Scott Peterson

In NVMe, if the host ever has two writes to the same LBA in flight simultaneously, the behavior is undefined. NVMe does not guarantee commands complete in order. The only...

> The idea here is based on an assumption that when the host decides to failover, it will first destroy (i.e. disconnect) from the controller, or that the controller will...

> > The retried write(lbaX, data0) starts at GW2 at some time after the original write(lbaX, data0) started on GW1. Normally they should take about the same time to complete,...

> Think it is possible to distinguish "regular" host's Disconnect from disconnect caused by IO delayed in librbd. > > 1. Add new "io_start_tick" parameter to the struct spdk_nvmf_request. Initialize...

There's an NVMe technical proposal in progress that seems to address most or all of the gaps in the KATO mechanism noted here. I'll send the details by email to...

> Another option @caroav suggested is having a ceph context per initiator, this will solve the resources concern and allow us simply to abort all pending I/Os per initiator. This...

See https://github.com/ceph/ceph-nvmeof/issues/63#issuecomment-1421280054

Stats from the SPDK NVMe-oF targets seem to be covered by #37, but that might not include the discovery services, whatever glue logic is required to enable all these Prometheus...

I found [this kernel patch](https://lore.kernel.org/linux-nvme/[email protected]/T/) from early 2021 that adds a new "nvme monitor" command to do just what we want here (monitor discovery controllers for changes, and promptly act...

Complete tests are a feature, right?