On Wed, May 26, 2021 at 7:44 PM Matteo Croce [off-list ref] wrote:

On Wed, May 19, 2021 at 4:33 PM Matteo Croce [off-list ref] wrote:

quoted

On Tue, May 18, 2021 at 9:52 AM Prasanna Kalever [off-list ref] wrote:

quoted

quoted

BTW, loop has similar issue, and patch of 'block: add a sequence number to disks'
is added for addressing this issue, what do you think of that generic
approach wrt. this nbd's issue? such as used the exposed sysfs sequence number
for addressing this issue?

https://lore.kernel.org/linux-block/YH81n34d2G3C4Re+@gardel-login/#r (local)

If I understand the changes and the background of the fix correctly, I
think with that fix author is trying to monotonically increase the seq
number and add it to the disk on every single device map/attach and
expose it through the sysfs, which will help the userspace processes
further to correlate events for particular and specific devices that
reuse the same loop device.

Yes, but nothing prevents to use diskseq in nbd, and increase it on reconfigure.
That would allow to detect if the device has been reconfigured.

Regards,
--
per aspera ad upstream

FYI: I just sent a v2 of the diskseq change

https://lore.kernel.org/linux-block/20210520135622.44625-1-mcroce@linux.microsoft.com/ (local)

Thanks, Matteo, I will take a look.

Just to set the expectation here, I don't have any thoughts on
leverage the diskseq number for nbd as part of this patch. This patch
is trying to solve a different problem which is more severe for us
than helping to identify the reconfigured events.

That all said, once diskseq patches are merged, I will surely open a
new patch with the required changes in nbd, to leverage diskseq
number.

Best Regards,
--
Prasanna

--
per aspera ad upstream