smeehrrr

For what it's worth, running ethtool -K eth0 tso off made the problem go away for me. I don't know what I'm giving up in terms for performance but it was a good enough temporary workaround to get all my files copied over. I did not have any of your additional issues with terminal, etc, just a momentary hitch that was long enough to interrupt file copies but not long enough to make ssh drop.

I started hitting this today after upgrading to 6.10.3. Attaching diagnostics just in case that's helpful. nova-diagnostics-20220809-1753.zip

I've been running them since 2014 with no problems, on up to date versions of Unraid. What kinds of issues are people seeing? What are the models of the recommended LSI cards?

I have a couple of 20-drive Norco-case servers in my rack, one of which is super limited in spec and the other of which is beefier and driving a bunch of VMs in addition to serving files. I've recently noticed that the VM server is overloaded and it's causing media streaming from it to glitch, so I'd like to upgrade the lesser server to something more capable and offload some of the VMs from one machine to the other. I'd like to salvage some of the hardware in the process, and that's where I'm getting confused. The server to be upgraded is currently running an ASUS P5BV-M motherboard and has two Supermicro AOC-SAS2LP-MV8 SAS cards in it on PCI-Ex8 slots. I'd like to dump that motherboard and replace it with something that can hold a virtualization capable processor and 64GB of RAM, but for the life of me I cannot figure out which motherboard is going to have the correct combination of PCI-E slots to let those cards run at full efficiency. Anyone have a mobo recommendation that won't break the bank?

This doesn't seem to work for me anymore, and I suspect it's related to the CSRF work. Has anyone found a recent way to spin up all drives from a URL or can anyone confirm that this old way is still working for them?

In my experience, if you have the bugged firmware then Unraid will not boot after you add the jumper.

If anyone gets wake-on-ARP working, would you please post your config? This is basically the only feature that I still need out of my Unraid and I'm willing to replace my hardware to get it.

Anyone have a list of NICs that support wake on ARP with Unraid? I desperately want this functionality. I thought wake on unicast was what I needed but that doesn't work.

Looks like my best bet is to RMA the drive that doesn't work with the jumper attached and make sure everything is jumpered in the future. Here's a question that I haven't been able to figure out from the various threads on this subject: Assuming that Unraid eventually supports these drives natively (sans jumper), will there be any performance penalty for having the jumper on there? Web sources suggest all that it's doing is adding one to the sector offsets, so wouldn't think so, but maybe someone has a definitive answer.

Found some jumpers and gave it a try. I have two WD20EARS in the server. Both of them apparently worked without the jumpers. With the jumpers, one of them works and the other one goes into the same error loop that captain_video saw above. Took the jumper off that one and it works again. Just weird.

My parity drive is, unfortunately, also a WD20EARS. Looks like I need to find some jumpers and get to work.

And the performance gap is due to a software deficiency in Slackware?

Hm, I installed a couple of WD20EARS to my Unraid before realizing that this thread existed. I didn't add any jumpers, and everything appears to be working fine. Are there some problems I should be looking for going forward?

Sorry, I forgot to put the sarcasm tags in there.

That's actually the manual for a different motherboard. Asus has cleverly named two different boards "P5B-VM" and "P5BV-M". The former is an older desktop board, the latter is the server mobo we're talking about here.