[SOLVED] HELP!!! undo a pvesm remove to CEPH storage

[pcao]

A collegue does an "pvesm remove VMs" and the qemu and lxv volumes named VMs disappear from the GUI

The doc says what the command pvesm remove

"Remove storage pools. This does not delete any data, and does not disconnect or unmount anything. It just removes the storage configuration".

But I dont make backup of this file !! (shame on me, I know)

Question: How can I get back my storage back ?

I try this

rbd: VMs
        monhost 10.0.0.1:6789 10.0.0.2:6789 10.0.0.3:6789
        pool mon_pool
        content rootdir,images
        username ceph

or some variant:

• without port in variavle monhsot
• without username

Any idea would be nice.

Context: proxmox 5.4 a old promox do its jobs very well for a long time

 

[fabian]

what the right data is depends on how your ceph cluster and pool was setup.. is this a hyperconverged setup (ceph running on the PVE nodes) or an external cluster?

 

[pcao]

yes this is a simple proxmox setup with ceph included in all 3 nodes running proxmox

 

[pcao]

with pvesm status something with rados_connect error:

rados_connect failed - Invalid argument
rados_connect failed - Invalid argument
Name             Type     Status           Total            Used       Available        %
VMs               rbd   inactive               0               0               0    0.00%
local             dir     active        98559220        79597204        13912468   80.76%
local-lvm     lvmthin     active       850669568               0       850669568    0.00%

So I copy /etc/ceph/ceph.client.admin.keyring to /etc/pve/priv/ceph/VMs.keyring but without change on the error message

 

[fabian]

what does ceph osd pool ls say? what about pveversion -v?

 

[pcao]

It' work !!!!
With storage.cfg

rbd: VMs
        monhost 10.0.0.1:6789 10.0.0.2:6789 10.0.0.3:6789
        pool mon_pool
        content rootdir,images

and pvesm status :

Name             Type     Status           Total            Used       Available        %
VMs               rbd     active      3645897417      1121278921      2524618496   30.75%
local             dir     active        98559220        79598884        13910788   80.76%
local-lvm     lvmthin     active       850669568               0       850669568    0.00%

Now I have to retreive the info in users.cfg

to be continue ...

 

[pcao]

the file users.cfg has its time changed but no impact to the GUI roles
edit: SOLVED !