[rlug] Proxmox ceph pe un nod reconstruit was boot de pe un disc clonat cu grub rescue

[Paul Lacatus]

revin cu update la povestea de ieri

Clusterul pare sa mearga ok , Ceph nu da nici o eroare , aria a fost 
reconstruita dar in sysylog la nodul reconstruit apare un warning

Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: WARNING:ceph-crash:post 
/var/lib/ceph/crash/2023-12-19T09:05:39.204755Z_2b65546d-25fd-4b14-bf05-5cb95ca3ad30 
as client.admin failed: 2023-12-20T10:19:47.424+0200 7f499a33e6c0 -1 
auth: unable to find a keyring on 
/etc/pve/priv/ceph.client.admin.keyring: (13) Permission denied
Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: 2023-12-20T10:19:47.424+0200 
7f499a33e6c0 -1 auth: unable to find a keyring on 
/etc/pve/priv/ceph.client.admin.keyring: (13) Permission denied
Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: 2023-12-20T10:19:47.424+0200 
7f499a33e6c0 -1 auth: unable to find a keyring on 
/etc/pve/priv/ceph.client.admin.keyring: (13) Permission denied
Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: 2023-12-20T10:19:47.424+0200 
7f499a33e6c0 -1 auth: unable to find a keyring on 
/etc/pve/priv/ceph.client.admin.keyring: (13) Permission denied
Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: 2023-12-20T10:19:47.424+0200 
7f499a33e6c0 -1 monclient: keyring not found
Dec 20 10:19:47 pve-m710-3 ceph-crash[553]: [errno 13] RADOS permission 
denied (error connecting to the cluster)

presupun ca a fost o problema cand am reintrodus nodul nou in cluster si 
in ssh hosts la nodul care face managementul clusterului era semnatura 
nodului vechi.  Am rezolvat ulterior treaba asta dar e posibil sa nu fi 
facut totul ok cand am refacut configurarile de ceph pe nodul nou .

O parere , o idee ? Ca vad ca va plictisiti asteptand sa vina Mosu' cu 
bradu.

Paul

On 19.12.2023 16:37, Paul Lacatus via RLUG wrote:
> configuratia era de la proxmox cu partitie de boot, EFI si root. Dar 
> intrand cu un ventoy si boot proxmox debug am vazut ca ceva aiurea 
> i-am dat eu la clonezilla ca pe discul clonat nu aveam decat partitia 
> de root si lipseau  boot si EFI si cum clonarea a durat 12 ore am zis 
> pas.  Asta cu toate ca am facut clonare de device nu de partitie. Am 
> sters nodul din cluster si am reinstalat de la capat un nod cu acelasi 
> nume, acelasi ip,  bagat in cluster si refacut osd de la ceph. Acum se 
> curata si rebuild ceph. VM care oricum erau in ceph pool le-am 
> restaurat din Proxmox backup server si acum pare ca e ok totul . Mai 
> verific ca nu sunt inca 100% convins
>
>
> Paul
>
> On 19.12.2023 12:50, Alex 'CAVE' Cernat via RLUG wrote:
>> On 19-Dec-23 09:10, Paul Lacatus via RLUG wrote:
>>> Salut
>>>
>>> La o masina cu proxmox ssd de pe care booteza proxmox a inceput sa 
>>> dea erori. L-am clonat cu clonezilla pe un disc de dimensiune 
>>> apropiata . Discul nou porneste in grub rescue . cum fac sa bootez 
>>> si sa repar grub ?
>>>
>>> Multumesc
>>>
>>> Paul
>>
>> salut
>>
>> daca vorbim de proxmox atunci foarte probabil avem de-a face cu lvm 
>> sau zfs care mai complica treaba; eventual si cu combinatii de EFI / 
>> secure boot care sporesc "distractia"
>>
>> care era configuratia exacta?
>>
>> Alex
>>
>>
>> _______________________________________________
>> RLUG mailing list
>> RLUG at lists.lug.ro
>> http://lists.lug.ro/mailman/listinfo/rlug_lists.lug.ro
>
> _______________________________________________
> RLUG mailing list
> RLUG at lists.lug.ro
> http://lists.lug.ro/mailman/listinfo/rlug_lists.lug.ro