Expand LV data from 8 to 24 TB?

[proxtest]

I have a server running a RAID6 with 8 TB useable VG space. Now i add 4 x 4 TB and grow the RAID6 and want to put all the new space into the LV-Pool data. Now i read something about fs size max 16 TB (no 64Bit Flag on root?!) but i think it dosn't affects me because i use lvm with thin provisioning? There are many old Infomations about 4MiB PE Size and so forth.
So i'm looking for data about maximal PV, VG and LV size in proxmox-ve: 4.4-86 (running kernel: 4.4.49-1-pve?

What is the maximal size of lv thin metadata? Still 16GB? Will this be enough for 24TB of LV's data? Should i switch to non thin provisioning LV's?
Can i expand my LV data (Proxmox layout) from 8 TB to 24 TB without issue?

Regards

 

[proxtest]

i do the Ingrid

Looks like the default is already the maximal size of metadata?

data pve twi-aotz-- 7,22t 4,92 0,18
[data_tdata] pve Twi-ao---- 7,22t
[data_tmeta] pve ewi-ao---- 15,81g
[lvol0_pmspare] pve ewi------- 128,00m

So i think i will only add space to the data section.

 

[dcsapak]

please note that lvmthin becomes unstable if the metadata pool is full, so you should probably monitor the metadata pool usage

 

[proxtest]

No doubt about this! I try to find some docs but can only find older ones, so still don't know the limits!

The allowed range of metadata space sizes is 2 MiB - 16 GiB and the LVM documentation (lvmthin(7)) suggests 1 GiB as the default value which should be okay for most of the use cases.

http://www-rhstorage.rhcloud.com/blog/vpodzime/lvm-thin-provisioning

Still true?

 

[fabian]

No doubt about this! I try to find some docs but can only find older ones, so still don't know the limits!

The allowed range of metadata space sizes is 2 MiB - 16 GiB and the LVM documentation (lvmthin(7)) suggests 1 GiB as the default value which should be okay for most of the use cases.

http://www-rhstorage.rhcloud.com/blog/vpodzime/lvm-thin-provisioning

Still true?

I am not sure how you get the 1 GiB as suggestion (it is used in some of the examples). The actual size depends on the chunk size of your pool:

   Chunk size

       The  size of data blocks managed by a thin pool can be specified with the --chunksize option when the thin pool LV is created.  The
       default unit is kilobytes and the default value is 64KiB.  The value must be a power of two between 4KiB and 1GiB.

       When a thin pool is used primarily for the thin provisioning feature, a larger value is optimal.  To optimize for a  lot  of  snap‐
       shotting, a smaller value reduces copying time and consumes less space.

       Command to display the thin pool LV chunk size:
       lvs -o+chunksize VG/ThinPoolLV

       Example
       # lvs -o name,chunksize
         pool0 64.00k

       lvm.conf(5) thin_pool_chunk_size
       controls the default chunk size used when creating a thin pool.

   Size of pool metadata LV

       The amount of thin metadata depends on how many blocks are shared between thin LVs (i.e. through snapshots).  A thin pool with many
       snapshots may need a larger metadata LV.

       The range of supported metadata LV sizes is 2MiB to 16GiB.
       The default size is estimated with the formula:
       ThinPoolLVSize / ThinPoolLVChunkSize * 64b.

       When creating a thin metadata LV explicitly, the size is specified in the lvcreate command.  When a command automatically creates a
       thin metadata LV, the --poolmetadatasize option can be used specify a non-default size.  The default unit is megabytes.

so I'd suggest checking your chunk_size ("lvs -o name,chunk_size") and plugging it into the formula for the default size. if you want to be on the safe side, add some more

 

[proxtest]

1 GB? I don't know what u talking about. I was asking about the 16 GB limit on thin pools.

My metadata is already near 16 GB as you can see in my second post.

data pve twi-aotz-- 7,22t 4,92 0,18
[data_tdata] pve Twi-ao---- 7,22t
[data_tmeta] pve ewi-ao---- 15,81g
[lvol0_pmspare] pve ewi------- 128,00m

LV............VG....Attr.......... LSize...Pool..Origin..Data%...Meta%
data.........pve...twi-aotz--..10,86t........................3,30......0,18

And i see 0,18% in use for some Snapshots i think.
So my question was: Is there still the 16 GB Limit for the thin pool? Can and should i resize it to more? Or does it mean 16GB per LV / VM-disc?
No, i think its a shared pool.?

I don't think i will ever use so much but i want to make sure i don't tap into the shitbox.

Did you have some newer docs about the limits?
Only for interest: Why the root fs has no 64Bit flag?

 

[fabian]

yes, AFAIK the docs are still correct there and 16 GiB is the upper limit for the metadata volume. for the default chunk size of 64k, LVM would allocate 10G of metadata for 10T of data, so your 16GiB is already with ~60% extra (assuming your pool uses the default chunk size!).

 

[proxtest]

OK thanks for clearing it!

Raid is still reshaping, after that i should have around 18TB in my pv, so it fits perfect.