finding out about QLogic features

finding out about QLogic features

[Thomas Meller]

My colleagues are going to hang me if my system produces further path failovers.

I use a QLogic 2312-based HBA (FastT FC2-133) in failover mode on a LinuX HA-cluster.

Everything works fine, but the driver causes several failovers per day without an urgent
reason.

We're using a FastT 700 via McData fabric switches.

Driver version 6.06-fo, Firmware 1.25, Kernel 2.4.18.

Maybe I need to tune the driver's parameters to extend some timeouts. Maybe it's as simple
as an upgrade.
I found out that a FastT requires more retries than are set by default.
QLogic support has given me some hints what to cause the scsi subsystem to do.

In short: I have to set a bigger retry count when a device proclaims a not-ready condition.

>From my experience it is best to understand first how it works before setting weird
parameters.
Especially due to the use within a failover cluster it could cause timing problems.

Are there any docs around to make a non-programmer understand the HBA-sd_mod interaction
technique and it's implications?

Is this the right mailing list for me?

Thanks in advance

-- 
Thomas Meller
mailto: thomas.meller@t-systems.ch
mailto: thomas.meller@gmx.net
----
...Our continuing mission: To seek out knowledge of C, to explore strange
UNIX commands, and to boldly code where no man page has ever been seen