Sponsored Content
Operating Systems Solaris Qlogic HBA card in E450 with errors Post 302243106 by DukeNuke2 on Friday 3rd of October 2008 05:39:24 PM
Old 10-03-2008
from what i see, your hba's are fine. the errors looks like a faulty scsi device. so maybe one of your disks is bad...
 

10 More Discussions You Might Find Interesting

1. Solaris

HBA Card

Hello...I got a problem here. I have HBAs installed in one of the domains in E20 and I gotta find out the brand and firmware. But when I do a prtconf, I cannot see the HBA. For other servers installed with HBAs, when I do prtconf on them I can see either qlc or em. Can you guys enlighten me how to... (5 Replies)
Discussion started by: xiaochensg
5 Replies

2. UNIX for Advanced & Expert Users

Sun internal disks and HBA card

I have a V490 that has couple internal disks attached to the same dual-port HBA card that I'm planning attach to the SAN for additional storage. Would I sacrifice performance if I use the same HBA? Is it better to throw in a new HBA instead? (1 Reply)
Discussion started by: kiem
1 Replies

3. AIX

Need HBA card driver

Hi All, SUN emulex HBA card has been installed on P520 box. When i run the cfgmgr i am getting the following message. lpar2#cfgmgr cfgmgr: 0514-621 WARNING: The following device packages are required for device support but are not currently installed.... (0 Replies)
Discussion started by: b_manu78
0 Replies

4. Solaris

Qlogic HBA's - lputil / lpfc.conf

Hi all Hope your all well and happy in solaris land :-) Recently purchased some Qlogic HBA's for a few solaris servers. Was told that you cant have mixed chipsets and as this client is all qlogic, have to have hba's using qlogic. 1) Is that correct ? Cant mix different chipsets of... (0 Replies)
Discussion started by: sbk1972
0 Replies

5. Solaris

Qlogic HBA + Veritas - Part 2

Hi all, New to the world of SAN, so forgive me for any ambigious terminology. I have a sun 6800 chassis with two domains, both domains connected to a San ancor sanbox ( 2 x Storeedge T3 / Mcdata switch ). Current Qlogic HBA's are 1Gb glc adaptors, version of qlc drivers... (1 Reply)
Discussion started by: sbk1972
1 Replies

6. Solaris

Solaris9 - qlogic HBAs - luxadm reports "Found path to 0 HBA ports"

Hi All, I'm actually trying to configure a Qlogic HBA on a solaris9 connecting in a Brocade SAN. I installed Solaris9 and now trying to look at FC devices.. bash-2.05# luxadm -e port Found path to 0 HBA ports bash-2.05# luxadm probe No Network Array enclosures found in /dev/es... (7 Replies)
Discussion started by: ilan
7 Replies

7. Solaris

How to enable hba port of a two port hba card

Hi, i have a two port hba card in solaris 10 system one of which is disabled. How to enable and configure the other port. ---------- Post updated at 08:17 AM ---------- Previous update was at 03:40 AM ---------- please help :( (7 Replies)
Discussion started by: rishiraaz
7 Replies

8. Solaris

Which FC HBA card support Solaris9

I have JNI FCE-6460-N fc card , it is detected in OK prompt.....I connected Sun StorEDGE A5000 FC storage. in OK prompt, PROBE-SCSI-ALL showing A5000 fc hdds details in ok prompt..... but in OS level.....not showing the hdd, i tried in DEVFSADM command and FORMAT command not showing A5000... (3 Replies)
Discussion started by: anysystem
3 Replies

9. Red Hat

lun masking at qlogic HBA level

need to do LUN masking at qlogic HBA level..at RHEL 5.x I am using qla* module at part of the Unix kernel, it is not OEM drivers that come with the HBA card, so I can't use Sansurfer to do this.. Is there a way to do this at all ?? :wall: (0 Replies)
Discussion started by: ppchu99
0 Replies

10. Red Hat

HBA card faulty

Hi, How to determine if HBA card in getting faulty from linux host ? any pointers. (2 Replies)
Discussion started by: saurabh84g
2 Replies
pci(4)																	    pci(4)

