On Fri, 11 Aug 2000, David Teigland wrote:
> Thanks to all for the input. I should have provided some more background
> information. I work on the GFS project and we primarily use Fibre Channel. I
> know SCA parallel SCSI drives are the way to go, but it still sounds like a
> touchy issue. I've seen my share of scsi mid-layer errors which lock up
> the machine, so I wanted to try and get a clearer picture of things.
Okay, I knew that sistina looked familiar. :)
> - Hot-swapping SCA disks on the bus should be relatively reliable if it's done
> with care. It sounds like if any transfer is happening during a swap you're
> in serious danger of crashing everthing. The scsi drivers can be prompted to
> add or remove devices. I wonder if multiple hosts put a wrench in things
> here.
Well I would certainly say that multiple hosts are not going to make
hot-swapping easier. I've not tried this, so I guessing at this. I can
say for sure, you need to make sure all hosts aren't doing any
transfers. Then when you add the device you can't have the driver sending
a bus reset, if the other host is going to reset in turn.
> - The other important issue is hosts which crash at any time, including during
> a transfer. It sounds like the drivers on other machines will currently
> start a reset-war, but the drivers could be improved to avoid this and
> hopefully keep using the devices as they were.
The are not buts about it. The drivers need to be able to detect and deal
with reset-wars, or there is no way to have two hosts on the same bus.
> - A similar problem for devices which crash abruptly.
> - How about adding machines to the bus and then booting them up?
Some (most?) SCSI controllers issue a reset on boot.
> By the look of things here, it is not reasonable to use GFS with multiple hosts
> on a shared SCSI bus if you're interested in HA. If any machine or disk
> crashes, all your devices are probably in trouble. Stopping all machines' I/O
> (and maybe unmounting everyone) to add or remove storage would also be
> prohibitive.
Okay 1 true bus (i.e. one cable winding between all devices) may not work,
but what if you had something in the middle performing arbitration. A
device that could filter out bus resets, and could isolate each host from
eachother and from the drives for that matter. This device could also
totally change the view of the drives to the host computers to perform any
sort of partitioning or LVM that you could want. And since this thing
speaks SCSI you could use out-of-band SCSI data sent to the SCSI-generic
device to manage it.
Hey, this magical device sounds awful like this 3 channel RAID controller
I have sitting behind me. (And for your GFS people it does support
DLOCKs.)
-Chris
--
Two penguins were walking on an iceburg. The first one said to the
second, "you look like you are wearing a tuxedo." The second one said,
"I might be..."
--David Lynch, Twin Peaks
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
