Questions[hba](server)

Last week I updated my kernel using the regular DNF update process. On reboot, none of the drives in my mdadm arrays were detected. I narrowed down the issue to a problem with the kernel and used an older kernel to boot. I waited a week for the next kernel update and ran the process again, hoping it was just a bad kernel, but the problem has not been resolved. Dmesg shows kernel: mpt2sas_cm0: failure at drivers/scsi/mpt3sas/mpt3sas_scsih.c:10790/_scsih_probe()! on both 5.8 kernels. After some research, I'm pretty sure that Fedora has dropped support for sas2116 controllers (https://access.redhat.com/discussions/3722151?page=2), so I either need to find drivers for this card or replace it with a 12gb/s card.

What do you suggest? Where can I find drivers for this card for Fedora 32 and how do I install them, or what would be a good 12gb/s card to replace it with (the card currently drives a 24 bay supermicro backplane with 12 drives installed)? The card I currently use is a LSI 9201-16e.

Is it possible to connect Point to Point two Hyper-V servers using FC HBA to replicate data without connection to external storage?

I'm trying to figure out if there are any differences between the "R" vs. the "S" of this SAS controller. As far as I can tell, the "S" is a Sun Microsystems re-branded version. I can't find much solid documentation on the "S" version.

Most documentation that I find is on Oracle's documentation website, but mostly it's about setting up RAID on the HBA, not the specs themselves. I know Intel and (apparently) Sun re-brand the controller as their own but I'm curious to know if there are any minor/major differences that Sun may have made.

Thanks!

I'm trying to understand some SAN infrastructure and I was hoping some of you with more experience than me could help me understand scaling with a SAN.

Imagine that you have some computer servers that have a HBA. They connect either directly or via a switch to a SAN Controller. The SAN Controller then provides one or more LUNs which are mapped to most likely a RAID array on a storage device.

So if I understand correctly, the "controller" represents a performance bottleneck. If you need lots of performance then you add more controllers with connections to their own storage, which then get mapped to the servers that need them.

I imagine you can get some very high performance controllers with huge storage capacities and lower performance controllers with a lower maximum performance? But if you have a switch you can then add several lower performance controllers to your network as you need them?

Please tear apart my understanding if I have it wrong, but I'm trying to work out how you connect HBAs from a server to storage without the fabric simply representing "magic".

I've been given two hpc cluster nodes, each with an infiniband hca.

Is it possible to connect them back-to-back, without using a switch ?