[Support] binhex - Krusader


Recommended Posts

2 minutes ago, JQNE said:

I have always used only search option to locate krusader.

Then there were probably 2 results, and assuming that you're running the latest CA with the new GUI, then you're the first to find a minor oversite - on the main display's it isn't showing that one of those 2 results is "private" (ie: the install from dockerHub).  *Alternatively, and I'd have to test is that it only displayed one, but because of the identical naming and repository it displayed the private one.  In which case you've found a bug in CA :( 

 

Go to statistics, click on Private Templates and hit the red-x on anything you're not actually using (esp. krusader) and try the whole thing again.

Link to comment
1 hour ago, JQNE said:

Why not? Something wrong?

It isn't necessary to share disks. If for some reason you want to control which disks get used you can do this with User Share settings.

 

Sharing disks has a number of ways it can get you into trouble, including data loss.

 

I always recommend just working with User Shares, at least until you get a good idea of how User Shares actually work "under the hood" so you will know how to avoid the problems if you ever decide you do need to access the disks directly for some reason.

 

I never put the disks on the network and I know how this stuff all works. I always just work with the disks directly on the server if the need arises.

Link to comment
36 minutes ago, trurl said:

It isn't necessary to share disks. If for some reason you want to control which disks get used you can do this with User Share settings.

 

Sharing disks has a number of ways it can get you into trouble, including data loss.

 

I always recommend just working with User Shares, at least until you get a good idea of how User Shares actually work "under the hood" so you will know how to avoid the problems if you ever decide you do need to access the disks directly for some reason.

 

I never put the disks on the network and I know how this stuff all works. I always just work with the disks directly on the server if the need arises.

Thanks for info! ✌️ I once tested is there any usage difference when using share or straight harddisk on win vm. So I left disk share on. I also have had no luck to get krusader work, and I have been moving stuf inside server with root share acces on Osx Finder... And also tested MC.. But still had little slow speeds. I have tested various versions of krusader, but still no luck with gui.

Link to comment
10 minutes ago, JQNE said:

Thanks for info! ✌️ I once tested is there any usage difference when using share or straight harddisk on win vm. So I left disk share on. I also have had no luck to get krusader work, and I have been moving stuf inside server with root share acces on Osx Finder... And also tested MC.. But still had little slow speeds. I have tested various versions of krusader, but still no luck with gui.

Writes to the parity array will be slower than the normal write speed of the disk being written, because parity must be updated at the same time. This actually involves reading the data and parity disk, calculating the parity change the new data will cause, and writing the data and parity disk. This is how it works whether you are working directly with a data disk, or whether you are working with a user share.

 

But you actually bypass some features of Unraid if you work directly with the disks.

 

User Shares can span disks. This allows you to have a folder with some of its files on one disk, and other files on another disk.

 

And User Shares can be cached. This allows new files to be written to cache and later moved to the parity array. The speed of writing to cache isn't impacted by the parity update, and people often use SSDs as cache so caching user shares can be a way to write faster.

 

Is there something specific you want to do with Krusader? It isn't going to help with the speed problems since they are just the way parity works. I always just use Midnight Commander since it is built-in and I have been using it in Unraid since long before dockers were introduced.

 

There is another method of doing parity writes that is somewhat faster. This thread explains the 2 different methods of writing parity:

 

 

Link to comment
18 hours ago, trurl said:

Writes to the parity array will be slower than the normal write speed of the disk being written, because parity must be updated at the same time. This actually involves reading the data and parity disk, calculating the parity change the new data will cause, and writing the data and parity disk. This is how it works whether you are working directly with a data disk, or whether you are working with a user share.

 

But you actually bypass some features of Unraid if you work directly with the disks.

 

User Shares can span disks. This allows you to have a folder with some of its files on one disk, and other files on another disk.

 

And User Shares can be cached. This allows new files to be written to cache and later moved to the parity array. The speed of writing to cache isn't impacted by the parity update, and people often use SSDs as cache so caching user shares can be a way to write faster.

 

Is there something specific you want to do with Krusader? It isn't going to help with the speed problems since they are just the way parity works. I always just use Midnight Commander since it is built-in and I have been using it in Unraid since long before dockers were introduced.

 

There is another method of doing parity writes that is somewhat faster. This thread explains the 2 different methods of writing parity:

 

 

Nothing really different than MC can do. I thought krusader would be easier and faster to use. But MC is looking verta string for me because it works.

 

I have 3 drives, and only one of them is filling up. I have high Water choosed across all shares.  

 

I been thinkin to use turbo write but it has its own drawbacks ✌️

Link to comment
19 minutes ago, JQNE said:

I have 3 drives, and only one of them is filling up. I have high Water choosed across all shares.

There are 3 settings for a User Share that Unraid considers when deciding which disk to write: Split Level, Minimum Free, and Allocation Method.

