabled, enable the hotplug service in the following procedure. The hot plug function is available only on the control domain and root domain.
To use the hot plug function, the hotplug service of Oracle Solaris needs to be enabled. Check the status of the service. If it is disabled, enable the hotplug service in the following procedure. The hot plug function is available only on the control domain and root domain.
# svcs hotplug ← If the STATE is disabled, enable the service. # svcadm enable hotplug ← Enables the hotplug service. |
For details on the cfgadm command used here, see "2 Dynamically Configuring Devices" in the Managing Devices in Oracle Solaris 11.2 or "Dynamically Configuring Devices (Tasks)" in the Oracle Solaris Administration: Devices and File Systems.
Note - To perform PCI Express card maintenance using the hot plug function, confirm that the card is hot-pluggable. To do so, check "Appendix A Cards That Support PCI Hot Plug and Dynamic Reconfiguration" in the Fujitsu M10/SPARC M10 Systems PCI Card Installation Guide. |
Enabling the removal of a PCI Express card
- Log in as a super-user to Oracle Solaris on the logical domain (control domain or root domain) to which the target PCI Express card is assigned.
- Check the mounting position (Ap_Id) of the target PCI Express card from the logical domain (control domain or root domain).
# cfgadm -a Ap_Id Type Receptacle Occupant Condition BB#0-PCI#0 pci-pci/hp connected configured ok |
- Release the target PCI Express card from the logical domain (control domain or root domain).
# cfgadm -c unconfigure <Ap_Id> |
- Stop the power supply to the target PCI Express card.
# cfgadm -c disconnect <Ap_Id> |
- Confirm that the PCI Express card has been released from the logical domain.
# cfgadm -a Ap_Id Type Receptacle Occupant Condition BB#0-PCI#0 unknown disconnected unconfigured unknown |
- In the above example, [Receptacle] and [Occupant] display "disconnected" and "unconfigured," respectively, which indicates that the PCI Express card mounted at BB#0-PCI#0 has been released from the logical domain.
- Set the ATTENTION LED of the target PCI Express card cassette to blink.
For removal, this work is not required.
# cfgadm -x led=attn,mode=blink <Ap_Id> |
- Check the following LEDs to determine whether the PCI Express card can be removed.- The ATTENTION LED (amber) of the PCI Express card is blinking.- The POWER LED of the PCI Express card is off.
For removal, the ATTENTION LED (amber) does not blink. Make sure that the POWER LED is off.
Enabling the removal of an internal disk
(1) Identifying the disk slot
To perform active replacement of an internal disk drive, you need to know the physical device name or logical device name of a drive to be installed or removed. If a disk error occurs in the system, from the system console, you can normally check for messages regarding a disk that is likely to fail or that has already failed. This information is also recorded in the /var/adm/messages file.
In these error messages, a failed internal disk drive is normally indicated with a physical device name or logical device name. In addition, depending on the application, the slot number of the disk may be reported.
Procedure to confirm the mounting position information on the HDD varies depending on the operating system.
For details, see "Using the diskinfo Command (Oracle Solaris 11 with SRU 11.4.27.82.1 or later applied)."
- For Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied
For details, see "Using the format command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)" or "Using the diskinfo command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)."
- For Oracle Solaris 10
For details, see "Using the diskinfo command (Oracle Solaris 10)."
In these error messages, a failed internal disk drive is normally indicated with a physical device name or logical device name. In addition, depending on the application, the slot number of the disk may be reported.
Procedure to confirm the mounting position information on the HDD varies depending on the operating system.
- For Oracle Solaris 11 with SRU 11.4.27.82.1 or later applied
For details, see "Using the diskinfo Command (Oracle Solaris 11 with SRU 11.4.27.82.1 or later applied)."
- For Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied
For details, see "Using the format command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)" or "Using the diskinfo command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)."
- For Oracle Solaris 10
For details, see "Using the diskinfo command (Oracle Solaris 10)."
