11-27-2006
VxVM/Hba cards/Solaris 10 FS issue
Hi all,
I have a compatibly issues with a new hardware
My config is the following one :
Solaris 10
V490 Hardware
2 X HBA Qlogic 2310f cards
An HP EVA 5000 for storage ...
Veritas VxVM 4.1 MP1
When I reboot my server, VxVM is trying to mount/use the volume before it's discovered/available with the SAN.
The Temp solution is to mount manually all the volumes with the following command : vxvol -g loc1vg startall & mountall
The other workaround is to reboot my server each time with the command : reboot -- -r
If I reboot my server with Init 6 :
hba1: QLogic Fibre Channel Host Adapter fcode version 2.00.09
hba1: Firmware v3.3.12 (ipx)
QLogic Fibre Channel Driver v4.20 Instance: 2
hba2: QLogic Fibre Channel Host Adapter fcode version 2.00.09
hba2: Firmware v3.3.12 (ipx)
Hostname: svdlux69
NOTICE: VxVM vxdmp V-5-0-34 added disk array DISKS, datype = Disk
cpqccl288: found array controller device at tgt32, lun0
cpqccl288: Vendor/Product ID = COMPAQ HSV110 (C)COMPAQ
cpqccl291: found array controller device at tgt35, lun0
cpqccl291: Vendor/Product ID = COMPAQ HSV110 (C)COMPAQ
cpqccl33: found array controller device at tgt33, lun0
cpqccl33: Vendor/Product ID = COMPAQ HSV110 (C)COMPAQ
cpqccl34: found array controller device at tgt34, lun0
cpqccl34: Vendor/Product ID = COMPAQ HSV110 (C)COMPAQ
NOTICE: VxVM vxdmp V-5-0-148 enabled path 118/0x30 belonging to the dmpnode 282/0x18
NOTICE: VxVM vxdmp V-5-0-148 enabled path 118/0x38 belonging to the dmpnode 282/0x10
Nov 22 09:28:45 spagent[418]: Starting spagent.
checking ufs filesystems
/dev/vx/rdsk/bootdg/opt: is clean.
checking vxfs filesystems
Running parallel replay fsck ...
/dev/vx/rdsk/loc1vg/loc1lv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/opttivlv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/tivlv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/ccmslv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/bgslv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/toolslv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/ecclv:file system check failure, aborting ...
/dev/vx/rdsk/loc1vg/crashlv:file system check failure, aborting ...
The following filesystems could not be repaired :
/dev/vx/rdsk/loc1vg/loc1lv
/dev/vx/rdsk/loc1vg/ccmslv
/dev/vx/rdsk/loc1vg/opttivlv
/dev/vx/rdsk/loc1vg/tivlv
/dev/vx/rdsk/loc1vg/toolslv
/dev/vx/rdsk/loc1vg/bgslv
/dev/vx/rdsk/loc1vg/ecclv
/dev/vx/rdsk/loc1vg/crashlv
fsckall failed with exit code 1.
WARNING - Unable to repair one or more filesystems.
Run fsck manually (fsck filesystem...).
I tried to put Sleep commands on all the scripts files (veritas+ Hp) without success ....
Any ideas ??
Thanks,
Fabien
10 More Discussions You Might Find Interesting
1. Solaris
Has anyone ever relocated an HBA card after already having it configured in another PCI slot? I've found that the HBA instance #'s start incrementing past the previous. It results in me having to have my instance numbers in /kernel/drv/qla2300.conf be 4,5,6,7 instead of 0,1,2,3. Cleanup with... (5 Replies)
Discussion started by: kduffin
5 Replies
2. Solaris
Due to speed mismatch, the server kept crashing. Sun suggests that I need to move the two HBA cards from the 66Mhz to the 33Mhz slots. There are tons of disks that are connected to these HBAs and I have Veritas volume manager.
Does anyone have a list of what I need to do prior/after moving to... (3 Replies)
Discussion started by: kiem
3 Replies
3. UNIX for Advanced & Expert Users
Hi,
I'm trying to determine which HBA cards are installed on my Solaris 10 (Sun-Fire-V240) machine.
The relevant data I have is below, but from it I cannot ascertain for sure if I have Emulex 10000 or 11000 HBA cards.
Can anyone suggest how to determine my HBA hardware ?
Thanks,
Ron.
... (3 Replies)
Discussion started by: ronbarak
3 Replies
4. Solaris
Hi,
I'm trying to determine which HBA cards are installed on my Solaris 10 (Sun-Fire-V240) machine.
The relevant data I have is below, but from it I cannot ascertain for sure if I have Emulex 10000 or 11000 HBA cards.
Can anyone suggest how to determine my HBA hardware ?
Thanks,
Ron.
... (4 Replies)
Discussion started by: ronbarak
4 Replies
5. Solaris
Hiii... Every one....
I want to know the command to get the details and number of HBA cards
attached to a server.
I know that we can use "powermt display", but nither my cards are connected to the storage nor powerpath is active.
I am using SUN server with Solaris -9.
Thanks in... (4 Replies)
Discussion started by: Reboot
4 Replies
6. Solaris
Hi,
Quick question if anyone knows this. Is there a command I can use in Veritas Volume manager on Solaris that will tell me what the name of the SAN I am connected to? We have a number of SANs so I am unsure which one my servers are connected to. Thanks. (13 Replies)
Discussion started by: gwhelan
13 Replies
7. BSD
Hi,
The title is a bit misleading, but it seems I can't re-edit the title after creating the post - apologies.
