Re: stuck in rbd client accessing pool

Ken Peng <ken@xxxxxxxxxx> · Tue, 7 Jun 2016 09:42:53 +0800

Hi,
I think you can specify the pool name in client settings.
for example, in our environment,

# rbd ls
rbd: error opening pool rbd: (2) No such file or directory

# rbd ls -p block
f7470c3f-e051-4f3d-86ff-52e8ba78ac4a
022e9944-122c-4ad0-b652-9e52ba32e2c0

Here -p pool_name was specified. that works.

2016-06-07 0:01 GMT+08:00 strony zhang <strony.zhang@xxxxxxxxx>:
Hi Ken,

Thanks for your reply.
The ceph cluster runs well.

:~$ sudo ceph -s
    cluster 285441d6-c059-405d-9762-86bd91f279d0
     health HEALTH_OK
     monmap e1: 1 mons at {strony-pc=10.132.138.233:6789/0}
            election epoch 9, quorum 0 strony-pc
     osdmap e200: 2 osds: 2 up, 2 in
            flags sortbitwise
      pgmap v225126: 256 pgs, 1 pools, 345 bytes data, 10 objects
            10326 MB used, 477 GB / 488 GB avail
                 256 active+clean
  client io 0 B/s rd, 193 op/s rd, 0 op/s wr

$ ceph osd lspools
6 rbd,

I previously deleted some pools. So, the latest ID for the pool, 'rbd', is 6. I guess the client probably tries accessing the first pool by default and then got stuck. So, how can I change the pool ID into '0'?

Thanks,
Strony 

    On Monday, June 6, 2016 1:46 AM, Ken Peng <ken@xxxxxxxxxx> wrote:

 hello,
Does ceph cluster work right? run ceph -s and ceph -w for watching more details.

2016-06-06 16:17 GMT+08:00 strony zhang <strony.zhang@xxxxxxxxx>:
Hi,

I am a new learner in ceph. Now I install an All-in-one ceph on the host A. Then I tried accessing the ceph from another host B with librados and librbd installed.

From host B: I run python to access the ceph on host A.
>>> import rados
>>> cluster1 = rados.Rados(conffile='/etc/ceph/ceph.conf')
>>> cluster1.connect()
>>> print cluster1.get_fsid()
285441d6-c059-405d-9762-86bd91f279d0
>>> 
>>> import rbd
>>> rbd_inst = rbd.RBD()
>>> ioctx = cluster1.open_ioctx('rbd')
>>> rbd_inst.list(ioctx)
.... stuck in here; it never exits until the python program is killed manually.

But in Host A, I don't find any error info.
zq@zq-ubuntu:~$ rbd list -l
NAME  SIZE PARENT FMT PROT LOCK 
z1   1024M          2           
z2   1024M          2           
z3   1024M          2  

The ceph.conf and ceph.client.admin.keyring in host B are the same to those in host A. Any comments are appreciated.

Thanks,
Strony

_______________________________________________

ceph-users mailing list

ceph-users@xxxxxxxxxxxxxx

http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com