On Tue, 2012-02-28 at 07:58 -0700, Peter W. Morreale wrote: > > Hi all, > > I'm trying to get a LIO FC target up and running in our lab without > success. It has been suggested that I forward my environment to the > list for help. > > The goal is to export /dev/sdb from the target. (Yes, I know I should > use the disk-id, just a test atm) > > The x86_64 machines are: > > Initiator: > o openSUSE 12.1, fully patched. > o QLogic QLE2462 - PCI-Express Dual Channel 4Gb Fibre Channel HBA > o Qlogic 5.06 firmware. > o wwn's > /sys/class/fc_host/host41/node_name: 0x2000001b3209a0df > /sys/class/fc_host/host41/node_name: 0x2001001b3229a0df > > Target: > o openSUSE 12.1 with LIO kernel (last commit: 21fd618b0) > o QLogic QLE2462 - PCI-Express Dual Channel 4Gb Fibre Channel HBA > o Qlogic 5.06 firmware. > o targetcli built from the git trees at: > http://risingtidesystems.com/git/ > > > Note that the Qlogic bios on both machines have set the transfer speed > to 4GB. (no auto-negotiate) > > The target config is: > > lisa:~ # targetcli > Welcome to the targetcli shell: > > Copyright (c) 2011 by RisingTide Systems LLC. > > Visit us at http://www.risingtidesystems.com. > > Using loopback fabric module. > Using tcm_fc fabric module. > Can't load fabric module ib_srpt. > Using qla2xxx fabric module. > Using iscsi fabric module. > /> ls > o- / ......................................................................... [...] > o- > backstores .............................................................. [...] > | o- fileio ................................................... [0 > Storage Object] > | o- iblock ................................................... [1 > Storage Object] > | | o- lio-test ............................................. > [/dev/sdb activated] > | o- pscsi .................................................... [0 > Storage Object] > | o- rd_dr .................................................... [0 > Storage Object] > | o- rd_mcp ................................................... [0 > Storage Object] > o- > iscsi .............................................................. [0 > Target] > o- > loopback ........................................................... [0 > Target] > o- qla2xxx ........................................................... > [2 Targets] > | o- > 21:00:00:1b:32:09:48:df ........................................... > [enabled] > | | o- > acls ............................................................. [2 > ACLs] > | | | o- 20:00:00:1b:32:09:a0:df .................................. [1 > Mapped LUN] > | | | | o- mapped_lun0 ............................................... > [lun0 (rw)] > | | | o- 20:01:00:1b:32:29:a0:df .................................. [1 > Mapped LUN] > | | | o- mapped_lun0 ............................................... > [lun0 (rw)] > | | o- > luns .............................................................. [1 > LUN] > | | o- lun0 ....................................... [iblock/lio-test > (/dev/sdb)] > | o- > 21:01:00:1b:32:29:48:df ........................................... > [enabled] > | o- > acls ............................................................. [2 > ACLs] > | | o- 20:00:00:1b:32:09:a0:df .................................. [1 > Mapped LUN] > | | | o- mapped_lun0 ............................................... > [lun0 (rw)] > | | o- 20:01:00:1b:32:29:a0:df .................................. [1 > Mapped LUN] > | | o- mapped_lun0 ............................................... > [lun0 (rw)] > | o- > luns .............................................................. [1 > LUN] > | o- lun0 ....................................... [iblock/lio-test > (/dev/sdb)] > o- > tcm_fc ............................................................. [0 > Target] > /> > This all looks fine.. > Clearing dmesg and starting the target produces the following in dmesg. > Note that qla2xxx is loaded in the initiator and only one cable is > attached at this point: > <SNIP> > [84662.364725] qla2xxx [0000:0d:00.1]-00fa:12: QLogic Fibre Channed HBA > Driver: 8.03.07.12-k. > [84662.364728] qla2xxx [0000:0d:00.1]-00fb:12: QLogic QLE2462 - > PCI-Express Dual Channel 4Gb Fibre Channel HBA. > [84662.364742] qla2xxx [0000:0d:00.1]-00fc:12: ISP2432: PCIe (2.5GT/s > x4) @ 0000:0d:00.1 hdma+ host#=12 fw=5.06.05 (9496). > [84662.364745] qla_target(0): using 64 Bit PCI addressing > [84662.451949] qla2xxx [0000:0d:00.1]-00af:12: Performing ISP error > recovery - ha=ffff88024cac0000. > [84662.502098] qla2xxx [0000:0d:00.1]-0092:12: Loading via > request-firmware. > [84662.733062] qla2xxx [0000:0d:00.1]-5009:12: LIP occurred (f8e8). > [84662.744031] qla2xxx [0000:0d:00.1]-500a:12: LOOP UP detected (4 > Gbps). > [84662.843297] qla_target(0): session for wwn 21:01:00:1b:32:29:a0:df > (loop_id 0, s_id 0:0:e8, confirmed completion supported) added Ok, so a session from 21:01:00:1b:32:29:a0:df is able to login here.. > [84663.530987] qla2xxx [0000:0d:00.0]-00af:11: Performing ISP error > recovery - ha=ffff88024c640000. > [84663.581096] qla2xxx [0000:0d:00.0]-0092:11: Loading via > request-firmware. > [84684.303055] qla2xxx [0000:0d:00.0]-8038:11: Cable is unplugged... > [84684.303059] qla2xxx [0000:0d:00.0]-803b:11: Firmware ready **** > FAILED ****. > > This is from the other port not having a physical link.. > > Now for the initiator. Rmmod qla2xxx, dmesg -c, modprobe qla2xxx, > dmesg, and lsscsi: > > snerble:~ # dmesg > [419610.718692] qla2xxx [0000:00:00.0]-0005: : QLogic Fibre Channel HBA > Driver: 8.03.07.07-k. > [419610.718803] qla2xxx 0000:0e:00.0: PCI INT A -> GSI 18 (level, low) > -> IRQ 18 > [419610.718838] qla2xxx [0000:0e:00.0]-001d: : Found an ISP2432 irq 18 > iobase 0xffffc90011788000. > [419610.719144] qla2xxx 0000:0e:00.0: irq 64 for MSI/MSI-X > [419610.719153] qla2xxx 0000:0e:00.0: irq 65 for MSI/MSI-X > [419610.719240] qla2xxx [0000:0e:00.0]-0040:47: Configuring PCI space... > [419610.719246] qla2xxx 0000:0e:00.0: setting latency timer to 64 > [419610.751940] qla2xxx [0000:0e:00.0]-0061:47: Configure NVRAM > parameters... > [419610.787880] qla2xxx [0000:0e:00.0]-0078:47: Verifying loaded RISC > code... > [419610.800470] qla2xxx [0000:0e:00.0]-0092:47: Loading via > request-firmware. > [419611.031056] qla2xxx [0000:0e:00.0]-00c3:47: Allocated (64 KB) > EFT ... > [419611.031169] qla2xxx [0000:0e:00.0]-00c5:47: Allocated (1285 KB) for > firmware dump. > [419611.042192] scsi47 : qla2xxx > [419616.047040] qla2xxx [0000:0e:00.0]-00fa:47: QLogic Fibre Channed HBA > Driver: 8.03.07.07-k. > [419616.047044] qla2xxx [0000:0e:00.0]-00fb:47: QLogic QLE2462 - > PCI-Express Dual Channel 