https://bugzilla.novell.com/show_bug.cgi?id=289228 Summary: QLogic HBA Timout Product: SUSE LINUX 10.0 Version: Final Platform: x86-64 OS/Version: SLES 10 Status: NEW Severity: Normal Priority: P5 - None Component: Basesystem AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: matthias.rudolph@hds.com QAContact: qa@suse.de CC: matthias.rudolph@hds.com Found By: System Test I am an employee of Hitachi Data Systems. Our customers are using SLES 10 with QLogic HBAs with the SLES built-in driver and DM Multipath for path failover. In all operating systems we recommend to use a HBA timout of 30-60 seconds. But the Linux QLogic driver does not provide such a paramter. I know in SLES 10 there is a possebility to set a timeout in /sys/class: # cat /sys/class/fc_remote_ports/rport-1:0-0/dev_loss_tmo 35 # cat /sys/class/fc_remote_ports/rport-2:0-0/dev_loss_tmo 35 So it seems the timeout is set to 35 seconds. But all SAN errors/delays were immediately reported as an IO error: Jul 3 14:14:12 --> FC cable unplugged Jul 3 14:14:13 D-SRV-L-02 kernel: qla2xxx 0000:03:0b.0: LOOP DOWN detected (2). Jul 3 14:14:13 D-SRV-L-02 multipathd: 66:240: mark as failed Jul 3 14:14:13 D-SRV-L-02 multipathd: 1HITACHI_D60H02610000: remaining active paths: 1 Jul 3 14:14:13 D-SRV-L-02 kernel: sd 2:0:1:7: SCSI error: return code = 0x20000 Jul 3 14:14:13 D-SRV-L-02 kernel: end_request: I/O error, dev sdav, sector 28027520 And here my question: Why is the timeout not active? Regards Matthias Rudolph HDS Germany -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.