Re: Ceph Reweight Only works in same host?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,

Thank you for the prompt response.

> Such a heterogenous setup is only possible with a large number of OSDs

A heterogeneous setup is exactly what I am testing for. I wonder if the balancer will actually consider the WEIGHT when allocating the data to OSD.

Let's say I have 3 hosts,  with 10 OSDs in each host. 1st and 2nd host has an equivalent size, 10 TB each host, while 3rd host has 20TB.
In this scenario what will happen? Is it still the same where the 1st and 2nd host will hit storage full while leaving 10TB in the 3rd host?

If so, we should always try our best to balance the hosts to have similar storage space?

Thanks for any advice.
Isaiah



________________________________
From: Robert Sander <r.sander@xxxxxxxxxxxxxxxxxxx>
Sent: Monday, December 19, 2022 3:42 PM
To: ceph-users@xxxxxxx <ceph-users@xxxxxxx>
Subject:  Re: Ceph Reweight Only works in same host?

Hi,

Am 19.12.22 um 08:28 schrieb Isaiah Tang Yue Shun:

> Currently, I have 3 OSDs in 3 different hosts:
>
> # ceph osd tree:
>
> ID  CLASS  WEIGHT   TYPE NAME       STATUS  REWEIGHT  PRI-AFF
> -1         0.03908  root default
> -3         0.00980      host node1
>   0    hdd  0.00980          osd.0       up   1.00000  1.00000
> -9               0      host node2
> -5         0.00980      host node3
>   1    hdd  0.00980          osd.1       up   1.00000  1.00000
> -7         0.01949      host node4
>   2    hdd  0.01949          osd.2       up   1.00000  1.00000
>
> When I insert data into the pool, the data is equally distributed on 3 OSDs. (3 replicas)
> However, I want the osd.2 to hold more data since it has more space.

What additional data should OSD 2 hold?

The pool has a size of 3, i.e. all data objects have three copies.

The cluster only has 3 OSDs, so each OSD gets one copy.

In this configuration your available capacity is that of the smallest
OSD. You cannot use additional space in larger OSDs.

Such a heterogenous setup is only possible with a large number of OSDs
where the placement groups can be assigned more flexible to the OSDs.

Regards
--
Robert Sander
Heinlein Consulting GmbH
Schwedter Str. 8/9b, 10119 Berlin

http://www.heinlein-support.de

Tel: 030 / 405051-43
Fax: 030 / 405051-19

Zwangsangaben lt. §35a GmbHG:
HRB 220009 B / Amtsgericht Berlin-Charlottenburg,
Geschäftsführer: Peer Heinlein -- Sitz: Berlin

_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux