Frank1940

Yes, let it ride and see if it completes successfully. By the way, the situation you described is not unusual. By any chance, did someone tie and dress all of the SATA (data and power) up so that they would 'look pretty'? This is usually a recipe for disaster as moving one cable the slightest amount will often loosen one or more connectors. If I get inside my server case to do anything, the last thing I do before closing up the case to check each SATA connector to make sure it is securely pushed in tight-- working from inside out. And I have quick-change drive enclosures so that the case does not need to be opened for drive changes.

First, try using the Windows Explorer to as the destination window that you drag-and-drop the file into: If that does not work. MS is trying to add more and more security to SMB usage in Windows. Here is a thread to read. Start on page three as it contains the most recent stuff that MS is doing to the default settings. https://forums.unraid.net/topic/53172-windows-issues-with-unraid/ Do check to see if Your computer has updated on April 9th or later. (The April updates are a real disaster! They should be fixed with re-released updates sometime later this month...) If it was, see if you roll them back and see if that fixes things.

Have you found this one: https://windowsreport.com/network-error-0x8007003b-fix/ (Found with Google and 'Error 0x8007003B: An unexpected network error occurred' as the target.) As much as you may not like it, it appears that MS is 'owning' this one. I can tell you that my WIN10 Pro-64Bit (version 1809) is not having this problem. (If you don't know what version is currently running on your Windows computer, type winver in the search box and click on the result.) By the way, when asking about Win10 issues, always list the version number, variety --Home, pro, etc-- and whether 32-or-64 Bit. All of these have minor differences that make them unique! EDIT: Are you using a Mapped drive or working from the Network location in Windows Explorer?

Since no one else has jumped in, I will try to provide you with a bit of direction. As a point of disclosure, I do not use Plex on either of my servers at this time. Since you are using an unassigned drive to store Plex data, you will have to make changes to your Plex configuration to use this drive. If you have questions about how to do this, there should be a support thread for the Plex Docker that you are using. Asking your question there will probably get a quicker answer. (One issue that I have seen is that sometimes PLEX users will config it in a way that writes certain of its files to UnRAID's RAM disk. This will result in problems eventually!!!) As I understand it, reading is not an longevity issue with SSD's. It is writing that causes 'wear'. Even the effect of this can be minimized if you regularly run TRIM on the drive. Not all SSD's are not equal. There are difference in the memory architecture used and these have different failure modes. Next, the firmware on all SSD's will use wear-leveling by the use of all available cells. (This basically means, it does NOT use the same cells over and over again --when write-then-erase operations occur--- but spreads the writes over all of the available cells.) Realize that the time required to change from reading (or writing) one sector on a SSD to another other is always the same regardless of it 'location'. So having data on contiguous sectors is a not a condition for rapid reading or writing of data.

You might also want to check out these threads: https://forums.unraid.net/topic/77442-cannot-connect-to-unraid-shares-from-windows-10/ and https://forums.unraid.net/topic/53172-windows-issues-with-unraid/ With the latter, start on the third page as those are the most recent solutions.

When you get a drive that won't mount, set it aside (With a post-it note on it) and go onto the next one. If you don't get a second drive that fails to mount, you have the parity drive. If you get more than one drive, that is the time to rethink the issue.

First thing, that I saw was that you cache drive (sdf) is throwing up a ton of errors in your syslog. The first error that I found for this drive was on April 1st. There is no SMART report for it in the Diagnostics file. I noticed that it was formatted BTRFS. I do not use BTRFS but there are some folks who do. I hope that they will chime in with some advice.

Not a problem. If you want to give back a bit, you made the comment: Elaborate a bit more on this. You will probably not be the last one to have this problem. Be sure to indicate your MB (manufacturer and model) and where in the BIOS menu you did the changes and what changes you made. (I recognize every BIOS is a bit different but they are still similar enough that knowing where you had to go will give some valuable clues to the next person with the same problem.)

By 'vetting', I mean a google search asking if it is 'safe'. I did this when I downloaded it a couple of years ago. (It was safe at that time.) I would suspect that it still is but you should be the one to decide whether you want to use it. Basically, I would install it on a third computer (like a Windows or a Mac) and verify the speeds for each server. That way, you would be testing each server for its maximum transfer speed. (My tests, at the time, seem to indicate this utility does that very well!) If the speed to-and-from each server is much faster than you are seeing using Krusader, you have narrowed the problem considerably... By the way, I believe that a lot of folks have used rsync to do this type of transfer. It is a command-line program with a laundry list of options. (Type rsync in the terminal widow of the GUI to see them.) I have used a couple of times but I had a lot of hand-holding to help me through it.

Under light loads and simple configurations (no VLAN's, etc.), all of the unmanaged switches have a very similar performance level. There are differences between them as far as reliability, stability and expected longevity goes, but you can pretty much figure that out from reading the user reviews on Amazon and Newegg. Your Trendnet s80g switch model has been around for more than five years and is now in its second (at least) hardware version. These basic switches are using a technology that is a very mature so don't expect any exciting developments in home-market unmanaged switches in the foreseeable future.

The speeds at both ends have to be approximately the same. So it will always to the speed of the bottleneck. Again, what are the size of the files? File management overhead is an issue with rotating hard drives. Plus, I am not sure how 'efficient' Krusader is in handling external transfers. (I assume that you are using the Docker version of Krusader.) Have you checked the speed of transfer to your backup server using another computer. There is also Lan Speed test utility that I have used in the past. I would suggest that you 'vet' it before downloading. You can find it here: https://old.totusoft.com/lanspeed1.html You do have to use a fairly large file size to get around the RAM caching that Unraid will use if memory is available.

This is probably close to normal for the default setup. (This is because parity is updated as a part of the array write operation which results in a series of disk IO operations.) You can get a increase in write speeds if you change the "Tunable (md_write_method):" on the Settings >> Disk Settings tab to 'Reconstruct write'. However, this will spin up all of the array disks for all write operations. (The default method will spin up only the parity disk and the disk on which the data is stored.) In any case, the write speed will also be slower for large numbers of small files rather than a few very large files because of the file management overhead.

Hopefully, you realize that the Wireless connection will be the limiting factor on the transfer speed. Remember that the all of the devices that are using that radio will be sharing the same bandwidth. Devices using the 5GHz radio have the potential of higher throughput but there are many factors that can negate that. You may not realize it but a wired switch can have several pairs of computers running transfers at 1Gb in both directions simultaneously! (Your TrendNet switch can handle 4 pairs of 1Gb simultaneous transfers.) Wireless is a poor second choice anytime throughput is a concern.

Was the 'Laptop' on a wired connection or wireless connection? Are you using a cache drive on your server? There is also Lan Speed test utility that I have used in the past. I would suggest that you 'vet' it before downloading. You can find it here: https://old.totusoft.com/lanspeed1.html You do have to use a fairly large file size to get around the RAM caching that Unraid will use if memory is available.

You can turn the reporting of that Attribute off for just the cache drive under the Device properties for that drive only ( Main tab, click on the drive you need to modify under the Device column)