I solved  by creating new then removing old pool

发件人: 廖建锋<mailto:[email protected]>
发送时间: 2014-09-09 17:39
收件人: haomaiwang<mailto:[email protected]>
抄送: ceph-users<mailto:[email protected]>; 
ceph-users<mailto:[email protected]>
主题: Re: Re: [ceph-users] 回复: mix ceph verion with 0.80.5 and 0.85
I re-installed the whole cluster with ceph 0.85 and lost all my 10T data
Now i have another question is  which i have not way to re-create pool.

[cid:_Foxmail.1@2488d66a-606e-bac6-c084-a4641a98b3be]

264 => # ceph osd pool delete data data --yes-i-really-really-mean-it
Error EBUSY: pool 'data' is in use by CephFS



发件人: Haomai Wang<mailto:[email protected]>
发送时间: 2014-09-09 17:28
收件人: 廖建锋<mailto:[email protected]>
抄送: ceph-users<mailto:[email protected]>; 
ceph-users<mailto:[email protected]>
主题: Re: [ceph-users] 回复: mix ceph verion with 0.80.5 and 0.85
Hi,

Thanks for your report, I will fix it(https://github.com/ceph/ceph/pull/2429).

Because KeyValueStore is aimed as a experiment backend, we still don't have 
enough test suite to cover it.



On Tue, Sep 9, 2014 at 11:02 AM, 廖建锋 <[email protected]<mailto:[email protected]>> 
wrote:
Looks like it dosn't work,  i noticed that 0.85 added superblock to leveldb 
osd,   the osd which I alread have do not have superblock
is there anybody can tell me how to upgrade OSDs ?



发件人: ceph-users<mailto:[email protected]>
发送时间: 2014-09-09 10:32
收件人: ceph-users<mailto:[email protected]>
主题: [ceph-users] mix ceph verion with 0.80.5 and 0.85
dear,
     As there are a lot of bugs of keyvalue backend in 0.80.5 firely version ,  
So i want to upgrade to 0.85 for some osds which already down and unable to 
start
and keep some other osd with 0.80.5,I wondering ,  will it works?


_______________________________________________
ceph-users mailing list
[email protected]<mailto:[email protected]>
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com




--

Best Regards,

Wheat
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to