Any ideas what could be wrong here? It's been running fine for quite a while...

truffle-diagnostics-20221128-1636.zip

Nothing obvious, has it changed or still remains the same?

6 hours ago, JorgeB said:

Nothing obvious, has it changed or still remains the same?

It used to be around 170-200MB/s until very recently.

 

I run a Plex server and Plex was also running terribly when the Parity was running, which is also a new problem.

I meant is it still going at 20MB/s? It could have been a slower zone on one of the disks.

I was on vacation, so I didn't stop it. Yes, it had been running for 4 days (and needed 4-8 more, I don't remember). And when I start it now, I get the same speed.

Run the diskpeed controller benchmark, to rule out any bottlenecks there.

Everything there looks good. The only thing it seems I can't test is an external USB 3.2 enclosure with a couple of drives in it. While that might sound crazy, I tested it extensively when adding it about 6 months ago and it's seemingly working fine. Unless maybe there were some USB changes in one of the latest point releases?

 

 

41 minutes ago, flyize said:

The only thing it seems I can't test is an external USB 3.2 enclosure with a couple of drives in it.

That would be the main suspect, test read speed over SMB for those drives.

I'll be damned, you're right. Is there anything you can think of that would have changed with USB in the last month or so? I just did a quick CTRL+F for USB on the latest 6.11 releases, and I don't see anything obvious. I'm quite positive this worked without issue before.

Possibly some quirks added to the kernel for the USB bridge:

 

usb-storage 1-5.3.3:1.0: Quirks match for vid 174c pid 55aa: 400000

 

It might be downgrading the speed to USB 2.0, we really don't recommend USB for array/pool devices.

I'll be damned. That's it.

 

 

Anything I can do? When was this added? I'd like to roll back and test (unless you're sure that's the issue).

 

While I doubt it matters, this is the one I purchased.

https://www.amazon.com/gp/product/B07Y3WDHLD

3 minutes ago, flyize said:

When was this added?

I don't know, but if it was added likely there's a good reason, could be for example due to data corruption or other serious problem.

Would it be possible for me to find out more information from somewhere? I'd really rather not abandon this thing unless I have to...

Probably there's a comment in the kernel to why the quirk was added, but sorry, no idea how to find that.

Yeah, shouldn't there be a repo somewhere for us to check? It's GPL'd code...

Anyone else here know where I can find out more information about this change? Thanks!

@limetech any chance you might be able to share some quick wisdom here?

Edited December 1, 20223 yr by flyize

Anyone have any idea or am I just dead in the water with this USB enclosure?

Not even close to an expert at this.  But a quick Google search of "ASM1153 SATA 3Gb/s bridge" finds many hits going back ro 2017 and beyond, with other ASM chips at 6Gb/s.  Makes me think nothing may have changed in the base kernel/drivers of Linux in general.

On 12/4/2022 at 11:32 AM, ConnerVT said:

Not even close to an expert at this.  But a quick Google search of "ASM1153 SATA 3Gb/s bridge" finds many hits going back ro 2017 and beyond, with other ASM chips at 6Gb/s.  Makes me think nothing may have changed in the base kernel/drivers of Linux in general.

Thanks! 

 

Its definitely something that changed in the last couple releases of 6.11. It worked flawlessly before that.

I managed to 3d print an extra drive cage to go into my Unraid server. Sucks, but I guess I'll just sell this enclosure.