Data now showing in Hierarchical Pool

[kpate77]

Pool Drive: 2.2.2.928

I've had an existing pool setup for years with x2 duplication.  However, I started running out of disk space and decided to use Clouddrive to backup instead of x2 locally.  I removed duplication on my original pool, shrinking it from 120TB to 60TB used / 60 TB available.  The 60TB of data is showing in original pool as Unduplicated / Blue.

I created a new pool, intending to add the original pool (described above) and a Cloud Drive (or pool of cloud drives).  However, when I add the original pool to the new pool, I do not see the data (as unduplicated / blue).  It is showing as Other / Gray, and I can not see the folders under file duplication options.  

Do I need to seed this new pool somehow?

I read the blog posting about the hierarchical pools and followed the instructions on setting up the pools.  However the new pool does not seem to be reading the data from the existing pool.

 

 

[Viktor]

A new pool starts always empty and does not automatically take over existing data from an item added to it, regardless of whether the added item is a real drive or a pool.

It’s always the same mechanism: Adding a drive (or another pool) to a pool means, that a new (empty) PoolPart folder is created on the added drive or pool, what in case of pool hierarchies results in nested PoolPart folders (folder in folder) on involved single drives.

There are (unsupported) “seeding” methods by moving “other” data into PoolPart folders, but the more pools and nested PoolPart folders are involved, the more complicated can this become.

[kpate77]

Thanks for your reply!

I get what you are saying, and that is the behavior I am experiencing.

However, I am trying to accomplish what is written in the following blog post:

http://blog.covecube.com/2017/09/stablebit-drivepool-2-2-0-847-beta/

 

Basically:

Hybrid Pool = Local Pool + Cloud Pool,  with x2 duplication from Local to Cloud.  This is what is explained in the above blog post.

However when I create the above, there is nothing in my Hybrid pool to duplicate.  How do I accomplish what is in the blog post (assuming functionality hasn't changed from 2.2.0 to 2.2.2?  

[kpate77]

I've been reading more in the forums and I think I have a better understanding now.  

I think I should seed the Hybrid pool using the following method:  http://wiki.covecube.com/StableBit_DrivePool_Q4142489

Then on the Hybrid pool, choose file placement onto the Local Pool only.  Choose x2 duplication on Hybrid Pool. 

The Hybrid Pool will be the new pool that I read/write from going forward.  

I've tested this on a couple of files and it seems to accomplish what I am looking to do.

[Umfriend]

It is actually easy. Let's start with your starting position:
1. You had a Pool of local HDDs, say Pool A (or Local). On these HDDs you have a hidden PoolPart.* folder in which the data of Pool A is stored.
2. You created a CloudDrive (possibly added it to a new Pool, Pool B (or CloudPool), I'll assume that)
3. You created a new Pool C (or HYbridPool) and added Pool A and Pool B to it.

Now on HDDs of Pool A, there will be another hidden PoolPart.* folder (say level 2) _within_ the earlier PoolPart.* ( level 1) folder. Anything stored in level 1 is in Pool A ONLY. When you look at Pool C, the data in level 1 will show up as Other. This is in fact very similar to storing data on a HDD of Pool A in the root folder or outside of the level 1 PoolPart.* folder: It will show up in Pool A as Other.

So, for each HDD in Pool A, _move_ the contents of the level 1 folder to the level 2 folder (stopping DrivePool service first and restarting it when done). The that data is in Pool C, unduplicated and the re-measure pass will duplicate Pool C (by storing a duplicate in Pool B).

There is no need for Fileplacement rules (that is assuming you have x2 for everything in the Pool). And you cannot have files only on Local _and_ have x2 duplication.

[kpate77]

It seems to be working now.  Thank you both for your help!

 

 

[Christopher (Drashna)]

Just a heads up, the duplicated space on the drive is fine.  This is the metadata folder, most likely, which is set to x3 duplication, and probably acounts for all of the duplicated data on that pool.

And yeah, you'd just seed the drive "like normal".