Using the diskinfo Command (Oracle Solaris 11 with SRU 11.4.27.82.1 or later applied)
- Execute the diskinfo command to check the physical disk slot and BB-ID.(1) The device name and logical path name of the disk installed in the HDD4 of
BB#00
(2) The device name and logical path name of the disk installed in the HDD7 of
BB#00
# diskinfo D:devchassis-path c:occupant-compdev ------------------------------ --------------------- /dev/chassis/SYS/BB0/HDD0 - /dev/chassis/SYS/BB0/HDD1 - /dev/chassis/SYS/BB0/HDD2 - /dev/chassis/SYS/BB0/HDD3 - /dev/chassis/SYS/BB0/HDD4/disk c5t50000393D82954D6d0 (1) /dev/chassis/SYS/BB0/HDD5 - /dev/chassis/SYS/BB0/HDD6 - /dev/chassis/SYS/BB0/HDD7/disk c5t50000393B81B2446d0 (2) |
Using the format command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)
- Execute the showhardconf command to check the CMUL serial numbers, which show mounting position information, of the chassis.
XSCF> showhardconf SPARC M10-4S; + Serial:2081238017; Operator_Panel_Switch:Service; + System_Power:Off; System_Phase:Cabinet Power Off; Partition#0 PPAR_Status:Powered Off; Partition#1 PPAR_Status:Powered Off; BB#00 Status:Normal; Role:Master; Ver:2044h; Serial:2081238017; + FRU-Part-Number:CA07361-D202 A1 ; + Power_Supply_System: ; + Memory_Size:256 GB; CMUL Status:Normal; Ver:0101h; Serial:PP123001Y1 ; * BB#00 CMUL serial + FRU-Part-Number:CA07361-D941 C3 /7060911 ; + Memory_Size:128 GB; Type: A ; CPU#0 Status:Normal; Ver:4142h; Serial:00321144; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; CPU#1 Status:Normal; Ver:4142h; Serial:00322957; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; ------------------------Omitted------------------------ BB#01 Status:Normal; Role:Standby; Ver:2044h; Serial:2081230011; + FRU-Part-Number:CA07361-D202 A1 ; + Power_Supply_System: ; + Memory_Size:256 GB; CMUL Status:Normal; Ver:0101h; Serial:PP123203N0 ; * BB#01 CMUL serial + FRU-Part-Number:CA07361-D941 C3 /7060911 ; + Memory_Size:128 GB; Type: A ; CPU#0 Status:Normal; Ver:4142h; Serial:00320804; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; CPU#1 Status:Normal; Ver:4142h; Serial:00321030; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; MEM#00A Status:Normal; + Code:2c800118KSF1G72PZ-1G6E1 4531-1A94229F; + Type:04; Size:8 GB; ------------------------Omitted----------------------- |
- Execute the format command to confirm the physical disk slot.(1) to (5) in the following example specify the following:(1): Logical path name of the disk(2): The disk is mounted in the HDD00 slot of BB#01.(3): The disk is mounted in the HDD01 slot of BB#01.(4): The disk is mounted in the HDD00 slot of BB#00.(5): The disk is mounted in the HDD01 slot of BB#00.
# format Searching for disks...done AVAILABLE DISK SELECTIONS: 0. c2t50000394281B5312d0 <TOSHIBA-MBF2600RC-3706 cyl 64986 alt 2 hd 27 sec 668> <-- (1) /pci@8800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w50000394281b5312,0 /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD00/disk <-- (2) * The last four digits of the BB#1_CMUL serial number 1. c2t50000394281B59D6d0 <TOSHIBA-MBF2600RC-3706 cyl 64986 alt 2 hd 27 sec 668> <-- (1) /pci@8800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w50000394281b59d6,0 /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD01/disk <-- (3) 2. c0t500003942823C8C6d0 <TOSHIBA-MBF2600RC-3706 cyl 64986 alt 2 hd 27 sec 668> <-- (1) /pci@8000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w500003942823c8c6,0 /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD00/disk <-- (4) * The last four digits of the BB#0_CMUL serial number 3. c0t50000394281B517Ad0 <TOSHIBA-MBF2600RC-3706 cyl 64986 alt 2 hd 27 sec 668> <-- (1) /pci@8000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w50000394281b517a,0 /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD01/disk <-- (5) Specify disk (enter its number): |
Using the diskinfo command (Oracle Solaris 11 without SRU 11.4.27.82.1 or later applied)
- Execute the showhardconf command to check the CMUL serial numbers, which show mounting position information, of the chassis.
