Phibster

So testing Beta 7.0 on my second machine and I see some improvement but still approximately 7-8 seconds (down from about 15) but still not amazing so I will wait for the proper release before I upgrade my main server.

Thank you, will test it out on my secondary server first.

The folder is simply a list of folders (each folder has a file underneath it) but no files in that folder. I absolutely hate the icon views and use detailed view only. Even changing to the 'list' view still has the same slow response so it's not the extra details being read. It is possible windows is still being silly and trying to 'peek' into folders as well which would slow it down. Oh that would be good to see. I am also eagerly awaiting the move to kernel 6.2 which supports the intel ARC GPU which I have one sitting waiting so I can do some transcoding.

I have been persevering for a while with this one. I have a folder which has 18,000 folders underneath it. When I browse to this directory it takes windows 11 about 7 seconds (on server 1) to 15 seconds (on server 2) to go though and list all the folders, even if I was just in there. I have tried the Dynamix FolderCache plugin which did not help at all. Wondering if there is something in Windows or the SMB config that I can adjust to make this better. Server 1 which takes approximately 7 seconds to list the folder: Motherboard: Asus Pro WS WRX80E-SAGE SE WIFI CPU: Threadripper Pro 3975WX RAM: 128GB DDR4 ECC Drives: 10 x Seagate ST20000NM0002D, 8 data 2 parity Lan connection: 10gb Server 2 which takes approximately 15 seconds to list the folder: Motherboard: Asus Pro WS WRX80E-SAGE SE WIFI CPU: Threadripper Pro 3975WX RAM: 128GB DDR4 ECC Drives: 12 x Seagate ST16000NE000, 10 data 2 parity Lan connection: 10gb I have multiple Windows machines, one with 10gb LAN and the second with 2.5gb. Both are Windows 11 but were previously Windows 10 and exhibited the same behavior. No changes have been made to the server or clients, all stock settings. The time difference between the two servers leads me to believe it could potentially be a FUSE filesystem bottleneck. There is no noticeable CPU increase while the directory is being listed. Any help appreciated and if I have missed some critical information please ask :)

Just to add my 2 cents and thank everyone. I upgraded from SATA to SAS drives and was trying to work out how none of them showed up. I had a similar cable with the SATA power on the back of it. I fixed it by removing the 3.3V pin from the modular power supply plug (looked at molex vs sata connector. I just taped it with some electrical tape so that it couldn't short anything and hey presto! All drives working!! So thank you to everyone here.

I have 2 unraid servers with all of my main array data backed up between them. I want to replace all of the drives in the main server with brand new ones and I feel that the new config method is the easiest way to do this rather than multiple rebuilds. My questions are around the other server settings and data not held on the main array. Does new config: Erase things such as ethernet, server name etc? Mess with any docker images I have (all docker images/config reside on a separate SSD)? Is this just a case of assigning the drives back as they were and then re-creating the docker from existing images? I have searched and could not find any definitive answers.

Thank you, pcie_aspm=off solved it for me. Rampage V Extreme motherboard.