-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

My next thing would be to turn up debugging on all the monitors and
see if the survivors are having a problem forming a quorum.

- ----------------
Robert LeBlanc
PGP Fingerprint 79A2 9CA4 6CC4 45DD A904  C70E E654 3BB2 FA62 B9F1

On Thu, Aug 27, 2015 at 7:53 PM, [email protected]  wrote:
ceph version:0.94.2

ceph.conf:
[global]
fsid = 51b9437c-b49d-460c-b919-4bd86cc4d97d
mon_initial_members = ubuntu2, ubuntu3, ubuntu4
mon_host = 192.168.33.182,192.168.33.183,192.168.33.184
auth_cluster_required = cephx
auth_service_required = cephx
auth_client_required = cephx
filestore_xattr_use_omap = true

I don't have any special configuration

Thanks
发件人: Robert LeBlanc [mailto:[email protected]]
发送时间: 2015年8月28日 2:27
收件人: zhaomingyue 09440 (RD)
抄送: [email protected]
主题: Re: [ceph-users] shutdown primary monitor, status of the osds on
it will not change, and command like 'rbd create xxx' would block

- -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

The OSDs should be marked down within about 30 seconds. Can you
provide additional information such as ceph version and the ceph.conf
file.

Thanks,

- - ----------------
Robert LeBlanc
PGP Fingerprint 79A2 9CA4 6CC4 45DD A904  C70E E654 3BB2 FA62 B9F1

On Thu, Aug 27, 2015 at 12:23 AM, [email protected]  wrote:
Hi,
I'm testing ceph feature,and now have a probem make me confused.

Network construction is as shown in the figure below:
  serverB is primary monitor , it has 5 osds
  serverC is secondary monitor, it has 5 osds
  serverD is tertiary monitor,it has 5 osds
  client A use a rbd block via 'rbd map' command.

When shutdown serverB(primary monitor), input ‘ceph osd tree’ on
serverC or serverD will show all the osd is up and in, At this time
,if input ‘rbd create aaa –size 100’, it will block, after neally 18
minutes, osds on serverB change status to down, and command ‘rbd
create xxx’ return;

My question:
1.       Why rbd create will block about 18minutes? Is this because
the status of osds on serverB(the shutdown one) is not modified to
down? What is the time mechanism about this?

2.       When write objects, if primary osd think other osd is up, it
will always wait until other osd reply, it doesn’t have timeout
mechanism, isn’t it?

3.       The same environment , if I shutdown serverC(secondary
monitor) , ‘ceph osd tree’ show four osds down and another one osd is
up;

if I shutdown serverD,the result is five osds are all down or all up,
when all osd is up,the result is the same as shutdown server B;

shutdown primary monitor / secondary monitor / tertiary monitor, ceph
status is not the same,I’m very confused about this.Can someone help
me?

- - 
-------------------------------------------------------------------------------------------------------------------------------------
本邮件及其附件含有杭州华三通信技术有限公司的保密信息,仅限于发送给上面地址中列出
的个人或群组。禁止任何其他人以任何形式使用(包括但不限于全部或部分地泄露、复制、
或散发)本邮件中的信息。如果您错收了本邮件,请您立即电话或邮件通知发件人并删除本
邮件!
This e-mail and its attachments contain confidential information from
H3C, which is
intended only for the person or entity whose address is listed above.
Any use of the
information contained herein in any way (including, but not limited
to, total or partial
disclosure, reproduction, or dissemination) by persons other than the intended
recipient(s) is prohibited. If you receive this e-mail in error,
please notify the sender
by phone or email immediately and delete it!

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


- -----BEGIN PGP SIGNATURE-----
Version: Mailvelope v1.0.2
Comment: https://www.mailvelope.com
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=ndFY
- -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNATURE-----
Version: Mailvelope v1.0.2
Comment: https://www.mailvelope.com
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=g+ql
-----END PGP SIGNATURE-----
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to