On Fri, 28 Apr 2000, Kevin Quick wrote:
> On Fri, 28 Apr 2000, Eduardo Horvath wrote:
>
> : First of all Node or Port WWNs are not sufficient for this purpose. Let's
> : say you have a RAID box with two controllers. Each controllser has its
> : own WWN: WWN0 and WWN1. One of the controllers fails and needs to be
> : replaced. The new controller has a different WWN: WWN2. But it turns out
> : that the controller really wasn't bad, it just had a loose connection. So
> : it's used when a controller fails on another RAID box on the same
> : SAN. Now the original box has WWN1 and WWN2, but another box has WWN0 and
> : WWN3. The volumes are still in the original box, but now you have a new,
> : completely different set of volumes that magically appear attached to
> : WWN0.
> :
> : No, the only reliable way to identify a platter is through the VIPD page.
>
>
> This is not quite correct. A Fibre Channel PLOGI will identify both the
> Port's WWN (P_WWN) and the Node's WWN (N_WWN) on whose behalf the port is
> logging in. The Login acceptance carries the same information for the
> other side of the connection.
>
> In your scenario, WWN0 is P_WWN0 and should be recognized as having been
> moved from N_WWN0 to N_WWN1, thereby creating a different path.
No. NWWN is encoded in the controller. Most FC RAID boxes have dual
controllers, thus two different NWWNs. In addition to that each
controller can have multiple ports, each with the same NWWN but multiple
PWWNs. I presupposed dual controllers each with one port, so PWWN and
NWWN would be equivalent. I suppose I should have been a bit more clear
on that point. Few people seem to grasp the subtlety of the brokenness of
the specification in this area.
Here are some fun brain teasers:
Suppose you have a SCSI RAID box attached to a properly designed
Fibre-channel/SCSI bridge. How do you probe it? (Hint: REPORT_LUNS.)
In the previous configuration a LUN is destroyed. Then another
LUN is created to replace it. What happens? (Hint: REPORT_LUNS_HAS_CHANGED.)
Suppose you have two properly designed Fibre-channel/SCSI
bridges. A disk is plugged in to the SCSI bus on one bridge. The host
formats it. It is then removed and attached to the other bridge. What
happens? (Hint: VPID.)
I'm glad I don't have to worry about this any more.
Eduardo Horvath
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 30 2000 - 21:00:16 EST