Ceph Mgr Orchestrator Enabled but showing missing?

dmkjr

Member
Sep 6, 2021
10
0
6
44
Good morning community! I have recently created a new ceph cluster and we are looking good in terms of functionality. I have enabled ceph-mgr on one of the nodes and am currently snooping around to see what information I can gleam from the administrator panel. For the most part all is working but some sections are stating that it requires Orchestrator "Orchestrator is not available. ".

However, when I run
Code:
ceph orch status
I received the following:

Code:
Error ENOENT: No orchestrator configured (try `ceph orch set backend`)

However, when I run
CSS:
ceph mgr module enable orchestrator
shows it's already enabled.

I have tried the
Code:
ceph orch set backend
which doesn't present any options but appears to run successfully. Not sure what to try at this step, cannot find much information on the topic.

Thank you!
 
You would have to manually add the services and configuration to the dashboard.
How would you manually add the services, especially if orchestrator is already showing enabled? Kind of looks like a false positive as everything works.
 
I have also wondered about this. How is it now when I want to put a Ceph node into maintenance mode?

According to the ceph documentation this is how it is done:
Code:
ceph orch host ls
ceph orch host ok-to-stop <host>
ceph orch host maintenance enter <host> [ --force ]
ceph orch host maintenance exit <host>

does this function even exist in proxmox?
 
This maintenance mode is only necessary for orchestrator managed services that can easily be moved (e.g. MGR).
It does not exist in Proxmox.
Clear so far. To clarify, if I want to reboot a proxmox node with Ceph for maintenance, I have no way to tell the cluster?
It would be nice to be able to tell the cluster not to write anything to the node at this time.
 
So theres no way to enable ceph orch in proxmox ve? I installed the ceph dashboard and there might be some cool features that are only available witih a installed ceph orch in the dashboard. Did anyone manage to get this working without any hassle?


Code:
root@PMX2:~# ceph mgr module enable orchestrator
module 'orchestrator' is already enabled (always-on)

root@PMX2:~# ceph orch set backend orchestrator
Error EINVAL: Module 'orchestrator' is not enabled.
 Run `ceph mgr module enable orchestrator` to enable.
 
root@PMX2:~# ceph orch ls
Error ENOENT: No orchestrator configured (try `ceph orch set backend`)
 
Last edited:
i'm on the same boat. i want to configure OSD via ceph but i can't do it without orchestrator
 
it doesn't make sense to use multiple different pieces of software that both expect to fully manage something as complicated as a ceph cluster. if you want to use the orchestrator, I would suggest keeping your Ceph and PVE cluster separate from eachother and configuring the former as an external storage cluster in the latter. obviously I would recommend to just skip the orchestrator and use the Ceph integration in PVE, especially if you also want support for the Ceph part itself..
 
  • Like
Reactions: gurubert and jsterr
it doesn't make sense to use multiple different pieces of software that both expect to fully manage something as complicated as a ceph cluster. if you want to use the orchestrator, I would suggest keeping your Ceph and PVE cluster separate from eachother and configuring the former as an external storage cluster in the latter. obviously I would recommend to just skip the orchestrator and use the Ceph integration in PVE, especially if you also want support for the Ceph part itself..
the reason i wanted to use ceph gui is that I have a ssd and I want to use it as a DB disk but pve can't cut it equally for the HDD. I was told that to do it in ceph GUI.
However, I agreed with you, if PVE can do everything, no reason to use something else
 
the reason i wanted to use ceph gui is that I have a ssd and I want to use it as a DB disk but pve can't cut it equally for the HDD. I was told that to do it in ceph GUI.
However, I agreed with you, if PVE can do everything, no reason to use something else
You actually can? Just devide SSD size with hdd count and lower it by 10% or so and type the size into the wizard when creating the osds. You can also create osd with db device via cli. But if your ssd fails, your osds will also fail!
 
Code:
Error ENOENT: No orchestrator configured (try `ceph orch set backend`)
这条提示命令的末尾,需要加上对应的模块。
我是用 cephadm 部署的集群,我使用如下命令解决这个问题:
Code:
ceph orch set backend cephadm

在我的集群上,除了 cephadm 之外,还有 rook ,test_orchestrator 这两个模块可供选择。
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!