4Gb Fibre Channel HBA. > [419616.047059] qla2xxx [0000:0e:00.0]-00fc:47: ISP2432: PCIe (2.5GT/s > x4) @ 0000:0e:00.0 hdma+ host#=47 fw=5.06.05 (9496). > [419616.047086] qla2xxx 0000:0e:00.1: PCI INT B -> GSI 19 (level, low) > -> IRQ 19 > [419616.047154] qla2xxx [0000:0e:00.1]-001d: : Found an ISP2432 irq 19 > iobase 0xffffc900117a0000. > [419616.047486] qla2xxx 0000:0e:00.1: irq 66 for MSI/MSI-X > [419616.047493] qla2xxx 0000:0e:00.1: irq 67 for MSI/MSI-X > [419616.047566] qla2xxx [0000:0e:00.1]-0040:48: Configuring PCI space... > [419616.047573] qla2xxx 0000:0e:00.1: setting latency timer to 64 > [419616.079704] qla2xxx [0000:0e:00.1]-0061:48: Configure NVRAM > parameters... > [419616.115267] qla2xxx [0000:0e:00.1]-0078:48: Verifying loaded RISC > code... > [419616.126056] qla2xxx [0000:0e:00.1]-0092:48: Loading via > request-firmware. > [419616.357053] qla2xxx [0000:0e:00.1]-00c3:48: Allocated (64 KB) > EFT ... > [419616.357177] qla2xxx [0000:0e:00.1]-00c5:48: Allocated (1285 KB) for > firmware dump. > [419616.368177] scsi48 : qla2xxx > [419616.374827] qla2xxx [0000:0e:00.1]-500c:48: LIP reset occurred > (f8f7). > [419616.377724] qla2xxx [0000:0e:00.1]-5009:48: LIP occurred (f8f7). > [419616.390588] qla2xxx [0000:0e:00.1]-500a:48: LOOP UP detected (4 > Gbps). > [419616.885041] qla2xxx [0000:0e:00.1]-00fa:48: QLogic Fibre Channed HBA > Driver: 8.03.07.07-k. > [419616.885045] qla2xxx [0000:0e:00.1]-00fb:48: QLogic QLE2462 - > PCI-Express Dual Channel 4Gb Fibre Channel HBA. > [419616.885058] qla2xxx [0000:0e:00.1]-00fc:48: ISP2432: PCIe (2.5GT/s > x4) @ 0000:0e:00.1 hdma+ host#=48 fw=5.06.05 (9496). So it looks like the session is started at this point, but no LUN SCAN is occuring for some reason.. > [419633.086130] qla2xxx [0000:0e:00.0]-8038:47: Cable is unplugged... > [419633.086134] qla2xxx [0000:0e:00.0]-803b:47: Firmware ready **** > FAILED ****. And the other port reports a link layer failure as expected.. > snerble:~ # lsscsi > [0:0:1:0] cd/dvd TEAC DV-28E-V 1.AB /dev/sr0 > [4:0:0:0] cd/dvd Intel(R) RMM2 VDrive 1 0.01 /dev/sr3 > [5:0:0:0] cd/dvd Intel(R) RMM2 VDrive 2 0.01 /dev/sr1 > [6:0:0:0] cd/dvd Intel(R) RMM2 VDrive 3 0.01 /dev/sr2 > [7:0:0:0] cd/dvd Intel(R) RMM2 VDrive 4 0.01 /dev/sr4 > [8:0:8:0] enclosu ESG-SHV. SCA HSBP M11.... 1.41 - > [8:2:0:0] disk INTEL SROMBSAS18E 1.03 /dev/sda > snerble:~ # > > No device on the initiator (from the target :-). The above is exactly > what I see after a clean reboot of the initiator without the qla2xxx > module loaded. > > I have done this set of operations several times including removing all > configs from /etc/target and /root/.targetcli with a reboot, just to > ensure that no cruft is lying about. > > Suggestions? On the target, try enabling extended logging with qla2xxx.ko: modprobe qla2xxx ql2xextended_error_logging=0xffffff And enable pr_debug() statements in tcm_qla2xxx (note: debugfs needs to be mounted) with: echo 'module tcm_qla2xxx +p' > /debug/dynamic_debug/control Also, it would be helpful to run the initiator with ql2xextended_error_logging enabled as well. > The another test I can run would be build rtsadmin and try it... Yet > another would be to use two cables. Should I? > That's not necessary, as the configuration side looks fine. --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
