Re: Ceph RBD w/erasure coding

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

 



100% agree.  

I’ve seen claims that SSDs synch quickly so 2 is enough.   Such claims are shortsighted.  

I have personally witnessed cases of device failures and OSD flaps / crashes at just the wrong times resulted in no clear latest copy of PGs.  

There are cases where data loss isn’t catastrophic, but unless you’re sure, you want either R3 for durability or EC to minimize space amp.   A 2,2 profile, for example.  

> On Sep 15, 2024, at 10:35 AM, Joachim Kraftmayer <joachim.kraftmayer@xxxxxxxxx> wrote:
> 
> first comment on the replicated pools:
> the replication size for rbd pools of 2 is not suitable for production
> clusters. It is only a matter of time before you lose data.
> Joachim
> 
> 
>  www.clyso.com
> 
>  Hohenzollernstr. 27, 80801 Munich
> 
> Utting a. A. | HR: Augsburg | HRB: 25866 | USt. ID-Nr.: DE2754306
> 
> 
> 
>> Am Sa., 14. Sept. 2024 um 14:04 Uhr schrieb <przemek.kuczynski@xxxxxxxxx>:
>> 
>> Here You have guide
>> 
>> https://febryandana.xyz/posts/deploy-ceph-openstack-cluster/
>> 
>> in short
>> 
>> ceph osd pool create images 128
>> ceph osd pool set images size 2
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd pool create volumes 128
>> ceph osd pool set volumes size 2
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd pool create vms 128
>> ceph osd pool set vms size 2
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd erasure-code-profile set ec-22-profile k=2 m=2
>> crush-device-class=ssd
>> ceph osd erasure-code-profile ls
>> ceph osd erasure-code-profile get ec-22-profile
>> 
>> ceph osd pool create images_data 128 128 erasure ec-22-profile
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd pool create volumes_data 128 128 erasure ec-22-profile
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd pool create vms_data 128 128 erasure ec-22-profile
>> while [ $(ceph -s | grep creating -c) -gt 0 ]; do echo -n .;sleep 1; done
>> 
>> ceph osd pool ls detail
>> 
>> ceph osd pool set images_data allow_ec_overwrites true
>> ceph osd pool set volumes_data allow_ec_overwrites true
>> ceph osd pool set vms_data allow_ec_overwrites true
>> 
>> ceph osd pool application enable volumes rbd
>> ceph osd pool application enable images rbd
>> ceph osd pool application enable vms rbd
>> ceph osd pool application enable volumes_data rbd
>> ceph osd pool application enable images_data rbd
>> ceph osd pool application enable vms_data rbd
>> On ceph.conf You need put as below
>> 
>> [client.glance]
>> rbd default data pool = images_data
>> 
>> [client.cinder]
>> rbd default data pool = volumes_data
>> 
>> [client.nova]
>> rbd default data pool = vms_data
>> for permission you probably also need add
>> 
>> caps mon = "allow r, allow command \\"osd blacklist\\", allow command
>> \\"osd blocklist\\", allow command \\"blacklistop\\", allow command
>> \\"blocklistop\\""
>> Newer versions might not work with blacklist anymore.
>> _______________________________________________
>> 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
_______________________________________________
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