 

Split Level has the highest precedence. When choosing a disk to write, if Split Level says the file should go on the same disk other files are already on then that is the disk that will be chosen.

 

If a disk has less than Minimum Free remaining, then Unraid will choose a different disk (except Split Level has precedence).

 

High-water Allocation is a good compromise between using all (included) disks without constantly switching disks simply because one disk temporarily has more free space than another. It is a little complicated to explain but fortunately there are already explanations in the wiki:

 

https://wiki.unraid.net/Un-Official_UnRAID_Manual#High_Water

Link to comment

Did I select the wrong share? shouldn't it be under "user"

 

I just used binhex-krusader to copy a very large media folder and for some reason it just filled up a single drive and stopped?

 

I have copied to the "root\UNRAID\user\movies" and I have the shared folder setting to "High-Water" any way to do the copy with krusader and not filling up the individual disks? 

 

Its movies: So split level is 2

 

image.thumb.png.010877f94a399ecb942c417efdd6322b.png

 

Link to comment

I just did a unBALANCE and then restarted Krusader and startet the transfer to the share under user (Not user0) I dont have any cash on this share and I dont have any parity drives (Waiting to add them after the copying is finished)

 

But Krusader is now again filling up disc nr 1 - transferring from my old server by SMB it looks like it levels the data?

Any help is much appreciated

Link to comment

Linux sees all user shares mounted at /mnt/user. Move and Rename are the same command in Linux, mv.

 

So if you move files between user shares, it will rename them with a different path on the same disk they are on, instead of copying them to another destination disk and then deleting them from the source.

 

You will have to do the copy then delete yourself instead of a move. If you copy from source to destination, it will write a new file at the destination using the user share settings. Then you can delete from the source.

Link to comment

I think you need to set your split level higher.  This quote is from the following section in the Unraid 6 manual found here:

 

         https://wiki.unraid.net/index.php/UnRAID_Manual_6#Parameters

    

Quote

Split level defines the highest level in the share directory hierarchy which can be split among multiple disks. In the Movie share example above, setting Split level to 1 only permits any object created directly under the Movie directory to be allocated to any disk according to the Allocation method. Thus, when we create the Alien subdirectory, it may reside on any of the data disks; however, when we create a file or another directory within the Movies/Alien directory, this object is at level 2, and will be created on whatever disk the Movies/Alien directory actually resides on.

 

I realize that it is virtually 'legalese' and very difficult to understand.  (At one time, there was a Wiki entry that had a much clear description of how the split level worked but I can't seem to be able to find it any more.)  The split level assignment number has caused confusion for years.  It is intended to keep the .VOB files in the VIDEO_TS directory of a DVD on the same hard disk so that the movie/show/video does not pause should the .VOB files be split across two disks.  

Link to comment
34 minutes ago, Frank1940 said:

Here is that section  (Found it with Google):

 

      https://wiki.unraid.net/Un-Official_UnRAID_Manual#Split_level

I did read this and came to the conclusion that split level 2 was perfect (All is mkv files) so dir is in a relative flat structure

But maybe I need to count the sub dir of the shares?

 

Movies/Kids Movies/Movie name/Movie file.mkv --> Would this be split level 2 or 3?

 

No The transfers are from Mounted share inside Krusader to the shared Unraid folder, so it should move to the next drive...

 

Update it just moved to the next drive but it didnt leave any room left on the disc (I have set 50G free space on all shares?)

image.thumb.png.36e599c7823f6e9bc840403a34f50c73.png

 

Looking at the folder it looks fine all movies in each folder on the same drive - like I wanted ?

 

Link to comment
1 hour ago, trurl said:

Linux sees all user shares mounted at /mnt/user. Move and Rename are the same command in Linux, mv.

 

So if you move files between user shares, it will rename them with a different path on the same disk they are on, instead of copying them to another destination disk and then deleting them from the source.

 

You will have to do the copy then delete yourself instead of a move. If you copy from source to destination, it will write a new file at the destination using the user share settings. Then you can delete from the source.

No move - I only copy from SMB share to local Unraid share folder!

Yes the move was done with unBALANCE and that worked perfectly, and just to remove the over allocation on drive 1 - only problem is that now Krusader have filled up drive 1 again... I think it works perfectly initiating the transfer on my Synology --> Unraid so maybe just keep it flowing in one direction 

Edited by casperse
Link to comment
9 minutes ago, casperse said:

I did read this and came to the conclusion that split level 2 was perfect (All is mkv files) so dir is in a relative flat structure

But maybe I need to count the sub dir of the shares?

 

Movies/Kids Movies/Movie name/Movie file.mkv --> Would this be split level 2 or 3?

 

No The transfers are from Mounted share inside Krusader to the shared Unraid folder, so it should move to the next drive...

