I’ll try the prepare/activiate commands again. I spent the least amount of time with them since activate _always_ failed for me. I’ll go back and check my logs, but probably because I was attempting to activate the same location I used in the ‘prepare’ instead of the partition 1 like you suggest (which is exactly how it is show in the documentation example).
I seemed to get the closest to a working cluster using the ‘manual’ commands below. I could try changing the XFS mount point to be on a partition of the hdd I’m using for the osd. mkdir /var/lib/ceph/osd/ceph-$OSD mkfs -t xfs -f /dev/sd$i mount -t xfs /dev/sd$i /var/lib/ceph/osd/ceph-$OSD ceph-osd -i $OSD --mkfs --mkkey --osd-journal /dev/md0p$PART What I find most confusing using ceph-deploy with multiple osds on the same host is that when ‘ceph-deploy osd create [data] [journal]’ completes there is no osd directory for each osd under: [root@ceph0 ceph]# ll /var/lib/ceph/osd/ total 0 [root@ceph0 ceph]# From: ceph-users [mailto:ceph-users-boun...@lists.ceph.com] On Behalf Of Jason King Sent: Thursday, August 14, 2014 8:13 PM To: ceph-us...@ceph.com Subject: Re: [ceph-users] How to create multiple OSD's per host? 2014-08-15 7:56 GMT+08:00 Bruce McFarland <bruce.mcfarl...@taec.toshiba.com<mailto:bruce.mcfarl...@taec.toshiba.com>>: This is an example of the output from ‘ceph-deploy osd create [data] [journal’ I’ve noticed that all of the ‘ceph-conf’ commands use the same parameter of ‘–name=osd.’ Everytime ceph-deploy is called. I end up with 30 osd’s – 29 in the prepared and 1 active according to the ‘ceph-disk list’ output and only 1 osd that has a xfs mount point. I’ve tried both with all data/journal devices on the same ceph-deploy command line and issuing 1 ceph-deploy cmd for each OSD data/journal pair (easier to script). + ceph-deploy osd create ceph0:/dev/sdl:/dev/md0p17 [ceph_deploy.conf][DEBUG ] found configuration file at: /root/.cephdeploy.conf [ceph_deploy.cli][INFO ] Invoked (1.5.10): /usr/bin/ceph-deploy osd create ceph0:/dev/sdl:/dev/md0p17 [ceph_deploy.osd][DEBUG ] Preparing cluster ceph disks ceph0:/dev/sdl:/dev/md0p17 [ceph0][DEBUG ] connected to host: ceph0 [ceph0][DEBUG ] detect platform information from remote host [ceph0][DEBUG ] detect machine type [ceph_deploy.osd][INFO ] Distro info: CentOS 6.5 Final [ceph_deploy.osd][DEBUG ] Deploying osd to ceph0 [ceph0][DEBUG ] write cluster configuration to /etc/ceph/{cluster}.conf [ceph0][INFO ] Running command: udevadm trigger --subsystem-match=block --action=add [ceph_deploy.osd][DEBUG ] Preparing host ceph0 disk /dev/sdl journal /dev/md0p17 activate True [ceph0][INFO ] Running command: ceph-disk -v prepare --fs-type xfs --cluster ceph -- /dev/sdl /dev/md0p17 [ceph0][DEBUG ] Information: Moved requested sector from 34 to 2048 in [ceph0][DEBUG ] order to align on 2048-sector boundaries. [ceph0][DEBUG ] The operation has completed successfully. [ceph0][DEBUG ] meta-data=/dev/sdl1 isize=2048 agcount=4, agsize=244188597 blks [ceph0][DEBUG ] = sectsz=512 attr=2, projid32bit=0 [ceph0][DEBUG ] data = bsize=4096 blocks=976754385, imaxpct=5 [ceph0][DEBUG ] = sunit=0 swidth=0 blks [ceph0][DEBUG ] naming =version 2 bsize=4096 ascii-ci=0 [ceph0][DEBUG ] log =internal log bsize=4096 blocks=476930, version=2 [ceph0][DEBUG ] = sectsz=512 sunit=0 blks, lazy-count=1 [ceph0][DEBUG ] realtime =none extsz=4096 blocks=0, rtextents=0 [ceph0][DEBUG ] The operation has completed successfully. [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-osd --cluster=ceph --show-config-value=fsid [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_mkfs_options_xfs [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_fs_mkfs_options_xfs [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_mount_options_xfs [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-conf --cluster=ceph --name=osd. --lookup osd_fs_mount_options_xfs [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/bin/ceph-osd --cluster=ceph --show-config-value=osd_journal_size [ceph0][WARNIN] DEBUG:ceph-disk:Journal /dev/md0p17 is a partition [ceph0][WARNIN] WARNING:ceph-disk:OSD will not be hot-swappable if journal is not the same device as the osd data [ceph0][WARNIN] DEBUG:ceph-disk:Creating osd partition on /dev/sdl [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/sbin/sgdisk --largest-new=1 --change-name=1:ceph data --partition-guid=1:a96b4af4-11f4-4257-9476-64a6e4c93c28 --typecode=1:89c57f98-2fe5-4dc0-89c1-f3ad0ceff2be -- /dev/sdl [ceph0][WARNIN] INFO:ceph-disk:Running command: /sbin/partprobe /dev/sdl [ceph0][WARNIN] INFO:ceph-disk:Running command: /sbin/udevadm settle [ceph0][WARNIN] DEBUG:ceph-disk:Creating xfs fs on /dev/sdl1 [ceph0][WARNIN] INFO:ceph-disk:Running command: /sbin/mkfs -t xfs -f -i size=2048 -- /dev/sdl1 [ceph0][WARNIN] DEBUG:ceph-disk:Mounting /dev/sdl1 on /var/lib/ceph/tmp/mnt.8xAu31 with options noatime [ceph0][WARNIN] INFO:ceph-disk:Running command: /bin/mount -t xfs -o noatime -- /dev/sdl1 /var/lib/ceph/tmp/mnt.8xAu31 [ceph0][WARNIN] DEBUG:ceph-disk:Preparing osd data dir /var/lib/ceph/tmp/mnt.8xAu31 [ceph0][WARNIN] DEBUG:ceph-disk:Creating symlink /var/lib/ceph/tmp/mnt.8xAu31/journal -> /dev/md0p17 [ceph0][WARNIN] DEBUG:ceph-disk:Unmounting /var/lib/ceph/tmp/mnt.8xAu31 [ceph0][WARNIN] INFO:ceph-disk:Running command: /bin/umount -- /var/lib/ceph/tmp/mnt.8xAu31 [ceph0][WARNIN] INFO:ceph-disk:Running command: /usr/sbin/sgdisk --typecode=1:4fbd7e29-9d25-41b8-afd0-062c0ceff05d -- /dev/sdl [ceph0][WARNIN] INFO:ceph-disk:calling partx on prepared device /dev/sdl [ceph0][WARNIN] INFO:ceph-disk:re-reading known partitions will display errors [ceph0][WARNIN] INFO:ceph-disk:Running command: /sbin/partx -a /dev/sdl [ceph0][WARNIN] BLKPG: Device or resource busy [ceph0][WARNIN] error adding partition 1 [ceph0][INFO ] Running command: udevadm trigger --subsystem-match=block --action=add [ceph0][INFO ] checking OSD status... [ceph0][INFO ] Running command: ceph --cluster=ceph osd stat --format=json [ceph_deploy.osd][DEBUG ] Host ceph0 is now ready for osd use. From: Bruce McFarland Sent: Thursday, August 14, 2014 11:45 AM To: 'ceph-us...@ceph.com<mailto:ceph-us...@ceph.com>' Subject: How to create multiple OSD's per host? I’ve tried using ceph-deploy but it wants to assign the same id for each osd and I end up with a bunch of “prepared” ceph-disk’s and only 1 “active”. If I use the manual “short form” method the activate step fails and there are no xfs mount points on the ceph-disks. If I use the manual “long form” it seems like I’m the closest to getting active ceph-disks/osd’s but the monitor always shows the osds as “down/in” and the ceph-disks don’t persist over a boot cycle. Is there a document anywhere that anyone knows of that explains a step by step process for bringing up multiple osd’s per host – 1 hdd with ssd journal partition per osd? Thanks, Bruce _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com<mailto:ceph-users@lists.ceph.com> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com Hi Bruce, I depolyed 3 OSDs per host a few days ago and it worked pretty well. Examined your log and guess you should try: ceph-deploy osd prepare ceph0:/dev/sdl:/dev/md0p17 ceph-deploy osd activate ceph0:/dev/sdl1:/dev/md0p17 Or you could create a partition on /dev/sdl by yourself and ceph-deploy osd create ceph0:/dev/sdl1:/dev/md0p17 Hope this works. Jason
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com