[SOLVED] Thin volume error

[Maher Khalil]

Hello
Can anyone explain what is this error?

snapshotting 'drive-scsi0' (local-lvm:vm-101-disk-0)
WARNING: You have not turned on protection against thin pools running out of space.
WARNING: Set activation/thin_pool_autoextend_threshold below 100 to trigger automatic extension of thin pools before they get full.
Logical volume "snap_vm-101-disk-0_update_2024-02-19" created.
WARNING: Sum of all thin volume sizes (1.70 TiB) exceeds the size of thin pool pve/data and the size of whole volume group (<893.25 GiB).
TASK OK

 

[sb-jw]

Can anyone explain what is this error?

This is a warning and not an error.

Stupid question, have you read what it says? And if so, what exactly is it that you don't understand?
In my opinion, everything you need to know is in the message.

 

[bbgeek17]

None of the messages that you posted are errors, they are all pre-fixed with "WARNING:". The difference is that you when you hit an Error - the operation will fail. Except in some Microsoft cases where you see "Error: Success".

As @sb-jw said the messages are pretty self-explanatory:

WARNING: You have not turned on protection against thin pools running out of space.

https://access.redhat.com/documenta...isioned-volumes#:~:text=Create a thin volume:

WARNING: Set activation/thin_pool_autoextend_threshold below 100 to trigger automatic extension of thin pools before they get full.
Logical volume "snap_vm-101-disk-0_update_2024-02-19" created.

You can configure your system to automatically expand the underlying volumes. However, I suspect you dont have available space to do so

WARNING: Sum of all thin volume sizes (1.70 TiB) exceeds the size of thin pool pve/data and the size of whole volume group (<893.25 GiB).

You overprovisioned your space. Imagine that you sold 50 auto-tires over the internet, but only have 40 to ship and you are hoping that 10 will get cancelled, or you might find the 10 before the customers come knocking down on your door...

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Maher Khalil]

Hell bbgeek17
Proxmox shows that I use only 11% out of the pool capacity (attached an image)

 

[bbgeek17]

Proxmox shows that I use only 11% out of the pool capacity (attached an image)

Usable: how much space can you use if you write only encrypted data to disk. Rounding things up a 1GB disk has 1GB usable (less with overhead)
Used: how much space you actually written to disk, ie 200MB out of 1GB
Provisioned: how much space you pretended you have in hopes it will never get all used, ie you thinly provisioned 1TB on a 1GB disk. The LVM layer allows you do so. When you write 1GB you will run out of space.
Effective: how much pretend space you might have if all of your data is compressible and deduplicative.

Your usable is: 893.25 GiB
Your used is: 89.99 GiB
Your provisioned: 1.7 TiB

Do you see now why you received a warning? If all your space consumers will start writing now to the volumes you created, you will run out of space before they can reach 100% of what they think they have.
It may never happen, this is why its a Warning and not an Error. But if it does happen there is a high chance of data corruption, and then you will be back with an Error. The developers of LVM and PVE dont want to hear from you in that case - they WARNED you.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Maher Khalil]

I understand now that there is no error.
In my plan, I will not add any new machines to this pool.
So, error will not come since the only VM on this pool have 600GB which is less of usable (893GB ) and provisioned (1.7 TB)
Am I right?

 

[bbgeek17]

Your numbers dont add up, so I cant provide you with any assurance of your correctness.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Maher Khalil]

i have only 1 VM
VM DISK SIZE = 600 GB (ALLOCATED BUT NOT YEST USED)
POOL SIZE = 893 GB (11% used because of the above VM)
provisioned space = 1.7TB

 

[Dunuin]

i have only 1 VM
VM DISK SIZE = 600 GB (ALLOCATED BUT NOT YEST USED)
POOL SIZE = 893 GB (11% used because of the above VM)
provisioned space = 1.7TB

If that would be true only 600GB should be provisioned. 1.7TB are provisioned, so that VM either got more disks than that 600GB one or you store more guest on it. We can't really help without you providing us with more data like vgs and lvs.

 

[Maher Khalil]

Thank you very much.
I appreciate your help.
I will close this ticket.

 

[EllyMae]

These are the least helpful responses I've ever come across. Has anyone considered how one might want to avoid an error in the future? Isn't that what warnings are for? Otherwise, the warning is completely useless and should not have been issued.

 

[Neobin]

These are the least helpful responses I've ever come across. Has anyone considered how one might want to avoid an error in the future? Isn't that what warnings are for? Otherwise, the warning is completely useless and should not have been issued.

So, why did you not take the opportunity and provide a, in your opinion, more helpful response?!
Only complaining, without doing it "better"...
(BTW: I, and seemingly the OP too, find the answers helpful...)

Regarding the topic:
If you do not want to run into problems:

• Set up proper monitoring with thresholds and alerts
• Do not overprovision
• Do not use thin-provisioning at all

https://forum.proxmox.com/threads/protection-against-thin-pool.121635/post-579586