XSCF> showhardconf SPARC M10-4S; + Serial:2081238017; Operator_Panel_Switch:Service; + System_Power:Off; System_Phase:Cabinet Power Off; Partition#0 PPAR_Status:Powered Off; Partition#1 PPAR_Status:Powered Off; BB#00 Status:Normal; Role:Master; Ver:2044h; Serial:2081238017; + FRU-Part-Number:CA07361-D202 A1 ; + Power_Supply_System: ; + Memory_Size:256 GB; CMUL Status:Normal; Ver:0101h; Serial:PP123001Y1 ; * BB#00 CMUL serial + FRU-Part-Number:CA07361-D941 C3 /7060911 ; + Memory_Size:128 GB; Type: A ; CPU#0 Status:Normal; Ver:4142h; Serial:00321144; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; CPU#1 Status:Normal; Ver:4142h; Serial:00322957; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; ------------------------Omitted------------------------ BB#01 Status:Normal; Role:Standby; Ver:2044h; Serial:2081230011; + FRU-Part-Number:CA07361-D202 A1 ; + Power_Supply_System: ; + Memory_Size:256 GB; CMUL Status:Normal; Ver:0101h; Serial:PP123203N0 ; * BB#01 CMUL serial + FRU-Part-Number:CA07361-D941 C3 /7060911 ; + Memory_Size:128 GB; Type: A ; CPU#0 Status:Normal; Ver:4142h; Serial:00320804; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; CPU#1 Status:Normal; Ver:4142h; Serial:00321030; + Freq:3.000 GHz; Type:0x10; + Core:16; Strand:2; MEM#00A Status:Normal; + Code:2c800118KSF1G72PZ-1G6E1 4531-1A94229F; + Type:04; Size:8 GB; ------------------------Omitted----------------------- |
- Execute the diskinfo command to confirm the physical disk slot.(1) to (4) in the example below specify the following:(1): The device path name and logical path name of the disk mounted on the HDD0 of BB#01.(2): The device path name and logical path name of the disk mounted on the HDD1 of BB#01.(3): The device path name and logical path name of the disk mounted on the HDD0 of BB#00.(4): The device path name and logical path name of the disk mounted on the HDD1 of BB#00.
# diskinfo D:devchassis-path c:occupant-compdev ----------------------------------------------------------- ------------------ /dev/chassis/SYS/BB0/CMUL/HDD0 - /dev/chassis/SYS/BB0/CMUL/HDD1 - /dev/chassis/SYS/BB0/CMUL/HDD2 - /dev/chassis/SYS/BB0/CMUL/HDD3 - /dev/chassis/SYS/BB0/CMUL/HDD4 - /dev/chassis/SYS/BB0/CMUL/HDD5 - /dev/chassis/SYS/BB0/CMUL/HDD6 - /dev/chassis/SYS/BB0/CMUL/HDD7 - /dev/chassis/SYS/BB1/CMUL/HDD0 - /dev/chassis/SYS/BB1/CMUL/HDD1 - /dev/chassis/SYS/BB1/CMUL/HDD2 - /dev/chassis/SYS/BB1/CMUL/HDD3 - /dev/chassis/SYS/BB1/CMUL/HDD4 - /dev/chassis/SYS/BB1/CMUL/HDD5 - /dev/chassis/SYS/BB1/CMUL/HDD6 - /dev/chassis/SYS/BB1/CMUL/HDD7 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD00/disk c4t50000394281B5312d0 <-- (1) * The last four digits of the BB#01_CMUL serial number /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD01/disk c4t50000394281B59D6d0 <-- (2) /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD02 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD03 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD04 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD05 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD06 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d31bf/03N0_HDD07 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD00/disk c2t500003942823C8C6d0 <-- (3) * The last four digits of the BB#00_CMUL serial number /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD01/disk c2t50000394281B517Ad0 <-- (4) /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD02 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD03 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD04 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD05 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD06 - /dev/chassis/FUJITSU-BBEXP.500000e0e06d237f/01Y1_HDD07 - |
Using the diskinfo command (Oracle Solaris 10)
- Execute the diskinfo command to confirm the physical disk slot and logical system board (LSB).(1) to (4) in the example below specify the following:(1): The logical path name of the disk mounted on the HDD0.(2): The device path of the disk mounted on the HDD0 of LSB#0.(3): The logical path name of the disk mounted on the HDD1.(4): The device path of the disk mounted on the HDD1 of LSB#0.
