[SPDK] Re: NVMe Multipath subsystem controller IDs
Jonathan Teh <jonathan.teh at mayadata.io> Fri, 09 Apr 2021 15:43:51 +0100
| Newsgroups | dev.linux.lists.spdk |
|---|---|
| Message-ID | <[email protected]> |
On 08/04/2021 23:09, 松本周平 wrote: > Hi Jonathan, Ben, Hi Matsumoto-san, > I'm not an expert in this area but supporting multiple SPDK target > instances is also important to iSCSI. > I see the value of per-subsystem controller ID range and it is already > added to Linux target. > > But we have little auto test for multiple SPDK target instances. > Could you enhance test/nvmf/target/multitarget.sh to verify the feature? To verify the controller ID range has changed, an initiator needs to connect to the controller and a specified subsystem. For that, I added a test to test/nvmf/target/nvme_cli.sh that extracts the controller ID from the output of "nvme id-ctrl". > And do we have negative side effect if we support per-subsystem controller > ID range? Not really AFAIK, the more limited controller ID range means a lower limit on the number of simultaneous connections to an individual controller for a subsystem but there's typically only one connection anyway.