[PVE-User] Create proxmox cluster / storage question.
elacunza at binovo.es
Fri Feb 28 15:05:35 CET 2020
El 28/2/20 a las 14:43, Leandro Roggerone escribió:
> Regarding your question , what is the tarjet use for this server.
> I have a dell R610 with 6 drive bays.
> Today I have 4 (2TB) drives in Raid5 , resulting a 5.5TB capacity.
> I will add 2 ssd drives later in raid1 for applications that need more read
> The purpose for this server is to run proxmox with some VMs for external
> and internal access.
> Im planning to build a second server and create a cluster just to have more
> redundancy and availability.
> I would like to set all I can now that server is not in production and
> minimize risk later.
> Thats why im asking so many questions.
Asking is good, but we need info to be able to help you ;)
When you talk about redundacy and availability, do you want...
- HA? (automatic restart of VMs in the other node in case one server fails)
- Be able to move VMs from one server to the other "fast" (without
copying the disks)?
If your answer is yes to any of the previous questions, you have to look
at using a NFS server or deploying Ceph.
If it's no, then we can talk about local storage in your servers. What
RAID card do you have in that server? Does it have write cache (non
volatile of battery-backed) If it doesn't have such, RAID5 could prove
slow (eat quite CPU), I suggest you use 2xRAID1 or a RAID10 setup. Also,
please bear in mind that RAID5 with "big" disks is considered quite
unsecure (risk of having a second disk failure during recovery is high).
> El vie., 28 feb. 2020 a las 5:49, Eneko Lacunza (<elacunza at binovo.es>)
>> Hola Leandro,
>> El 27/2/20 a las 17:29, Leandro Roggerone escribió:
>>> Hi guys , i'm still tunning my 5.5 Tb server.
>>> While setting storage options during install process, I set 2000 for hd
>>> size, so I have 3.5 TB free to assign later.
>>> my layout is as follows:
>>> root at pve:~# lsblk
>>> NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
>>> sda 8:0 0 5.5T 0 disk
>>> ├─sda1 8:1 0 1007K 0 part
>>> ├─sda2 8:2 0 512M 0 part
>>> └─sda3 8:3 0 2T 0 part
>>> ├─pve-swap 253:0 0 8G 0 lvm [SWAP]
>>> ├─pve-root 253:1 0 1T 0 lvm /
>>> ├─pve-data_tmeta 253:2 0 9G 0 lvm
>>> │ └─pve-data 253:4 0 949.6G 0 lvm
>>> └─pve-data_tdata 253:3 0 949.6G 0 lvm
>>> └─pve-data 253:4 0 949.6G 0 lvm
>>> sr0 11:0 1 1024M 0 rom
>>> My question is:
>>> Is it possible to expand sda3 partition later without service outage ?
>>> Is it possible to expand pve group on sda3 partition ?
>> You don't need to expand sda3 really. You can just create a new
>> partition, create a new PV with it and add the new PV to pve VG.
>>> In case to create a proxmox cluster, what should I do with that 3.5 TB
>> I don't know really how to reply to this. If you're building a cluster,
>> I suggest you configure some kind of shared storage; NFS server or Ceph
>> cluster for example.
>>> Is there a best partition type suited for this ? Can I do it without
>>> service outage?
>> For what?
>>> I have not any service running yet , so I can experiment what it takes.
>>> Any thought about this would be great.
>> Maybe you can start telling us your target use for this server/cluster.
>> Also some detailed spec of the server would help; for example does it
>> have a RAID card with more than one disk, or you're using a 6TB single
>> Zuzendari Teknikoa / Director Técnico
>> Binovo IT Human Project, S.L.
>> Telf. 943569206
>> Astigarragako bidea 2, 2º izq. oficina 11; 20180 Oiartzun (Gipuzkoa)
>> pve-user mailing list
>> pve-user at pve.proxmox.com
> pve-user mailing list
> pve-user at pve.proxmox.com
Zuzendari Teknikoa / Director Técnico
Binovo IT Human Project, S.L.
Astigarragako bidea 2, 2º izq. oficina 11; 20180 Oiartzun (Gipuzkoa)
More information about the pve-user