Jump to content

Unraid 653 locks up after deletion of many files


CDebarry

Recommended Posts

Just now, CDebarry said:

Ohhh so maybe cache is used when deleting files even though these folks are not on cache? Investigating now.

I don't know what you mean here. If you were deleting files that were on cache then cache would be where the files were being deleted from. It's as simple as that. If the files you were deleting were not on cache, then cache would not be involved.

Link to comment
1 minute ago, CDebarry said:

I mean really remove the cache drive completely.

Unless you really have some good reason to think that might be the right way to fix it then you should wait on further advice. It doesn't seem to me you even know which disk you were deleting data from.

 

Why exactly are you trying to delete so much data anyway?

Link to comment
33 minutes ago, trurl said:

Unless you really have some good reason to think that might be the right way to fix it then you should wait on further advice. It doesn't seem to me you even know which disk you were deleting data from.

 

Why exactly are you trying to delete so much data anyway?

I'm deleting old and unused files so backup to cloud is faster and cheaper. They are small images files. I'm deleting from share, so I guess it shouldn't matter what disk they're at.

 

Anyway data is on disks 1, 2 and 3 but definitely not on share.

Link to comment
9 hours ago, CDebarry said:

Anyway data is on disks 1, 2 and 3 but definitely not on share.

This statement doesn't make any sense. Not on what "share"?

 

Maybe some of this would make more sense to you if understood how user shares actually work under the hood. It is really pretty simple.

 

The user shares are simply the top level folders on cache and array with the same name. If you have a top level folder named "media" on disk1 and a top level folder named "media" on disk2, for example, all the files in the "media" folder on both of those disks are in the user share named "media".

 

Any time you create a user share, Unraid creates a top level folder, named the same as that user share, on cache or array as needed, according to the settings for that user share.

 

Conversely, if you have a top level folder on cache or array, it is a user share named for that folder. It doesn't matter how that top level folder got created. It is a user share even if you didn't tell Unraid to create that user share. If you create a top level folder on cache or array, it is a user share with default settings until you make some settings for it.

 

So, assuming the data on disks 1,2,3 are in some folder, they are definitely in some user share.

 

Link to comment

And you seem to be somewhat confused about cache as well. Cache isn't some mysterious place where data comes and goes on its way to and from the array. It is just another disk (or maybe a pool).

 

Unlike the array disks, parity isn't involved, so it can be written faster than the array disks. It is typically one (as you have) or more SSDs and so will perform even faster than the spinners in the array. But it also will typically have a lot less capacity than the array. Because of the way user shares work, you can think of the capacity of the array as being all of the array disks combined.

 

There is a process called "mover" that will move data on schedule from cache to array (or from array to cache) depending on the Use cache setting for each user share.

 

So, if you were deleting data from a user share, and some of that data happened to be on cache at the time, then the cache disk would be involved in the delete of the part of the data that was on cache. It's as simple as that.

 

Just remember that cache disk(s) is also part of the user shares just like the array disks are.

Link to comment

So, how exactly were you deleting these files on disks 1,2,3? I ask this question because you didn't seem to think they were in a share.

 

People will often just work with the shares and not the individual disks. And, in fact, that is often all that you need to do. It is possible to get yourself into trouble if you try to work with the individual disks unless you know what you're doing.

Link to comment

Hi Everyone,

 

Thanks for all replies. Maybe I cant express my self well enough, sorry.

I understand how everything works! :D

 

I know data is written to the cache (witch in m case is an 120gb ssd) and then moved to disks with parity - (these are 5 mechanical drives). Not all shares are using cache due to its low capacity.

image.thumb.png.a0931f747f4f5870730ca7c7e29b63f5.png

 

I only work on the files from the shares..

image.thumb.png.19a6f4d3f777239b3c53e99c00fd3aad.pngThe folder I have to delete is the one called "_digital2". Data is located on disks 1, 2 and 3 like I said.

 

I just updated OS version to the latest: 667.

I'll try again.

Link to comment
34 minutes ago, CDebarry said:

delete is the one called "_digital2".

That last screenshot indicates the folder named "_DIGITAL2" is in the User Share named docxpress. And just in case you aren't aware, it is not called "_digital2" because linux is case-sensitive.

 

34 minutes ago, CDebarry said:

Not all shares are using cache due to its low capacity.

With so little cache, I would reserve it for use only by your dockers and VMs; i.e., the appdata, system, and domain shares that are cache-prefer. All other shares set to cache-no. No wonder you were filling it up.

 

It has enough capacity for some other things as well if you want fast storage for a little bit of stuff, but it isn't big enough to cache writes to the array in my opinion.

Link to comment

I guess you still didn't understand me.

 

I was DELETING content from share 'docxpress' folder '_digital2' when system frooze. It's not supposed to completely halt while deleting files much less use cache for deletion as far as I know.

 

Get it?

Link to comment

Hi everyone,

 

After upgrading to the latest version the problem was solved. Im able to delete a huge amount of data at once as I need to.

 

Also, the cache drive is definitely not filling up even though I still see the message about it in the log.

 

Don't know why.

Thanks again.

 

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...