[ceph-users] creating+incomplete issues

2015-10-27 Thread Wah Peng
hello, After installing ceph I tried to watch it with ceph -w, 2015-10-28 14:54:08.035995 mon.0 [INF] pgmap v82: 192 pgs: 104 active+degraded+remapped, 88 creating+incomplete; 0 bytes data, 36775 MB used, 113 GB / 156 GB avail 2015-10-28 14:54:12.327050 mon.0 [INF] pgmap v83: 192 pgs: 104 act

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
Hello, $ ceph osd stat osdmap e18: 2 osds: 2 up, 2 in this is what it shows. does it mean I need to add up to 3 osds? I just use the default setup. thx. On 2015/10/28 星期三 19:53, Gurjar, Unmesh wrote: Are all the OSDs being reported as 'up' and 'in'? This can be checked by executing 'c

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
:38, Lindsay Mathieson wrote: On 29 October 2015 at 10:29, Wah Peng mailto:wah_p...@yahoo.com.sg>> wrote: $ ceph osd stat osdmap e18: 2 osds: 2 up, 2 in this is what it shows. does it mean I need to add up to 3 osds? I just use the default setup. If you went wi

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
四 8:55, Robert LeBlanc wrote: Please paste 'ceph osd tree'. Robert LeBlanc Sent from a mobile device please excuse any typos. On Oct 28, 2015 6:54 PM, "Wah Peng" mailto:wah_p...@yahoo.com.sg>> wrote: Hello, Just did it, but still no good health. can you he

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
wow this sounds hard to me. can you show the details? thanks a lot. On 2015/10/29 星期四 9:01, Robert LeBlanc wrote: You need to change the CRUSH map to select osd instead of host. Robert LeBlanc Sent from a mobile device please excuse any typos. On Oct 28, 2015 7:00 PM, "Wah

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
Behalf Of Wah Peng Sent: 2015年10月29日 9:14 To: Robert LeBlanc Cc: Lindsay Mathieson; Gurjar, Unmesh; ceph-users@lists.ceph.com Subject: Re: [ceph-users] creating+incomplete issues wow this sounds hard to me. can you show the details? thanks a lot. On 2015/10/29 星期四 9:01, Robert LeBlanc wrote: Y

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
h-map/#editing-a-crush-map Robert LeBlanc Sent from a mobile device please excuse any typos. On Oct 28, 2015 7:46 PM, "Wah Peng" mailto:wah_p...@yahoo.com.sg>> wrote: Is there a ceph sub-command existing instead of changing the config file? :) On 2015/10/29 星期四 9:24

Re: [ceph-users] creating+incomplete issues

2015-10-28 Thread Wah Peng
I have changed this line, step chooseleaf firstn 0 type osd the type from "host" to "osd". Now the health looks fine: $ ceph health HEALTH_OK Thanks for all the helps. On 2015/10/29 星期四 10:35, Wah Peng wrote: Hello, this shows the content of crush-map file, what con

[ceph-users] Input/output error

2015-10-29 Thread Wah Peng
hello, do you know why this happens when I did it following the official ducumentation. $ sudo rbd map foo --name client.admin rbd: add failed: (5) Input/output error the OS kernel, $ uname -a Linux ceph.yygamedev.com 3.2.0-23-generic #36-Ubuntu SMP Tue Apr 10 20:39:51 UTC 2012 x86_64 x86

Re: [ceph-users] Input/output error

2015-10-29 Thread Wah Peng
onfirm if the rbd module is loaded (sudo modprobe rbd) on the ceph-client node and give it a retry. If you still encounter the issue, post back the snippet of error logs in syslog or dmesg to take it forward. Regards, Unmesh G. -Original Message- From: ceph-users [mailto:ceph-users-boun...@li

Re: [ceph-users] Input/output error

2015-10-29 Thread Wah Peng
$ ceph -v ceph version 0.80.10 (ea6c958c38df1216bf95c927f143d8b13c4a9e70) thanks. On 2015/10/29 星期四 18:23, Ilya Dryomov wrote: What's your ceph version and what does dmesg say? 3.2 is*way* too old, you are probably missing more than one required feature bit. See http://docs.ceph.com/docs/ma

[ceph-users] two or three replicas?

2015-11-02 Thread Wah Peng
Hello, for production application (for example, openstack's block storage), is it better to setup data to be stored with two replicas, or three replicas? is two replicas with better performance and lower cost? Thanks. ___ ceph-users mailing list cep

[ceph-users] rollback fail?

2015-11-10 Thread wah peng
Hello, I follow these steps trying to rollback a snapshot but it seems fail, the files are lost. root@ceph3:/mnt/ceph-block-device# rbd snap create rbd/bar@snap1 root@ceph3:/mnt/ceph-block-device# rbd snap ls rbd/bar SNAPID NAME SIZE 2 snap1 10240 MB root@ceph3:/mnt/ceph-block-device

[ceph-users] about PG_Number

2015-11-12 Thread wah peng
Hello, what's the disadvantage if setup PG_Number too large or too small against OSD number? Thanks. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Re: [ceph-users] about PG_Number

2015-11-13 Thread wah peng
why data lost happens? thanks. On 2015/11/13 星期五 16:13, Vickie ch wrote: In the other side, pg number too large but OSD number too small that have a chance to cause data lost. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com