NAME
pci, pcie - configuration files for PCI and PCI Express device drivers The Peripheral Component Interconnect (PCI) bus is a little endian bus. PCI Express (PCIe) and PCI-X are successors to PCI. All three types of devices share the same configuration parameters. What is specified here for PCI devices applies to PCI-X 1.0 devices as well. All three types of devices are self-identifying, which means that these devices provide configuration parameters to the system that allow the system to identify the device and its driver. The configuration parameters are represented in the form of name-value pairs that can be retrieved using the DDI property interfaces. See ddi_prop_lookup(9F) for details. The bus properties of PCI devices or logical bus properties of PCIe devices are derived from PCI configuration space, or supplied by the Fcode PROM, if it exists. Therefore, driver configuration files are not necessary for these devices. On some occasions, drivers for PCI and PCIe devices can use driver configuration files to provide driver private properties through the global property mechanism. See driver.conf(4) for further details. Driver configuration files can also be used to augment or override prop- erties for a specific instance of a driver. All bus drivers of PCI and PCIe devices recognize the following properties: reg An arbitrary length array where each element of the array consists of a 5-tuple of 32-bit values. Each array element describes a logi- cally contiguous mappable resource on the PCI bus or PCIe device tree. The first three values in the 5-tuple describe the PCI address of the mappable resource. The first tuple contains the following infor- mation: Bits 0 - 7 8-bit register number Bits 8 - 10 3-bit function number Bits 11 - 15 5-bit device number Bits 16 - 23 8-bit bus number Bits 24 - 25 2-bit address space type identifier Bits 31 - 28 Register number extended bits 8:11 for extended config space. Zero for conventional configuration space. The address space type identifier can be interpreted as follows: 0x0 configuration space 0x1 I/O space 0x2 32-bit memory space address 0x3 64-bit memory space address The bus number is a unique identifying number assigned to each PCI bus or PCIe logical bus within its domain. The device number is a unique identifying number assigned to each device on a PCI bus or PCIe logical bus. Note that a device number is unique only within the set of device numbers for a particular bus or logical bus. Each PCI or PCIe device can have one to eight logically independent functions, each with its own independent set of configuration reg- isters. Each function on a device is assigned a function number. For a device with only one function, the function number must be 0. The register number fields select a particular register within the set of configuration registers corresponding to the selected func- tion. When the address space type identifier indicates configuration space, non-zero register number extended bits select registers in extended configuration space. The second and third values in the reg property 5-tuple specify the 64-bit address of the mappable resource within the PCI or PCIe address domain. The second 32-bit tuple corresponds to the high order four bytes of the 64-bit address. The third 32-bit tuple corre- sponds to the low order bytes. The fourth and fifth 32-bit values in the 5-tuple reg property specify the size of the mappable resource. The size is a 64-bit value, where the fourth tuple corresponds to the high order bytes of the 64-bit size and the fifth corresponds to the low order. The driver can refer to the elements of this array by index, and construct kernel mappings to these addresses using ddi_regs_map_set- up(9F). The index into the array is passed as the rnumber argument of ddi_regs_map_setup(9F). At a high-level interrupt context, you can use the ddi_get* and ddi_put* family of functions to access I/O and memory space. However, access to configuration space is not allowed when running at a high-interrupt level. interrupts This property consists of a single-integer element array. Valid interrupt property values are 1, 2, 3, and 4. This value is derived directly from the contents of the device's configuration-interrupt-pin register. A driver should use an index value of 0 when registering its interrupt handler with the DDI interrupt interfaces. All PCI and PCIe devices support the reg property. The device number and function number as derived from the reg property are used to con- struct the address part of the device name under /devices. Only devices that generate interrupts support an interrupts property. Occasionally it might be necessary to override or augment the configuration information supplied by a PCI or PCIe device. This change can be achieved by writing a driver configuration file that describes a prototype device node specification containing the additional proper- ties required. For the system to merge the prototype node specification into an actual device node, certain conditions must be met. o First, the name property must be identical. The value of the name property needs to match the binding name of the device. The binding name is the name chosen by the system to bind a driver to a device and is either an alias associated with the driver or the hardware node name of the device. o Second, the parent property must identify the PCI bus or PCIe logical bus. o Third, the unit-address property must identify the card. The format of the unit-address property is: DD[,F] where DD is the device number and F is the function number. If the function number is 0, only DD is specified. Example 1: Sample Configuration File An example configuration file called ACME,scsi-hba.conf for a PCI driver called ACME,scsi-hba follows: # # Copyright (c) 1995, ACME SCSI Host Bus Adaptor # ident "@(#)ACME,scsi-hba.conf 1.1 96/02/04" name="ACME,scsi-hba" parent="/pci@1,0/pci@1f,4000" unit-address="3" scsi-initiator-id=6; hba-advanced-mode="on"; hba-dma-speed=10; In this example, a property scsi-initiator-id specifies the SCSI bus initiator id that the adapter should use, for just one particular instance of adapter installed in the machine. The name property identifies the driver and the parent property to identify the particular bus the card is plugged into. This example uses the parent's full path name to identify the bus. The unit-address property identifies the card itself, with device number of 3 and function number of 0. Two global driver properties are also created: hba-advanced-mode (which has the string value on) and hba-dma-speed (which has the value 10 M bit/s). These properties apply to all device nodes of the ACME,scsi-hba. Configuration files for PCIe devices are similar. Shown below is an example configuration file called ACME,pcie-widget.conf for a PCIe driver called ACME,pcie-widget. # # Copyright (c) 2005, ACME PCIe Widget Adapter # ident "@(#)ACME,pcie-widget.conf 1.1 05/11/14" name="ACME,pcie-widget" parent="/pci@780" unit-address="2,1" debug-mode=12; In this example, we provide a property debug-mode for a particular PCIe device. As before, the logical bus is identified by the pathname of the parent of the device. The device has a device number of 2, and a function number of 1. See attributes(5) for descriptions of the following attributes: +-----------------------------+-----------------------------+ | ATTRIBUTE TYPE | ATTRIBUTE VALUE | +-----------------------------+-----------------------------+ |Architecture |SPARC, | +-----------------------------+-----------------------------+ driver.conf(4), attributes(5), ddi_intr_add_handler(9F), ddi_prop_lookup(9F), ddi_regs_map_setup(9F) IEEE 1275 PCI Bus Binding http://playground.sun.com/1275/bindings/pci/pci-express.txt PCIe devices support an extended configuration space unavailable to PCI devices. While PCIe devices can be operated using a PCI device driver, operating them using a PCIe device driver can make use of the extended properties and features made available only in the extended configuration space. 13 May 2005 pci(4)
All times are GMT -4. The time now is 09:28 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy