On 3/5/21 9:31 PM, Sagi Grimberg wrote:

quoted

Compare that to the 'standard', non-CMIC nvme, where with the same
setup MD would detach the nvme on its own:

# cat /proc/mdstat
Personalities : [raid10]
md127 : active (auto-read-only) raid10 nvme2n1[1]
       4189184 blocks super 1.2 2 near-copies [2/1] [_U]
       bitmap: 0/1 pages [0KB], 65536KB chunk

unused devices: <none>
# nvme list
Node             SN                   Model       Namespace
Usage                      Format           FW Rev
---------------- --------------------
---------------------------------------- ---------
-------------------------- ---------------- --------
/dev/nvme0n1     SLESNVME1            QEMU NVMe Ctrl       1         
17.18  GB /  17.18  GB    512   B +  0 B   1.0
/dev/nvme2n1     SLESNVME3            QEMU NVMe Ctrl       1          
4.29  GB /   4.29  GB    512   B +  0 B   1.0

And yes, this is exactly the same setup, the only difference being the
CMIC setting for the NVMe device.

I agree with Christoph that we should do exactly the same for all.

Hannes, My understanding here is that you want the device to go away
after the last path disappeared because it breaks md, why don't you
want to have this also for fabrics?

Oh, I would _love_ to have it for fabrics per default.
If you agree with it I can resend Keiths original patch, which solves
the issue without the need for any additional settings.

You mention that in fabrics one can manually disconnect, but why should
the user resort to a manual disconnect?

Is something else broken with this behavior? Maybe I am missing
something here?

I'm just trying to understand why the current logic was build into
fabrics in the first place.
But if no-one remembers (or there was none) I'd happily convert to the
PCI logic.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		           Kernel Storage Architect
hare@suse.de			                  +49 911 74053 688
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), GF: Felix Imendörffer

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme