Hi,
how did you create the OSDs? Were they built from scratch with the
respective command options (--block.db /dev/<SSD>)?
You could check what the bluestore tool tells you about the block.db:
ceph1:~ # ceph-bluestore-tool show-label --dev
/var/lib/ceph/osd/ceph-21/block | grep path
"path_block.db": "/dev/ceph-journals/bluefsdb-21",
Does it point to the right device(s)?
Regards,
Eugen
Zitat von Hervé Ballans <herve.ball...@ias.u-psud.fr>:
Hi,
By testing the command on my side, it gives me the right information
(modulo the fact that the disk is a nvme and not ssd) :
# ceph osd metadata 1 |grep bluefs_db
"bluefs_db_access_mode": "blk",
"bluefs_db_block_size": "4096",
"bluefs_db_dev": "259:3",
"bluefs_db_dev_node": "nvme0n1",
"bluefs_db_driver": "KernelDevice",
"bluefs_db_model": "Dell Express Flash PM1725a 800GB SFF ",
"bluefs_db_partition_path": "/dev/nvme0n1p2",
"bluefs_db_rotational": "0",
"bluefs_db_serial": " S39YNX0JC02801",
"bluefs_db_size": "80016834560",
"bluefs_db_type": "ssd",
I notice that in your case, the db_model returns the information on
the PERC card and not on the disk...
Maybe this is where the missing information comes from ?
Le 22/09/2018 à 01:24, Brett Chancellor a écrit :
Hi all. Quick question about osd metadata information. I have
several OSDs setup with the data dir on HDD and the db going to a
partition on ssd. But when I look at the metadata for all the OSDs,
it's showing the db as "hdd". Does this effect anything? And is
there anyway to change it?
$ sudo ceph osd metadata 1
{
"id": 1,
"arch": "x86_64",
"back_addr": "<removed>:6805/2053608",
"back_iface": "eth0",
"bluefs": "1",
"bluefs_db_access_mode": "blk",
"bluefs_db_block_size": "4096",
"bluefs_db_dev": "8:80",
"bluefs_db_dev_node": "sdf",
"bluefs_db_driver": "KernelDevice",
"bluefs_db_model": "PERC H730 Mini ",
"bluefs_db_partition_path": "/dev/sdf2",
"bluefs_db_rotational": "1",
"bluefs_db_size": "266287972352",
* "bluefs_db_type": "hdd",*
_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
