The osd data directories are on /mnt/osd$id which are btrfs.
The mon data directory is /mnt/local/mon$id which is xfs. (/mnt/local
is the mount point)
Here is the ceph.conf file (note: secure authentication is turned off
at the moment):
; global
[global]
; enable secure authentication
; auth supported = cephx
; monitors
; You need at least one. You need at least three if you want to
; tolerate any node failures. Always create an odd number.
[mon]
mon data = /mnt/local/mon$id
; some minimal logging (just message traffic) to aid debugging
debug ms = 0
[mon.0]
host = 172.16.0.130
mon addr = 172.16.0.130:6789
[mon.1]
host = 172.16.0.128
mon addr = 172.16.0.128:6789
[mon.2]
host = 172.16.0.129
mon addr = 172.16.0.129:6789
; mds
; You need at least one. Define two to get a standby.
[mds]
; where the mds keeps its secret encryption keys
keyring = /mnt/local/keyring.$name
[mds.node-172-16-0-130]
host = 172.16.0.130
[mds.node-172-16-0-128]
host = 172.16.0.128
[mds.node-172-16-0-129]
host = 172.16.0.129
mds standby replay = true
mds standby for name = node-172-16-0-130
[osd]
; This is where the btrfs volume will be mounted.
osd data = /mnt/osd$id
; Ideally, make this a separate disk or partition. A few GB
; is usually enough; more if you have fast disks. You can use
; a file under the osd data dir if need be
; (e.g. /mnt/osd$id/journal), but it will be slower than a
; separate disk or partition.
osd journal = /mnt/osd$id/journal
osd class dir = /lib/rados-classes
; If the OSD journal is a file, you need to specify the size. This
is specified in MB.
osd journal size = 512
[osd.0]
host = 172.16.0.130
[osd.1]
host = 172.16.0.130
[osd.2]
host = 172.16.0.130
[osd.3]
host = 172.16.0.130
[osd.24]
host = 172.16.0.128
[osd.25]
host = 172.16.0.128
[osd.26]
host = 172.16.0.128
[osd.27]
host = 172.16.0.128
[osd.48]
host = 172.16.0.129
[osd.49]
host = 172.16.0.129
[osd.50]
host = 172.16.0.129
[osd.51]
host = 172.16.0.129
On Thu, Nov 3, 2011 at 12:23 PM, Tommi Virtanen
<tommi.virtanen@xxxxxxxxxxxxx> wrote:
> On Thu, Nov 3, 2011 at 12:09, Mandell Degerness <mandell@xxxxxxxxxxxxxxx> wrote:
>> We are currently running all services on each of the three nodes.
>> (mon, mds, and several osd processes).
>
> Is the osd data directory on the same filesystem as the mon data dir,
> or are they perhaps on different disks? (Hoping for the former, for
> the sake of xfs credibility ;)
>
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