 

Update it just moved to the next drive but it didnt leave any room left on the disc (I have set 50G free space on all shares?)

image.thumb.png.36e599c7823f6e9bc840403a34f50c73.png

 

Looking at the folder it looks fine all movies in each folder on the same drive - like I wanted ?

 

Print out this section on paper:

 


   Capture.JPG.f913979f2dfeeb7b1f699c6c2c45ea65.JPG

 

Now write the name of each folder (not any file names, as the split level strictly deals with folders/directories) on the paper.  There is not such thing as a 'sub-dir"!  It is either a folder or a file.  I do believe at this point, you need a split level of at least three.  (By the way, I doubt if you truly need any split level if you are using .mkv files for media storage.  Each show or movie is in one (and only one) file and that means that there will be no pause in the playing of any movie or show while a second disk spins up.  You may have a slight delay while loading file names but that is usually only a few seconds.) 

Link to comment
7 minutes ago, trurl said:

Another possibility that I have seen in other situations, such as rsync, is that the application will create all the folders in advance, then copy the files to those already created folders. Since the disk has plenty of space for the empty folders, that is where they wind up.

Yes you are completely correct - it just finished with 68 empty folders... so cleaning up and starting unBALANCE up again.

Link to comment
6 minutes ago, Frank1940 said:

Print out this section on paper:

 

image.png.0b2330e74b2295ef54e06f1ed21b888d.png
   

 

Now write the name of each folder (not any file names, as the split level strictly deals with folders/directories) on the paper.  There is not such thing as a 'sub-dir"!  It is either a folder or a file.  I do believe at this point, you need a split level of at least three.  (By the way, I doubt if you truly need any split level if you are using .mkv files for media storage.  Each show or movie is in one (and only one) file and that means that there will be no pause in the playing of any movie or show while a second disk spins up.  You may have a slight delay while loading file names but that is usually only a few seconds.) 

Ok bad choice of words 😊 but you know what I mean saying "Sub-dir"

I just wanted to have one single folder with the movie in that folder and if possible poster, nfo files etc,,,,

 

And my homework was exactly what your link says:

image.png.d8060c92cefe9f79b6fe3aa89f3aab8d.png

 

Losing a drive would then mean losing, a "full" set of movies with info, making it possible to identify what was lost.

 

And for TV Shows

image.png.33d2f9124372e37ec2d2110b62e69eea.png

 

I think the level is (ok) its the way copying is done in Unix.... just have to keep using External transfers to the unRAID server

Link to comment

FIle copy crashing/freezing krusader when copying a large file ~10GB to QNAP NAS. I have a SMB mapping from KRUSADER to a share on my QNAP NAS and I have been unsuccessful copying large files from the unRAID NAS to the QNAP NAS, starts out fine and ~2GB of data copied, KRUSADER crashes and I can't restart it w/ out rebooting unRAID.

 

Is this just a limitation of KRUSADER or do I have a configuration issue? I have read just about suggestion posted and re-installed KRUSADER w/ same results.

 

Both NAS are attached w/ bonded interfaces and configured correctly on the switch.

 

image.thumb.png.ffefee7ac46c15e6f619e8f979dc881a.png

 

Thanks

Edited by Govnah
Link to comment

I under probably an issue here with Krusader and what troubles me is this..... I have connected an external usb hdd to copy all its contents to the unraid and therefore the mac users can work from there their files.  The external hdd is formated in hfs+ file system and even though i managed to copy all 394gb ok, it couldn't copy the following files/directories

.DocumentRevisions-v100 --x

.TemporaryItems               -wr  (w=write, r=read, x=?)

.Trashes                           -xw

.journal                             ---

.journal_info_block            --- (under user:root, Group:root)

 

Right clicking TemporaryItems, Trashes and Document Revisions inside Krusader and going to permissions->advanced permisions and under user, checking show entries, seemed to do the trick and copy the above files to the unrad (Not sure if they are needed though to exist at all - I am not a mac user and probably never will but our designers use them) but for .journal and .journal_info_block   everything in properties are greyd out and can t do anything. Is there a way through terminal or something to take ownership kind of and copy them (again not sure if i need them at first place

 

Thank you)

Link to comment

In Linux, directories that begin with a dot (.) are considered to be hidden directories.  Also, a Linux directory must have the "execute" attribute turned on to be browse-able. (i.e., to be able to list the contents of the directory.)  I suspect that they are housekeeping files and directories used by the Apple OS and for internal use by other programs.   

 

I am not an Apple user but I would suggest that you use Google (Google is your friend if you have to support different computing environments..) and see if you can find out what these contents of these directories are.  I know that the Apple OS's are based on UNIX and the file permissions are probably very similar.  What you have run into is always an issue when you mount a disk using a 'foreign' format on another OS.   

 

One thing to try is to copy these files/directories to your server using SMB from an Apple computer.  You have done the heavy lifting by getting the data files across.  The permissions will probably get converted as they come across...

Edited by Frank1940
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.