I have been handed a variety of servers, whereas quite a few of these are older FreeBSD (4.8, thank you legacy). I now have one of these servers which has a HBA card installed, and... (0 Replies)
Discussion started by: brightstorm
0 Replies
8. Solaris
This is going to mean nothing to most....
I'm working on a E4500 server running Solaris 2.5.1 and VXVM 2.5.5 :eek:
I've replaced the rootmirror and would like to monitor progress of the resync
It's only a 2Gb /partition so it shouldn't take too long.
In later versions
vxtask list
... (5 Replies)
Discussion started by: Martincorneuk
5 Replies
9. Red Hat
hi all,
I am trying to view the hba cards on a rhel7 server (storage says they are there, but I am not seeing them.
I tried lspci -nn | grep -i hba
lspci |grep qla
I tried looking in /sys/class/fc_host.. but it is empty.
I am not sure if I am looking in the right spot... (0 Replies)
Discussion started by: gartie
0 Replies
10. Solaris
hi everone
i am experiencing disconnectivity from storage to My database machines. i have doubt on my FC cards in my sparc T4-2 machines. i want to know how should i check from my solaris that whether the cards are ok or not because once the disks get disconnect from SAN so it come back again but... (5 Replies)
Discussion started by: janakors
5 Replies
ISP(4) BSD Kernel Interfaces Manual ISP(4)
NAME
isp -- Qlogic based SCSI and FibreChannel SCSI Host Adapters
SYNOPSIS
isp* at pci? dev? function? (PCI)
isp* at sbus? slot ? offset ? (SBus)
scsibus* at isp?
DESCRIPTION
This driver provides access to SCSI or FibreChannel devices.
SCSI features include support for Ultra SCSI and wide mode transactions for SCSI, and LVD (for the ISP1080 and ISP1280),
Fibre Channel support uses FCP SCSI profile for FibreChannel. and uses Class 3 connections only. Support is available for Public and Pri-
vate loops. Command tagging is supported for all (in fact, FibreChannel requires tagging).
CONFIGURATION
An optional flags 0x80 appended to the above isp* declarations will disable the download of driver firmware, which means you use whatever
firmware is running on the card. If no firmware is running on the card, the driver cannot operate the card.
An optional flags 0x40 appended to the above isp* declarations (can be OR'd in with the other config flags option) will keep the driver from
looking at device or bus NVRAM settings (this is in case NVRAM is just wrong and you have the card in a platform where it is inconvenient to
change NVRAM settings on the card).
HARDWARE
Supported cards include:
ISP1000
SBus Fast Wide, Ultra Fast Wide cards, Single Ended or Differential cards.
PTI SBS440
Performance Technology ISP1000 variants.
ISP1020
Qlogic 1020 Fast Wide and Differential Fast Wide PCI cards.
ISP1040
Qlogic 1040 Ultra Wide and Differential Ultra Wide PCI cards.
PTI SBS450
Performance Technology ISP1040 variants.
Qlogic 1240
Qlogic 1240 Dual Bus Ultra Wide and Differential Ultra Wide PCI cards.
Qlogic 1080
Qlogic 1280 LVD Ultra2 Wide PCI cards.
Qlogic 1280
Qlogic 1280 Dual Bus LVD Ultra2 Wide PCI cards.
Qlogic 2100
Qlogic 2100 and 2100A Copper and Optical Fibre Channel Arbitrated Loop
Qlogic 2102
Qlogic Dual Loop 2100A Optical Fibre Channel Arbitrated Loop PCI cards.
Qlogic 2200
Qlogic 2200 Copper and Optical Fibre Channel Arbitrated Loop PCI cards.
Qlogic 2202
Qlogic 2200 Dual Bus Optical Fibre Channel Arbitrated Loop PCI cards.
Qlogic 2204
Qlogic 2200 Quad Bus Optical Fibre Channel Arbitrated Loop PCI cards.
Qlogic 2300
Qlogic 2300 2-Gigabit Optical Fibre Channel PCI cards.
Qlogic 2312
Qlogic 2300 2-Gigabit Dual Channel Optical Fibre Channel PCI cards.
PTI SBS470
Performance Technology ISP2100 variants.
Antares P-0033
Antares Microsystems ISP2100 variants.
Qlogic 2400
Qlogic 2400 4-Gigabit Optical Fibre Channel PCI cards.
Qlogic 2500
Qlogic 2500 8-Gigabit Optical Fibre Channel PCI cards.
SEE ALSO
cd(4), intro(4), scsi(4), sd(4), st(4)
AUTHORS
The isp driver was written by Matthew Jacob for NASA/Ames Research Center.
BUGS
The driver currently ignores some NVRAM settings.
The driver currently doesn't do error recovery for timed out commands very gracefully.
Sometimes, when booting, the driver gets stuck waiting for the Fibre Channel firmware to tell it that the loop port database is ready. In
this case you'll see an announcement that the loop state has a value of 0x1. To unwedge the system, unplug and replug the fibre channel con-
nection, or otherwise cause a LIP (Loop Initialization Primitive sequence) - this will kick the firmware into getting unstuck.
BSD
June 24, 2009 BSD