[SPDK] Github issue #407 and libibverbs events

[SPDK] Github issue #407 and libibverbs events

[Szwed, Maciej]

[-- Attachment #1: Type: text/plain, Size: 480 bytes --]

Hi guys,
I was investigating github issue #407 and found out that we don't manage events in NVMf host side (perf in this case). We do that for NVMf target -> spdk_nvmf_process_cm/ib_event(). We should probably implement new poller that will handle incoming events in host side and change code accordingly. What do you think?

Here is documentation for reference:
https://pdfs.semanticscholar.org/9584/17ad2d476c0a4821925cf8474fabcb1229ce.pdf
Chapter 3.7

Regards,
Maciek

Re: [SPDK] Github issue #407 and libibverbs events

[Walker, Benjamin]

[-- Attachment #1: Type: text/plain, Size: 852 bytes --]

On Mon, 2018-09-10 at 08:54 +0100, Szwed, Maciej wrote:
> Hi guys,
> I was investigating github issue #407 and found out that we don’t manage
> events in NVMf host side (perf in this case). We do that for NVMf target ->
> spdk_nvmf_process_cm/ib_event(). We should probably implement new poller that
> will handle incoming events in host side and change code accordingly. What do
> you think?
>  
> Here is documentation for reference:
> https://pdfs.semanticscholar.org/9584/17ad2d476c0a4821925cf8474fabcb1229ce.pdf
> Chapter 3.7
>  

Yes, we should be doing this. It is theoretically possible to transparently
recover from connection errors if we were to handle them on both the target and
the initiator side. Now that the target logic is in place, it's a great time to
go improve the initiator behavior.

> Regards,
> Maciek