Hi,
It looks like at sdb size around 1.1 GBytes ceph (ceph version 12.2.0
(32ce2a3ae5239ee33d6150705cdb24d43bab910c) luminous (rc)) is not crashing
anymore.
Please don't increase the minimum disk size requirements unnecessarily - it
makes it more demanding to test new ceph features and operational
Hi,
/var/log/ceph/ceph-osd.0.log is attached.
My sdb is 128MB and sdc (journal) is 16MB:
[root@server0 ~]# ceph-disk list
/dev/dm-0 other, xfs, mounted on /
/dev/dm-1 swap, swap
/dev/sda :
/dev/sda1 other, 0x83
/dev/sda2 other, xfs, mounted on /boot
/dev/sda3 other, LVM2_member
/dev/sdb :
/de
Hi Marcin,
Can you reproduce the crash with 'debug bluestore = 20' set, and then
ceph-post-file /var/log/ceph/ceph-osd.0.log?
My guess is that we're not handling a very small device properly?
sage
On Thu, 31 Aug 2017, Marcin Dulak wrote:
> Hi,
>
> I have a virtual CentOS 7.3 test setup at:
Hi,
I have a virtual CentOS 7.3 test setup at:
https://github.com/marcindulak/github-test-local/blob/a339ff
7505267545f593fd949a6453a56cdfd7fe/vagrant-ceph-rbd-tutorial-centos7.sh
It seems to crash reproducibly with luminous, and works with kraken.
Is this a known issue?
[ceph_deploy.conf][DEBUG