# diskinfo -ap Enclosure path: 2081210007-physical-hba-0 Chassis Serial Number: 2081210007-physical-hba-0 Chassis Model: ORCL,SPARC64-X Enclosure path: /dev/es/ses0 Chassis Serial Number: 500000e0e06d233f Chassis Model: FUJITSU-BBEXP Label Disk name Vendor Product Vers -------- ---------------------- -------- -------------- ---- HDD_0 c0t50000393D8289180d0 TOSHIBA MBF2600RC 3706 <-- (1) Physical path -------------------------------- 0: /pci@8000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w50000393D8289180,0<--(2) * LSB#0 HDD_1 c0t50000393D82891D0d0 TOSHIBA MBF2600RC 3706 <-- (3) Physical path -------------------------------- 0: /pci@8000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@w50000393D82891D0,0<--(4) * LSB#0 |
- Device paths differ depending on the system configuration.
Table 5-6 lists examples of the device paths displayed for logical system boards (LSBs) in a SPARC M10 system.
This table can be used for mapping of the device paths of internal disk drives and the logical system board (LSB) numbers. The letters "WWN" in the table represent the WWN assigned to the disk.
LSB No. | Device path |
---|---|
LSB#0 | /pci@8000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#1 | /pci@8800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#2 | /pci@9000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#3 | /pci@9800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#4 | /pci@a000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#5 | /pci@a800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#6 | /pci@b000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#7 | /pci@b800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#8 | /pci@c000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#9 | /pci@c800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#10 | /pci@d000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#11 | /pci@d800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#12 | /pci@e000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#13 | /pci@e800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#14 | /pci@f000/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
LSB#15 | /pci@f800/pci@4/pci@0/pci@0/scsi@0/iport@f/disk@wWWN,0 |
- Execute the showboards command to confirm the BB-ID of the chassis in which the disk requiring maintenance is mounted.
Identify the BB-ID from the LSB number confirmed in step 1. The BB-ID of the chassis requiring maintenance is indicated by the system board (PSB) number.
XSCF> showboards -a PSB PPAR-ID(LSB) Assignment Pwr Conn Conf Test Fault ---- ------------ ----------- ---- ---- ---- ------- -------- 00-0 00(00) Assigned y y y Passed Normal * BB#0 * LSB#0 01-0 00(01) Assigned y y y Passed Normal |
(2) Releasing an internal disk from the system
- Execute the cfgadm command to check the configuration of the internal disk.
To replace or remove an internal disk, use the cfgadm -al command to check the Ap_ID of the disk requiring maintenance and make a note of it.
If multiple internal disks are mounted, check for the disk that matches the physical device name confirmed in "(1) Identifying the disk slot."
In the following execution example, Ap_ID is "c2::dsk/c2t50000394281B59D6d0".
# cfgadm -al Ap_Id Type Receptacle Occupant Condition <Omitted> c2 scsi-sas connected configured unknown c2::dsk/c2t50000394281B59D6d0 disk connected configured unknown c2::es/ses0 ESI connected configured unknown c2::smp/expd0 smp connected configured unknown c3 scsi-sas connected unconfigured unknown c4 scsi-sas connected unconfigured unknown c5 fc connected unconfigured unknown c6 fc connected unconfigured unknown c7 scsi-sas connected unconfigured unknown <Omitted> |
- Stop all applications from using the internal disk.
- Execute the cfgadm command to release the internal disk requiring maintenance from the system.
Enter the Ap_ID you noted in advance.
# cfgadm -c unconfigure <Ap_Id> |
- Execute the cfgadm command to confirm that the internal disk requiring maintenance is disconnected.
The disconnected internal disk is displayed as "unconfigured."
# cfgadm -al Ap_Id Type Receptacle Occupant Condition <Omitted> c2 scsi-sas connected configured unknown c2::dsk/c2t50000394281B4D9Ad0 disk connected unconfigured unknown c2::es/ses0 ESI connected configured unknown c2::smp/expd0 smp connected configured unknown c3 scsi-sas connected unconfigured unknown c4 scsi-sas connected unconfigured unknown c5 fc connected unconfigured unknown c6 fc connected unconfigured unknown c7 scsi-sas connected unconfigured unknown <Omitted> |
- Execute the cfgadm command to identify the location of the internal disk requiring maintenance by causing the CHECK LED of the disk to blink.
Enter the Ap_ID you noted in advance.
# cfgadm -x led=fault,mode=blink <Ap_ID> |
- Check the following LEDs to determine whether the internal disk requiring maintenance can be removed:- READY LED (green) of the internal disk: On- CHECK LED (amber) of the internal disk: Blinking
< Previous Page | Next Page >